diff options
Diffstat (limited to '.venv/lib/python3.12/site-packages/networkx/readwrite')
35 files changed, 13253 insertions, 0 deletions
diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/__init__.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/__init__.py new file mode 100644 index 00000000..a805c50a --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/__init__.py @@ -0,0 +1,17 @@ +""" +A package for reading and writing graphs in various formats. + +""" + +from networkx.readwrite.adjlist import * +from networkx.readwrite.multiline_adjlist import * +from networkx.readwrite.edgelist import * +from networkx.readwrite.pajek import * +from networkx.readwrite.leda import * +from networkx.readwrite.sparse6 import * +from networkx.readwrite.graph6 import * +from networkx.readwrite.gml import * +from networkx.readwrite.graphml import * +from networkx.readwrite.gexf import * +from networkx.readwrite.json_graph import * +from networkx.readwrite.text import * diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/adjlist.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/adjlist.py new file mode 100644 index 00000000..768af5ad --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/adjlist.py @@ -0,0 +1,310 @@ +""" +************** +Adjacency List +************** +Read and write NetworkX graphs as adjacency lists. + +Adjacency list format is useful for graphs without data associated +with nodes or edges and for nodes that can be meaningfully represented +as strings. + +Format +------ +The adjacency list format consists of lines with node labels. The +first label in a line is the source node. Further labels in the line +are considered target nodes and are added to the graph along with an edge +between the source node and target node. + +The graph with edges a-b, a-c, d-e can be represented as the following +adjacency list (anything following the # in a line is a comment):: + + a b c # source target target + d e +""" + +__all__ = ["generate_adjlist", "write_adjlist", "parse_adjlist", "read_adjlist"] + +import networkx as nx +from networkx.utils import open_file + + +def generate_adjlist(G, delimiter=" "): + """Generate a single line of the graph G in adjacency list format. + + Parameters + ---------- + G : NetworkX graph + + delimiter : string, optional + Separator for node labels + + Returns + ------- + lines : string + Lines of data in adjlist format. + + Examples + -------- + >>> G = nx.lollipop_graph(4, 3) + >>> for line in nx.generate_adjlist(G): + ... print(line) + 0 1 2 3 + 1 2 3 + 2 3 + 3 4 + 4 5 + 5 6 + 6 + + See Also + -------- + write_adjlist, read_adjlist + + Notes + ----- + The default `delimiter=" "` will result in unexpected results if node names contain + whitespace characters. To avoid this problem, specify an alternate delimiter when spaces are + valid in node names. + + NB: This option is not available for data that isn't user-generated. + + """ + directed = G.is_directed() + seen = set() + for s, nbrs in G.adjacency(): + line = str(s) + delimiter + for t, data in nbrs.items(): + if not directed and t in seen: + continue + if G.is_multigraph(): + for d in data.values(): + line += str(t) + delimiter + else: + line += str(t) + delimiter + if not directed: + seen.add(s) + yield line[: -len(delimiter)] + + +@open_file(1, mode="wb") +def write_adjlist(G, path, comments="#", delimiter=" ", encoding="utf-8"): + """Write graph G in single-line adjacency-list format to path. + + + Parameters + ---------- + G : NetworkX graph + + path : string or file + Filename or file handle for data output. + Filenames ending in .gz or .bz2 will be compressed. + + comments : string, optional + Marker for comment lines + + delimiter : string, optional + Separator for node labels + + encoding : string, optional + Text encoding. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_adjlist(G, "test.adjlist") + + The path can be a filehandle or a string with the name of the file. If a + filehandle is provided, it has to be opened in 'wb' mode. + + >>> fh = open("test.adjlist", "wb") + >>> nx.write_adjlist(G, fh) + + Notes + ----- + The default `delimiter=" "` will result in unexpected results if node names contain + whitespace characters. To avoid this problem, specify an alternate delimiter when spaces are + valid in node names. + NB: This option is not available for data that isn't user-generated. + + This format does not store graph, node, or edge data. + + See Also + -------- + read_adjlist, generate_adjlist + """ + import sys + import time + + pargs = comments + " ".join(sys.argv) + "\n" + header = ( + pargs + + comments + + f" GMT {time.asctime(time.gmtime())}\n" + + comments + + f" {G.name}\n" + ) + path.write(header.encode(encoding)) + + for line in generate_adjlist(G, delimiter): + line += "\n" + path.write(line.encode(encoding)) + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_adjlist( + lines, comments="#", delimiter=None, create_using=None, nodetype=None +): + """Parse lines of a graph adjacency list representation. + + Parameters + ---------- + lines : list or iterator of strings + Input data in adjlist format + + create_using : NetworkX graph constructor, optional (default=nx.Graph) + Graph type to create. If graph instance, then cleared before populated. + + nodetype : Python type, optional + Convert nodes to this type. + + comments : string, optional + Marker for comment lines + + delimiter : string, optional + Separator for node labels. The default is whitespace. + + Returns + ------- + G: NetworkX graph + The graph corresponding to the lines in adjacency list format. + + Examples + -------- + >>> lines = ["1 2 5", "2 3 4", "3 5", "4", "5"] + >>> G = nx.parse_adjlist(lines, nodetype=int) + >>> nodes = [1, 2, 3, 4, 5] + >>> all(node in G for node in nodes) + True + >>> edges = [(1, 2), (1, 5), (2, 3), (2, 4), (3, 5)] + >>> all((u, v) in G.edges() or (v, u) in G.edges() for (u, v) in edges) + True + + See Also + -------- + read_adjlist + + """ + G = nx.empty_graph(0, create_using) + for line in lines: + p = line.find(comments) + if p >= 0: + line = line[:p] + if not len(line): + continue + vlist = line.rstrip("\n").split(delimiter) + u = vlist.pop(0) + # convert types + if nodetype is not None: + try: + u = nodetype(u) + except BaseException as err: + raise TypeError( + f"Failed to convert node ({u}) to type {nodetype}" + ) from err + G.add_node(u) + if nodetype is not None: + try: + vlist = list(map(nodetype, vlist)) + except BaseException as err: + raise TypeError( + f"Failed to convert nodes ({','.join(vlist)}) to type {nodetype}" + ) from err + G.add_edges_from([(u, v) for v in vlist]) + return G + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_adjlist( + path, + comments="#", + delimiter=None, + create_using=None, + nodetype=None, + encoding="utf-8", +): + """Read graph in adjacency list format from path. + + Parameters + ---------- + path : string or file + Filename or file handle to read. + Filenames ending in .gz or .bz2 will be uncompressed. + + create_using : NetworkX graph constructor, optional (default=nx.Graph) + Graph type to create. If graph instance, then cleared before populated. + + nodetype : Python type, optional + Convert nodes to this type. + + comments : string, optional + Marker for comment lines + + delimiter : string, optional + Separator for node labels. The default is whitespace. + + Returns + ------- + G: NetworkX graph + The graph corresponding to the lines in adjacency list format. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_adjlist(G, "test.adjlist") + >>> G = nx.read_adjlist("test.adjlist") + + The path can be a filehandle or a string with the name of the file. If a + filehandle is provided, it has to be opened in 'rb' mode. + + >>> fh = open("test.adjlist", "rb") + >>> G = nx.read_adjlist(fh) + + Filenames ending in .gz or .bz2 will be compressed. + + >>> nx.write_adjlist(G, "test.adjlist.gz") + >>> G = nx.read_adjlist("test.adjlist.gz") + + The optional nodetype is a function to convert node strings to nodetype. + + For example + + >>> G = nx.read_adjlist("test.adjlist", nodetype=int) + + will attempt to convert all nodes to integer type. + + Since nodes must be hashable, the function nodetype must return hashable + types (e.g. int, float, str, frozenset - or tuples of those, etc.) + + The optional create_using parameter indicates the type of NetworkX graph + created. The default is `nx.Graph`, an undirected graph. + To read the data as a directed graph use + + >>> G = nx.read_adjlist("test.adjlist", create_using=nx.DiGraph) + + Notes + ----- + This format does not store graph or node data. + + See Also + -------- + write_adjlist + """ + lines = (line.decode(encoding) for line in path) + return parse_adjlist( + lines, + comments=comments, + delimiter=delimiter, + create_using=create_using, + nodetype=nodetype, + ) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/edgelist.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/edgelist.py new file mode 100644 index 00000000..393b64ed --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/edgelist.py @@ -0,0 +1,489 @@ +""" +********** +Edge Lists +********** +Read and write NetworkX graphs as edge lists. + +The multi-line adjacency list format is useful for graphs with nodes +that can be meaningfully represented as strings. With the edgelist +format simple edge data can be stored but node or graph data is not. +There is no way of representing isolated nodes unless the node has a +self-loop edge. + +Format +------ +You can read or write three formats of edge lists with these functions. + +Node pairs with no data:: + + 1 2 + +Python dictionary as data:: + + 1 2 {'weight':7, 'color':'green'} + +Arbitrary data:: + + 1 2 7 green +""" + +__all__ = [ + "generate_edgelist", + "write_edgelist", + "parse_edgelist", + "read_edgelist", + "read_weighted_edgelist", + "write_weighted_edgelist", +] + +import networkx as nx +from networkx.utils import open_file + + +def generate_edgelist(G, delimiter=" ", data=True): + """Generate a single line of the graph G in edge list format. + + Parameters + ---------- + G : NetworkX graph + + delimiter : string, optional + Separator for node labels + + data : bool or list of keys + If False generate no edge data. If True use a dictionary + representation of edge data. If a list of keys use a list of data + values corresponding to the keys. + + Returns + ------- + lines : string + Lines of data in adjlist format. + + Examples + -------- + >>> G = nx.lollipop_graph(4, 3) + >>> G[1][2]["weight"] = 3 + >>> G[3][4]["capacity"] = 12 + >>> for line in nx.generate_edgelist(G, data=False): + ... print(line) + 0 1 + 0 2 + 0 3 + 1 2 + 1 3 + 2 3 + 3 4 + 4 5 + 5 6 + + >>> for line in nx.generate_edgelist(G): + ... print(line) + 0 1 {} + 0 2 {} + 0 3 {} + 1 2 {'weight': 3} + 1 3 {} + 2 3 {} + 3 4 {'capacity': 12} + 4 5 {} + 5 6 {} + + >>> for line in nx.generate_edgelist(G, data=["weight"]): + ... print(line) + 0 1 + 0 2 + 0 3 + 1 2 3 + 1 3 + 2 3 + 3 4 + 4 5 + 5 6 + + See Also + -------- + write_adjlist, read_adjlist + """ + if data is True: + for u, v, d in G.edges(data=True): + e = u, v, dict(d) + yield delimiter.join(map(str, e)) + elif data is False: + for u, v in G.edges(data=False): + e = u, v + yield delimiter.join(map(str, e)) + else: + for u, v, d in G.edges(data=True): + e = [u, v] + try: + e.extend(d[k] for k in data) + except KeyError: + pass # missing data for this edge, should warn? + yield delimiter.join(map(str, e)) + + +@open_file(1, mode="wb") +def write_edgelist(G, path, comments="#", delimiter=" ", data=True, encoding="utf-8"): + """Write graph as a list of edges. + + Parameters + ---------- + G : graph + A NetworkX graph + path : file or string + File or filename to write. If a file is provided, it must be + opened in 'wb' mode. Filenames ending in .gz or .bz2 will be compressed. + comments : string, optional + The character used to indicate the start of a comment + delimiter : string, optional + The string used to separate values. The default is whitespace. + data : bool or list, optional + If False write no edge data. + If True write a string representation of the edge data dictionary.. + If a list (or other iterable) is provided, write the keys specified + in the list. + encoding: string, optional + Specify which encoding to use when writing file. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_edgelist(G, "test.edgelist") + >>> G = nx.path_graph(4) + >>> fh = open("test.edgelist", "wb") + >>> nx.write_edgelist(G, fh) + >>> nx.write_edgelist(G, "test.edgelist.gz") + >>> nx.write_edgelist(G, "test.edgelist.gz", data=False) + + >>> G = nx.Graph() + >>> G.add_edge(1, 2, weight=7, color="red") + >>> nx.write_edgelist(G, "test.edgelist", data=False) + >>> nx.write_edgelist(G, "test.edgelist", data=["color"]) + >>> nx.write_edgelist(G, "test.edgelist", data=["color", "weight"]) + + See Also + -------- + read_edgelist + write_weighted_edgelist + """ + + for line in generate_edgelist(G, delimiter, data): + line += "\n" + path.write(line.encode(encoding)) + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_edgelist( + lines, comments="#", delimiter=None, create_using=None, nodetype=None, data=True +): + """Parse lines of an edge list representation of a graph. + + Parameters + ---------- + lines : list or iterator of strings + Input data in edgelist format + comments : string, optional + Marker for comment lines. Default is `'#'`. To specify that no character + should be treated as a comment, use ``comments=None``. + delimiter : string, optional + Separator for node labels. Default is `None`, meaning any whitespace. + create_using : NetworkX graph constructor, optional (default=nx.Graph) + Graph type to create. If graph instance, then cleared before populated. + nodetype : Python type, optional + Convert nodes to this type. Default is `None`, meaning no conversion is + performed. + data : bool or list of (label,type) tuples + If `False` generate no edge data or if `True` use a dictionary + representation of edge data or a list tuples specifying dictionary + key names and types for edge data. + + Returns + ------- + G: NetworkX Graph + The graph corresponding to lines + + Examples + -------- + Edgelist with no data: + + >>> lines = ["1 2", "2 3", "3 4"] + >>> G = nx.parse_edgelist(lines, nodetype=int) + >>> list(G) + [1, 2, 3, 4] + >>> list(G.edges()) + [(1, 2), (2, 3), (3, 4)] + + Edgelist with data in Python dictionary representation: + + >>> lines = ["1 2 {'weight': 3}", "2 3 {'weight': 27}", "3 4 {'weight': 3.0}"] + >>> G = nx.parse_edgelist(lines, nodetype=int) + >>> list(G) + [1, 2, 3, 4] + >>> list(G.edges(data=True)) + [(1, 2, {'weight': 3}), (2, 3, {'weight': 27}), (3, 4, {'weight': 3.0})] + + Edgelist with data in a list: + + >>> lines = ["1 2 3", "2 3 27", "3 4 3.0"] + >>> G = nx.parse_edgelist(lines, nodetype=int, data=(("weight", float),)) + >>> list(G) + [1, 2, 3, 4] + >>> list(G.edges(data=True)) + [(1, 2, {'weight': 3.0}), (2, 3, {'weight': 27.0}), (3, 4, {'weight': 3.0})] + + See Also + -------- + read_weighted_edgelist + """ + from ast import literal_eval + + G = nx.empty_graph(0, create_using) + for line in lines: + if comments is not None: + p = line.find(comments) + if p >= 0: + line = line[:p] + if not line: + continue + # split line, should have 2 or more + s = line.rstrip("\n").split(delimiter) + if len(s) < 2: + continue + u = s.pop(0) + v = s.pop(0) + d = s + if nodetype is not None: + try: + u = nodetype(u) + v = nodetype(v) + except Exception as err: + raise TypeError( + f"Failed to convert nodes {u},{v} to type {nodetype}." + ) from err + + if len(d) == 0 or data is False: + # no data or data type specified + edgedata = {} + elif data is True: + # no edge types specified + try: # try to evaluate as dictionary + if delimiter == ",": + edgedata_str = ",".join(d) + else: + edgedata_str = " ".join(d) + edgedata = dict(literal_eval(edgedata_str.strip())) + except Exception as err: + raise TypeError( + f"Failed to convert edge data ({d}) to dictionary." + ) from err + else: + # convert edge data to dictionary with specified keys and type + if len(d) != len(data): + raise IndexError( + f"Edge data {d} and data_keys {data} are not the same length" + ) + edgedata = {} + for (edge_key, edge_type), edge_value in zip(data, d): + try: + edge_value = edge_type(edge_value) + except Exception as err: + raise TypeError( + f"Failed to convert {edge_key} data {edge_value} " + f"to type {edge_type}." + ) from err + edgedata.update({edge_key: edge_value}) + G.add_edge(u, v, **edgedata) + return G + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_edgelist( + path, + comments="#", + delimiter=None, + create_using=None, + nodetype=None, + data=True, + edgetype=None, + encoding="utf-8", +): + """Read a graph from a list of edges. + + Parameters + ---------- + path : file or string + File or filename to read. If a file is provided, it must be + opened in 'rb' mode. + Filenames ending in .gz or .bz2 will be uncompressed. + comments : string, optional + The character used to indicate the start of a comment. To specify that + no character should be treated as a comment, use ``comments=None``. + delimiter : string, optional + The string used to separate values. The default is whitespace. + create_using : NetworkX graph constructor, optional (default=nx.Graph) + Graph type to create. If graph instance, then cleared before populated. + nodetype : int, float, str, Python type, optional + Convert node data from strings to specified type + data : bool or list of (label,type) tuples + Tuples specifying dictionary key names and types for edge data + edgetype : int, float, str, Python type, optional OBSOLETE + Convert edge data from strings to specified type and use as 'weight' + encoding: string, optional + Specify which encoding to use when reading file. + + Returns + ------- + G : graph + A networkx Graph or other type specified with create_using + + Examples + -------- + >>> nx.write_edgelist(nx.path_graph(4), "test.edgelist") + >>> G = nx.read_edgelist("test.edgelist") + + >>> fh = open("test.edgelist", "rb") + >>> G = nx.read_edgelist(fh) + >>> fh.close() + + >>> G = nx.read_edgelist("test.edgelist", nodetype=int) + >>> G = nx.read_edgelist("test.edgelist", create_using=nx.DiGraph) + + Edgelist with data in a list: + + >>> textline = "1 2 3" + >>> fh = open("test.edgelist", "w") + >>> d = fh.write(textline) + >>> fh.close() + >>> G = nx.read_edgelist("test.edgelist", nodetype=int, data=(("weight", float),)) + >>> list(G) + [1, 2] + >>> list(G.edges(data=True)) + [(1, 2, {'weight': 3.0})] + + See parse_edgelist() for more examples of formatting. + + See Also + -------- + parse_edgelist + write_edgelist + + Notes + ----- + Since nodes must be hashable, the function nodetype must return hashable + types (e.g. int, float, str, frozenset - or tuples of those, etc.) + """ + lines = (line if isinstance(line, str) else line.decode(encoding) for line in path) + return parse_edgelist( + lines, + comments=comments, + delimiter=delimiter, + create_using=create_using, + nodetype=nodetype, + data=data, + ) + + +def write_weighted_edgelist(G, path, comments="#", delimiter=" ", encoding="utf-8"): + """Write graph G as a list of edges with numeric weights. + + Parameters + ---------- + G : graph + A NetworkX graph + path : file or string + File or filename to write. If a file is provided, it must be + opened in 'wb' mode. + Filenames ending in .gz or .bz2 will be compressed. + comments : string, optional + The character used to indicate the start of a comment + delimiter : string, optional + The string used to separate values. The default is whitespace. + encoding: string, optional + Specify which encoding to use when writing file. + + Examples + -------- + >>> G = nx.Graph() + >>> G.add_edge(1, 2, weight=7) + >>> nx.write_weighted_edgelist(G, "test.weighted.edgelist") + + See Also + -------- + read_edgelist + write_edgelist + read_weighted_edgelist + """ + write_edgelist( + G, + path, + comments=comments, + delimiter=delimiter, + data=("weight",), + encoding=encoding, + ) + + +@nx._dispatchable(graphs=None, returns_graph=True) +def read_weighted_edgelist( + path, + comments="#", + delimiter=None, + create_using=None, + nodetype=None, + encoding="utf-8", +): + """Read a graph as list of edges with numeric weights. + + Parameters + ---------- + path : file or string + File or filename to read. If a file is provided, it must be + opened in 'rb' mode. + Filenames ending in .gz or .bz2 will be uncompressed. + comments : string, optional + The character used to indicate the start of a comment. + delimiter : string, optional + The string used to separate values. The default is whitespace. + create_using : NetworkX graph constructor, optional (default=nx.Graph) + Graph type to create. If graph instance, then cleared before populated. + nodetype : int, float, str, Python type, optional + Convert node data from strings to specified type + encoding: string, optional + Specify which encoding to use when reading file. + + Returns + ------- + G : graph + A networkx Graph or other type specified with create_using + + Notes + ----- + Since nodes must be hashable, the function nodetype must return hashable + types (e.g. int, float, str, frozenset - or tuples of those, etc.) + + Example edgelist file format. + + With numeric edge data:: + + # read with + # >>> G=nx.read_weighted_edgelist(fh) + # source target data + a b 1 + a c 3.14159 + d e 42 + + See Also + -------- + write_weighted_edgelist + """ + return read_edgelist( + path, + comments=comments, + delimiter=delimiter, + create_using=create_using, + nodetype=nodetype, + data=(("weight", float),), + encoding=encoding, + ) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/gexf.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/gexf.py new file mode 100644 index 00000000..f830dd12 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/gexf.py @@ -0,0 +1,1066 @@ +"""Read and write graphs in GEXF format. + +.. warning:: + This parser uses the standard xml library present in Python, which is + insecure - see :external+python:mod:`xml` for additional information. + Only parse GEFX files you trust. + +GEXF (Graph Exchange XML Format) is a language for describing complex +network structures, their associated data and dynamics. + +This implementation does not support mixed graphs (directed and +undirected edges together). + +Format +------ +GEXF is an XML format. See http://gexf.net/schema.html for the +specification and http://gexf.net/basic.html for examples. +""" + +import itertools +import time +from xml.etree.ElementTree import ( + Element, + ElementTree, + SubElement, + register_namespace, + tostring, +) + +import networkx as nx +from networkx.utils import open_file + +__all__ = ["write_gexf", "read_gexf", "relabel_gexf_graph", "generate_gexf"] + + +@open_file(1, mode="wb") +def write_gexf(G, path, encoding="utf-8", prettyprint=True, version="1.2draft"): + """Write G in GEXF format to path. + + "GEXF (Graph Exchange XML Format) is a language for describing + complex networks structures, their associated data and dynamics" [1]_. + + Node attributes are checked according to the version of the GEXF + schemas used for parameters which are not user defined, + e.g. visualization 'viz' [2]_. See example for usage. + + Parameters + ---------- + G : graph + A NetworkX graph + path : file or string + File or file name to write. + File names ending in .gz or .bz2 will be compressed. + encoding : string (optional, default: 'utf-8') + Encoding for text data. + prettyprint : bool (optional, default: True) + If True use line breaks and indenting in output XML. + version: string (optional, default: '1.2draft') + The version of GEXF to be used for nodes attributes checking + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_gexf(G, "test.gexf") + + # visualization data + >>> G.nodes[0]["viz"] = {"size": 54} + >>> G.nodes[0]["viz"]["position"] = {"x": 0, "y": 1} + >>> G.nodes[0]["viz"]["color"] = {"r": 0, "g": 0, "b": 256} + + + Notes + ----- + This implementation does not support mixed graphs (directed and undirected + edges together). + + The node id attribute is set to be the string of the node label. + If you want to specify an id use set it as node data, e.g. + node['a']['id']=1 to set the id of node 'a' to 1. + + References + ---------- + .. [1] GEXF File Format, http://gexf.net/ + .. [2] GEXF schema, http://gexf.net/schema.html + """ + writer = GEXFWriter(encoding=encoding, prettyprint=prettyprint, version=version) + writer.add_graph(G) + writer.write(path) + + +def generate_gexf(G, encoding="utf-8", prettyprint=True, version="1.2draft"): + """Generate lines of GEXF format representation of G. + + "GEXF (Graph Exchange XML Format) is a language for describing + complex networks structures, their associated data and dynamics" [1]_. + + Parameters + ---------- + G : graph + A NetworkX graph + encoding : string (optional, default: 'utf-8') + Encoding for text data. + prettyprint : bool (optional, default: True) + If True use line breaks and indenting in output XML. + version : string (default: 1.2draft) + Version of GEFX File Format (see http://gexf.net/schema.html) + Supported values: "1.1draft", "1.2draft" + + + Examples + -------- + >>> G = nx.path_graph(4) + >>> linefeed = chr(10) # linefeed=\n + >>> s = linefeed.join(nx.generate_gexf(G)) + >>> for line in nx.generate_gexf(G): # doctest: +SKIP + ... print(line) + + Notes + ----- + This implementation does not support mixed graphs (directed and undirected + edges together). + + The node id attribute is set to be the string of the node label. + If you want to specify an id use set it as node data, e.g. + node['a']['id']=1 to set the id of node 'a' to 1. + + References + ---------- + .. [1] GEXF File Format, https://gephi.org/gexf/format/ + """ + writer = GEXFWriter(encoding=encoding, prettyprint=prettyprint, version=version) + writer.add_graph(G) + yield from str(writer).splitlines() + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_gexf(path, node_type=None, relabel=False, version="1.2draft"): + """Read graph in GEXF format from path. + + "GEXF (Graph Exchange XML Format) is a language for describing + complex networks structures, their associated data and dynamics" [1]_. + + Parameters + ---------- + path : file or string + File or file name to read. + File names ending in .gz or .bz2 will be decompressed. + node_type: Python type (default: None) + Convert node ids to this type if not None. + relabel : bool (default: False) + If True relabel the nodes to use the GEXF node "label" attribute + instead of the node "id" attribute as the NetworkX node label. + version : string (default: 1.2draft) + Version of GEFX File Format (see http://gexf.net/schema.html) + Supported values: "1.1draft", "1.2draft" + + Returns + ------- + graph: NetworkX graph + If no parallel edges are found a Graph or DiGraph is returned. + Otherwise a MultiGraph or MultiDiGraph is returned. + + Notes + ----- + This implementation does not support mixed graphs (directed and undirected + edges together). + + References + ---------- + .. [1] GEXF File Format, http://gexf.net/ + """ + reader = GEXFReader(node_type=node_type, version=version) + if relabel: + G = relabel_gexf_graph(reader(path)) + else: + G = reader(path) + return G + + +class GEXF: + versions = { + "1.1draft": { + "NS_GEXF": "http://www.gexf.net/1.1draft", + "NS_VIZ": "http://www.gexf.net/1.1draft/viz", + "NS_XSI": "http://www.w3.org/2001/XMLSchema-instance", + "SCHEMALOCATION": " ".join( + [ + "http://www.gexf.net/1.1draft", + "http://www.gexf.net/1.1draft/gexf.xsd", + ] + ), + "VERSION": "1.1", + }, + "1.2draft": { + "NS_GEXF": "http://www.gexf.net/1.2draft", + "NS_VIZ": "http://www.gexf.net/1.2draft/viz", + "NS_XSI": "http://www.w3.org/2001/XMLSchema-instance", + "SCHEMALOCATION": " ".join( + [ + "http://www.gexf.net/1.2draft", + "http://www.gexf.net/1.2draft/gexf.xsd", + ] + ), + "VERSION": "1.2", + }, + } + + def construct_types(self): + types = [ + (int, "integer"), + (float, "float"), + (float, "double"), + (bool, "boolean"), + (list, "string"), + (dict, "string"), + (int, "long"), + (str, "liststring"), + (str, "anyURI"), + (str, "string"), + ] + + # These additions to types allow writing numpy types + try: + import numpy as np + except ImportError: + pass + else: + # prepend so that python types are created upon read (last entry wins) + types = [ + (np.float64, "float"), + (np.float32, "float"), + (np.float16, "float"), + (np.int_, "int"), + (np.int8, "int"), + (np.int16, "int"), + (np.int32, "int"), + (np.int64, "int"), + (np.uint8, "int"), + (np.uint16, "int"), + (np.uint32, "int"), + (np.uint64, "int"), + (np.int_, "int"), + (np.intc, "int"), + (np.intp, "int"), + ] + types + + self.xml_type = dict(types) + self.python_type = dict(reversed(a) for a in types) + + # http://www.w3.org/TR/xmlschema-2/#boolean + convert_bool = { + "true": True, + "false": False, + "True": True, + "False": False, + "0": False, + 0: False, + "1": True, + 1: True, + } + + def set_version(self, version): + d = self.versions.get(version) + if d is None: + raise nx.NetworkXError(f"Unknown GEXF version {version}.") + self.NS_GEXF = d["NS_GEXF"] + self.NS_VIZ = d["NS_VIZ"] + self.NS_XSI = d["NS_XSI"] + self.SCHEMALOCATION = d["SCHEMALOCATION"] + self.VERSION = d["VERSION"] + self.version = version + + +class GEXFWriter(GEXF): + # class for writing GEXF format files + # use write_gexf() function + def __init__( + self, graph=None, encoding="utf-8", prettyprint=True, version="1.2draft" + ): + self.construct_types() + self.prettyprint = prettyprint + self.encoding = encoding + self.set_version(version) + self.xml = Element( + "gexf", + { + "xmlns": self.NS_GEXF, + "xmlns:xsi": self.NS_XSI, + "xsi:schemaLocation": self.SCHEMALOCATION, + "version": self.VERSION, + }, + ) + + # Make meta element a non-graph element + # Also add lastmodifieddate as attribute, not tag + meta_element = Element("meta") + subelement_text = f"NetworkX {nx.__version__}" + SubElement(meta_element, "creator").text = subelement_text + meta_element.set("lastmodifieddate", time.strftime("%Y-%m-%d")) + self.xml.append(meta_element) + + register_namespace("viz", self.NS_VIZ) + + # counters for edge and attribute identifiers + self.edge_id = itertools.count() + self.attr_id = itertools.count() + self.all_edge_ids = set() + # default attributes are stored in dictionaries + self.attr = {} + self.attr["node"] = {} + self.attr["edge"] = {} + self.attr["node"]["dynamic"] = {} + self.attr["node"]["static"] = {} + self.attr["edge"]["dynamic"] = {} + self.attr["edge"]["static"] = {} + + if graph is not None: + self.add_graph(graph) + + def __str__(self): + if self.prettyprint: + self.indent(self.xml) + s = tostring(self.xml).decode(self.encoding) + return s + + def add_graph(self, G): + # first pass through G collecting edge ids + for u, v, dd in G.edges(data=True): + eid = dd.get("id") + if eid is not None: + self.all_edge_ids.add(str(eid)) + # set graph attributes + if G.graph.get("mode") == "dynamic": + mode = "dynamic" + else: + mode = "static" + # Add a graph element to the XML + if G.is_directed(): + default = "directed" + else: + default = "undirected" + name = G.graph.get("name", "") + graph_element = Element("graph", defaultedgetype=default, mode=mode, name=name) + self.graph_element = graph_element + self.add_nodes(G, graph_element) + self.add_edges(G, graph_element) + self.xml.append(graph_element) + + def add_nodes(self, G, graph_element): + nodes_element = Element("nodes") + for node, data in G.nodes(data=True): + node_data = data.copy() + node_id = str(node_data.pop("id", node)) + kw = {"id": node_id} + label = str(node_data.pop("label", node)) + kw["label"] = label + try: + pid = node_data.pop("pid") + kw["pid"] = str(pid) + except KeyError: + pass + try: + start = node_data.pop("start") + kw["start"] = str(start) + self.alter_graph_mode_timeformat(start) + except KeyError: + pass + try: + end = node_data.pop("end") + kw["end"] = str(end) + self.alter_graph_mode_timeformat(end) + except KeyError: + pass + # add node element with attributes + node_element = Element("node", **kw) + # add node element and attr subelements + default = G.graph.get("node_default", {}) + node_data = self.add_parents(node_element, node_data) + if self.VERSION == "1.1": + node_data = self.add_slices(node_element, node_data) + else: + node_data = self.add_spells(node_element, node_data) + node_data = self.add_viz(node_element, node_data) + node_data = self.add_attributes("node", node_element, node_data, default) + nodes_element.append(node_element) + graph_element.append(nodes_element) + + def add_edges(self, G, graph_element): + def edge_key_data(G): + # helper function to unify multigraph and graph edge iterator + if G.is_multigraph(): + for u, v, key, data in G.edges(data=True, keys=True): + edge_data = data.copy() + edge_data.update(key=key) + edge_id = edge_data.pop("id", None) + if edge_id is None: + edge_id = next(self.edge_id) + while str(edge_id) in self.all_edge_ids: + edge_id = next(self.edge_id) + self.all_edge_ids.add(str(edge_id)) + yield u, v, edge_id, edge_data + else: + for u, v, data in G.edges(data=True): + edge_data = data.copy() + edge_id = edge_data.pop("id", None) + if edge_id is None: + edge_id = next(self.edge_id) + while str(edge_id) in self.all_edge_ids: + edge_id = next(self.edge_id) + self.all_edge_ids.add(str(edge_id)) + yield u, v, edge_id, edge_data + + edges_element = Element("edges") + for u, v, key, edge_data in edge_key_data(G): + kw = {"id": str(key)} + try: + edge_label = edge_data.pop("label") + kw["label"] = str(edge_label) + except KeyError: + pass + try: + edge_weight = edge_data.pop("weight") + kw["weight"] = str(edge_weight) + except KeyError: + pass + try: + edge_type = edge_data.pop("type") + kw["type"] = str(edge_type) + except KeyError: + pass + try: + start = edge_data.pop("start") + kw["start"] = str(start) + self.alter_graph_mode_timeformat(start) + except KeyError: + pass + try: + end = edge_data.pop("end") + kw["end"] = str(end) + self.alter_graph_mode_timeformat(end) + except KeyError: + pass + source_id = str(G.nodes[u].get("id", u)) + target_id = str(G.nodes[v].get("id", v)) + edge_element = Element("edge", source=source_id, target=target_id, **kw) + default = G.graph.get("edge_default", {}) + if self.VERSION == "1.1": + edge_data = self.add_slices(edge_element, edge_data) + else: + edge_data = self.add_spells(edge_element, edge_data) + edge_data = self.add_viz(edge_element, edge_data) + edge_data = self.add_attributes("edge", edge_element, edge_data, default) + edges_element.append(edge_element) + graph_element.append(edges_element) + + def add_attributes(self, node_or_edge, xml_obj, data, default): + # Add attrvalues to node or edge + attvalues = Element("attvalues") + if len(data) == 0: + return data + mode = "static" + for k, v in data.items(): + # rename generic multigraph key to avoid any name conflict + if k == "key": + k = "networkx_key" + val_type = type(v) + if val_type not in self.xml_type: + raise TypeError(f"attribute value type is not allowed: {val_type}") + if isinstance(v, list): + # dynamic data + for val, start, end in v: + val_type = type(val) + if start is not None or end is not None: + mode = "dynamic" + self.alter_graph_mode_timeformat(start) + self.alter_graph_mode_timeformat(end) + break + attr_id = self.get_attr_id( + str(k), self.xml_type[val_type], node_or_edge, default, mode + ) + for val, start, end in v: + e = Element("attvalue") + e.attrib["for"] = attr_id + e.attrib["value"] = str(val) + # Handle nan, inf, -inf differently + if val_type == float: + if e.attrib["value"] == "inf": + e.attrib["value"] = "INF" + elif e.attrib["value"] == "nan": + e.attrib["value"] = "NaN" + elif e.attrib["value"] == "-inf": + e.attrib["value"] = "-INF" + if start is not None: + e.attrib["start"] = str(start) + if end is not None: + e.attrib["end"] = str(end) + attvalues.append(e) + else: + # static data + mode = "static" + attr_id = self.get_attr_id( + str(k), self.xml_type[val_type], node_or_edge, default, mode + ) + e = Element("attvalue") + e.attrib["for"] = attr_id + if isinstance(v, bool): + e.attrib["value"] = str(v).lower() + else: + e.attrib["value"] = str(v) + # Handle float nan, inf, -inf differently + if val_type == float: + if e.attrib["value"] == "inf": + e.attrib["value"] = "INF" + elif e.attrib["value"] == "nan": + e.attrib["value"] = "NaN" + elif e.attrib["value"] == "-inf": + e.attrib["value"] = "-INF" + attvalues.append(e) + xml_obj.append(attvalues) + return data + + def get_attr_id(self, title, attr_type, edge_or_node, default, mode): + # find the id of the attribute or generate a new id + try: + return self.attr[edge_or_node][mode][title] + except KeyError: + # generate new id + new_id = str(next(self.attr_id)) + self.attr[edge_or_node][mode][title] = new_id + attr_kwargs = {"id": new_id, "title": title, "type": attr_type} + attribute = Element("attribute", **attr_kwargs) + # add subelement for data default value if present + default_title = default.get(title) + if default_title is not None: + default_element = Element("default") + default_element.text = str(default_title) + attribute.append(default_element) + # new insert it into the XML + attributes_element = None + for a in self.graph_element.findall("attributes"): + # find existing attributes element by class and mode + a_class = a.get("class") + a_mode = a.get("mode", "static") + if a_class == edge_or_node and a_mode == mode: + attributes_element = a + if attributes_element is None: + # create new attributes element + attr_kwargs = {"mode": mode, "class": edge_or_node} + attributes_element = Element("attributes", **attr_kwargs) + self.graph_element.insert(0, attributes_element) + attributes_element.append(attribute) + return new_id + + def add_viz(self, element, node_data): + viz = node_data.pop("viz", False) + if viz: + color = viz.get("color") + if color is not None: + if self.VERSION == "1.1": + e = Element( + f"{{{self.NS_VIZ}}}color", + r=str(color.get("r")), + g=str(color.get("g")), + b=str(color.get("b")), + ) + else: + e = Element( + f"{{{self.NS_VIZ}}}color", + r=str(color.get("r")), + g=str(color.get("g")), + b=str(color.get("b")), + a=str(color.get("a", 1.0)), + ) + element.append(e) + + size = viz.get("size") + if size is not None: + e = Element(f"{{{self.NS_VIZ}}}size", value=str(size)) + element.append(e) + + thickness = viz.get("thickness") + if thickness is not None: + e = Element(f"{{{self.NS_VIZ}}}thickness", value=str(thickness)) + element.append(e) + + shape = viz.get("shape") + if shape is not None: + if shape.startswith("http"): + e = Element( + f"{{{self.NS_VIZ}}}shape", value="image", uri=str(shape) + ) + else: + e = Element(f"{{{self.NS_VIZ}}}shape", value=str(shape)) + element.append(e) + + position = viz.get("position") + if position is not None: + e = Element( + f"{{{self.NS_VIZ}}}position", + x=str(position.get("x")), + y=str(position.get("y")), + z=str(position.get("z")), + ) + element.append(e) + return node_data + + def add_parents(self, node_element, node_data): + parents = node_data.pop("parents", False) + if parents: + parents_element = Element("parents") + for p in parents: + e = Element("parent") + e.attrib["for"] = str(p) + parents_element.append(e) + node_element.append(parents_element) + return node_data + + def add_slices(self, node_or_edge_element, node_or_edge_data): + slices = node_or_edge_data.pop("slices", False) + if slices: + slices_element = Element("slices") + for start, end in slices: + e = Element("slice", start=str(start), end=str(end)) + slices_element.append(e) + node_or_edge_element.append(slices_element) + return node_or_edge_data + + def add_spells(self, node_or_edge_element, node_or_edge_data): + spells = node_or_edge_data.pop("spells", False) + if spells: + spells_element = Element("spells") + for start, end in spells: + e = Element("spell") + if start is not None: + e.attrib["start"] = str(start) + self.alter_graph_mode_timeformat(start) + if end is not None: + e.attrib["end"] = str(end) + self.alter_graph_mode_timeformat(end) + spells_element.append(e) + node_or_edge_element.append(spells_element) + return node_or_edge_data + + def alter_graph_mode_timeformat(self, start_or_end): + # If 'start' or 'end' appears, alter Graph mode to dynamic and + # set timeformat + if self.graph_element.get("mode") == "static": + if start_or_end is not None: + if isinstance(start_or_end, str): + timeformat = "date" + elif isinstance(start_or_end, float): + timeformat = "double" + elif isinstance(start_or_end, int): + timeformat = "long" + else: + raise nx.NetworkXError( + "timeformat should be of the type int, float or str" + ) + self.graph_element.set("timeformat", timeformat) + self.graph_element.set("mode", "dynamic") + + def write(self, fh): + # Serialize graph G in GEXF to the open fh + if self.prettyprint: + self.indent(self.xml) + document = ElementTree(self.xml) + document.write(fh, encoding=self.encoding, xml_declaration=True) + + def indent(self, elem, level=0): + # in-place prettyprint formatter + i = "\n" + " " * level + if len(elem): + if not elem.text or not elem.text.strip(): + elem.text = i + " " + if not elem.tail or not elem.tail.strip(): + elem.tail = i + for elem in elem: + self.indent(elem, level + 1) + if not elem.tail or not elem.tail.strip(): + elem.tail = i + else: + if level and (not elem.tail or not elem.tail.strip()): + elem.tail = i + + +class GEXFReader(GEXF): + # Class to read GEXF format files + # use read_gexf() function + def __init__(self, node_type=None, version="1.2draft"): + self.construct_types() + self.node_type = node_type + # assume simple graph and test for multigraph on read + self.simple_graph = True + self.set_version(version) + + def __call__(self, stream): + self.xml = ElementTree(file=stream) + g = self.xml.find(f"{{{self.NS_GEXF}}}graph") + if g is not None: + return self.make_graph(g) + # try all the versions + for version in self.versions: + self.set_version(version) + g = self.xml.find(f"{{{self.NS_GEXF}}}graph") + if g is not None: + return self.make_graph(g) + raise nx.NetworkXError("No <graph> element in GEXF file.") + + def make_graph(self, graph_xml): + # start with empty DiGraph or MultiDiGraph + edgedefault = graph_xml.get("defaultedgetype", None) + if edgedefault == "directed": + G = nx.MultiDiGraph() + else: + G = nx.MultiGraph() + + # graph attributes + graph_name = graph_xml.get("name", "") + if graph_name != "": + G.graph["name"] = graph_name + graph_start = graph_xml.get("start") + if graph_start is not None: + G.graph["start"] = graph_start + graph_end = graph_xml.get("end") + if graph_end is not None: + G.graph["end"] = graph_end + graph_mode = graph_xml.get("mode", "") + if graph_mode == "dynamic": + G.graph["mode"] = "dynamic" + else: + G.graph["mode"] = "static" + + # timeformat + self.timeformat = graph_xml.get("timeformat") + if self.timeformat == "date": + self.timeformat = "string" + + # node and edge attributes + attributes_elements = graph_xml.findall(f"{{{self.NS_GEXF}}}attributes") + # dictionaries to hold attributes and attribute defaults + node_attr = {} + node_default = {} + edge_attr = {} + edge_default = {} + for a in attributes_elements: + attr_class = a.get("class") + if attr_class == "node": + na, nd = self.find_gexf_attributes(a) + node_attr.update(na) + node_default.update(nd) + G.graph["node_default"] = node_default + elif attr_class == "edge": + ea, ed = self.find_gexf_attributes(a) + edge_attr.update(ea) + edge_default.update(ed) + G.graph["edge_default"] = edge_default + else: + raise # unknown attribute class + + # Hack to handle Gephi0.7beta bug + # add weight attribute + ea = {"weight": {"type": "double", "mode": "static", "title": "weight"}} + ed = {} + edge_attr.update(ea) + edge_default.update(ed) + G.graph["edge_default"] = edge_default + + # add nodes + nodes_element = graph_xml.find(f"{{{self.NS_GEXF}}}nodes") + if nodes_element is not None: + for node_xml in nodes_element.findall(f"{{{self.NS_GEXF}}}node"): + self.add_node(G, node_xml, node_attr) + + # add edges + edges_element = graph_xml.find(f"{{{self.NS_GEXF}}}edges") + if edges_element is not None: + for edge_xml in edges_element.findall(f"{{{self.NS_GEXF}}}edge"): + self.add_edge(G, edge_xml, edge_attr) + + # switch to Graph or DiGraph if no parallel edges were found. + if self.simple_graph: + if G.is_directed(): + G = nx.DiGraph(G) + else: + G = nx.Graph(G) + return G + + def add_node(self, G, node_xml, node_attr, node_pid=None): + # add a single node with attributes to the graph + + # get attributes and subattributues for node + data = self.decode_attr_elements(node_attr, node_xml) + data = self.add_parents(data, node_xml) # add any parents + if self.VERSION == "1.1": + data = self.add_slices(data, node_xml) # add slices + else: + data = self.add_spells(data, node_xml) # add spells + data = self.add_viz(data, node_xml) # add viz + data = self.add_start_end(data, node_xml) # add start/end + + # find the node id and cast it to the appropriate type + node_id = node_xml.get("id") + if self.node_type is not None: + node_id = self.node_type(node_id) + + # every node should have a label + node_label = node_xml.get("label") + data["label"] = node_label + + # parent node id + node_pid = node_xml.get("pid", node_pid) + if node_pid is not None: + data["pid"] = node_pid + + # check for subnodes, recursive + subnodes = node_xml.find(f"{{{self.NS_GEXF}}}nodes") + if subnodes is not None: + for node_xml in subnodes.findall(f"{{{self.NS_GEXF}}}node"): + self.add_node(G, node_xml, node_attr, node_pid=node_id) + + G.add_node(node_id, **data) + + def add_start_end(self, data, xml): + # start and end times + ttype = self.timeformat + node_start = xml.get("start") + if node_start is not None: + data["start"] = self.python_type[ttype](node_start) + node_end = xml.get("end") + if node_end is not None: + data["end"] = self.python_type[ttype](node_end) + return data + + def add_viz(self, data, node_xml): + # add viz element for node + viz = {} + color = node_xml.find(f"{{{self.NS_VIZ}}}color") + if color is not None: + if self.VERSION == "1.1": + viz["color"] = { + "r": int(color.get("r")), + "g": int(color.get("g")), + "b": int(color.get("b")), + } + else: + viz["color"] = { + "r": int(color.get("r")), + "g": int(color.get("g")), + "b": int(color.get("b")), + "a": float(color.get("a", 1)), + } + + size = node_xml.find(f"{{{self.NS_VIZ}}}size") + if size is not None: + viz["size"] = float(size.get("value")) + + thickness = node_xml.find(f"{{{self.NS_VIZ}}}thickness") + if thickness is not None: + viz["thickness"] = float(thickness.get("value")) + + shape = node_xml.find(f"{{{self.NS_VIZ}}}shape") + if shape is not None: + viz["shape"] = shape.get("shape") + if viz["shape"] == "image": + viz["shape"] = shape.get("uri") + + position = node_xml.find(f"{{{self.NS_VIZ}}}position") + if position is not None: + viz["position"] = { + "x": float(position.get("x", 0)), + "y": float(position.get("y", 0)), + "z": float(position.get("z", 0)), + } + + if len(viz) > 0: + data["viz"] = viz + return data + + def add_parents(self, data, node_xml): + parents_element = node_xml.find(f"{{{self.NS_GEXF}}}parents") + if parents_element is not None: + data["parents"] = [] + for p in parents_element.findall(f"{{{self.NS_GEXF}}}parent"): + parent = p.get("for") + data["parents"].append(parent) + return data + + def add_slices(self, data, node_or_edge_xml): + slices_element = node_or_edge_xml.find(f"{{{self.NS_GEXF}}}slices") + if slices_element is not None: + data["slices"] = [] + for s in slices_element.findall(f"{{{self.NS_GEXF}}}slice"): + start = s.get("start") + end = s.get("end") + data["slices"].append((start, end)) + return data + + def add_spells(self, data, node_or_edge_xml): + spells_element = node_or_edge_xml.find(f"{{{self.NS_GEXF}}}spells") + if spells_element is not None: + data["spells"] = [] + ttype = self.timeformat + for s in spells_element.findall(f"{{{self.NS_GEXF}}}spell"): + start = self.python_type[ttype](s.get("start")) + end = self.python_type[ttype](s.get("end")) + data["spells"].append((start, end)) + return data + + def add_edge(self, G, edge_element, edge_attr): + # add an edge to the graph + + # raise error if we find mixed directed and undirected edges + edge_direction = edge_element.get("type") + if G.is_directed() and edge_direction == "undirected": + raise nx.NetworkXError("Undirected edge found in directed graph.") + if (not G.is_directed()) and edge_direction == "directed": + raise nx.NetworkXError("Directed edge found in undirected graph.") + + # Get source and target and recast type if required + source = edge_element.get("source") + target = edge_element.get("target") + if self.node_type is not None: + source = self.node_type(source) + target = self.node_type(target) + + data = self.decode_attr_elements(edge_attr, edge_element) + data = self.add_start_end(data, edge_element) + + if self.VERSION == "1.1": + data = self.add_slices(data, edge_element) # add slices + else: + data = self.add_spells(data, edge_element) # add spells + + # GEXF stores edge ids as an attribute + # NetworkX uses them as keys in multigraphs + # if networkx_key is not specified as an attribute + edge_id = edge_element.get("id") + if edge_id is not None: + data["id"] = edge_id + + # check if there is a 'multigraph_key' and use that as edge_id + multigraph_key = data.pop("networkx_key", None) + if multigraph_key is not None: + edge_id = multigraph_key + + weight = edge_element.get("weight") + if weight is not None: + data["weight"] = float(weight) + + edge_label = edge_element.get("label") + if edge_label is not None: + data["label"] = edge_label + + if G.has_edge(source, target): + # seen this edge before - this is a multigraph + self.simple_graph = False + G.add_edge(source, target, key=edge_id, **data) + if edge_direction == "mutual": + G.add_edge(target, source, key=edge_id, **data) + + def decode_attr_elements(self, gexf_keys, obj_xml): + # Use the key information to decode the attr XML + attr = {} + # look for outer '<attvalues>' element + attr_element = obj_xml.find(f"{{{self.NS_GEXF}}}attvalues") + if attr_element is not None: + # loop over <attvalue> elements + for a in attr_element.findall(f"{{{self.NS_GEXF}}}attvalue"): + key = a.get("for") # for is required + try: # should be in our gexf_keys dictionary + title = gexf_keys[key]["title"] + except KeyError as err: + raise nx.NetworkXError(f"No attribute defined for={key}.") from err + atype = gexf_keys[key]["type"] + value = a.get("value") + if atype == "boolean": + value = self.convert_bool[value] + else: + value = self.python_type[atype](value) + if gexf_keys[key]["mode"] == "dynamic": + # for dynamic graphs use list of three-tuples + # [(value1,start1,end1), (value2,start2,end2), etc] + ttype = self.timeformat + start = self.python_type[ttype](a.get("start")) + end = self.python_type[ttype](a.get("end")) + if title in attr: + attr[title].append((value, start, end)) + else: + attr[title] = [(value, start, end)] + else: + # for static graphs just assign the value + attr[title] = value + return attr + + def find_gexf_attributes(self, attributes_element): + # Extract all the attributes and defaults + attrs = {} + defaults = {} + mode = attributes_element.get("mode") + for k in attributes_element.findall(f"{{{self.NS_GEXF}}}attribute"): + attr_id = k.get("id") + title = k.get("title") + atype = k.get("type") + attrs[attr_id] = {"title": title, "type": atype, "mode": mode} + # check for the 'default' subelement of key element and add + default = k.find(f"{{{self.NS_GEXF}}}default") + if default is not None: + if atype == "boolean": + value = self.convert_bool[default.text] + else: + value = self.python_type[atype](default.text) + defaults[title] = value + return attrs, defaults + + +def relabel_gexf_graph(G): + """Relabel graph using "label" node keyword for node label. + + Parameters + ---------- + G : graph + A NetworkX graph read from GEXF data + + Returns + ------- + H : graph + A NetworkX graph with relabeled nodes + + Raises + ------ + NetworkXError + If node labels are missing or not unique while relabel=True. + + Notes + ----- + This function relabels the nodes in a NetworkX graph with the + "label" attribute. It also handles relabeling the specific GEXF + node attributes "parents", and "pid". + """ + # build mapping of node labels, do some error checking + try: + mapping = [(u, G.nodes[u]["label"]) for u in G] + except KeyError as err: + raise nx.NetworkXError( + "Failed to relabel nodes: missing node labels found. Use relabel=False." + ) from err + x, y = zip(*mapping) + if len(set(y)) != len(G): + raise nx.NetworkXError( + "Failed to relabel nodes: " + "duplicate node labels found. " + "Use relabel=False." + ) + mapping = dict(mapping) + H = nx.relabel_nodes(G, mapping) + # relabel attributes + for n in G: + m = mapping[n] + H.nodes[m]["id"] = n + H.nodes[m].pop("label") + if "pid" in H.nodes[m]: + H.nodes[m]["pid"] = mapping[G.nodes[n]["pid"]] + if "parents" in H.nodes[m]: + H.nodes[m]["parents"] = [mapping[p] for p in G.nodes[n]["parents"]] + return H diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/gml.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/gml.py new file mode 100644 index 00000000..891d7096 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/gml.py @@ -0,0 +1,879 @@ +""" +Read graphs in GML format. + +"GML, the Graph Modelling Language, is our proposal for a portable +file format for graphs. GML's key features are portability, simple +syntax, extensibility and flexibility. A GML file consists of a +hierarchical key-value lists. Graphs can be annotated with arbitrary +data structures. The idea for a common file format was born at the +GD'95; this proposal is the outcome of many discussions. GML is the +standard file format in the Graphlet graph editor system. It has been +overtaken and adapted by several other systems for drawing graphs." + +GML files are stored using a 7-bit ASCII encoding with any extended +ASCII characters (iso8859-1) appearing as HTML character entities. +You will need to give some thought into how the exported data should +interact with different languages and even different Python versions. +Re-importing from gml is also a concern. + +Without specifying a `stringizer`/`destringizer`, the code is capable of +writing `int`/`float`/`str`/`dict`/`list` data as required by the GML +specification. For writing other data types, and for reading data other +than `str` you need to explicitly supply a `stringizer`/`destringizer`. + +For additional documentation on the GML file format, please see the +`GML website <https://web.archive.org/web/20190207140002/http://www.fim.uni-passau.de/index.php?id=17297&L=1>`_. + +Several example graphs in GML format may be found on Mark Newman's +`Network data page <http://www-personal.umich.edu/~mejn/netdata/>`_. +""" + +import html.entities as htmlentitydefs +import re +import warnings +from ast import literal_eval +from collections import defaultdict +from enum import Enum +from io import StringIO +from typing import Any, NamedTuple + +import networkx as nx +from networkx.exception import NetworkXError +from networkx.utils import open_file + +__all__ = ["read_gml", "parse_gml", "generate_gml", "write_gml"] + + +def escape(text): + """Use XML character references to escape characters. + + Use XML character references for unprintable or non-ASCII + characters, double quotes and ampersands in a string + """ + + def fixup(m): + ch = m.group(0) + return "&#" + str(ord(ch)) + ";" + + text = re.sub('[^ -~]|[&"]', fixup, text) + return text if isinstance(text, str) else str(text) + + +def unescape(text): + """Replace XML character references with the referenced characters""" + + def fixup(m): + text = m.group(0) + if text[1] == "#": + # Character reference + if text[2] == "x": + code = int(text[3:-1], 16) + else: + code = int(text[2:-1]) + else: + # Named entity + try: + code = htmlentitydefs.name2codepoint[text[1:-1]] + except KeyError: + return text # leave unchanged + try: + return chr(code) + except (ValueError, OverflowError): + return text # leave unchanged + + return re.sub("&(?:[0-9A-Za-z]+|#(?:[0-9]+|x[0-9A-Fa-f]+));", fixup, text) + + +def literal_destringizer(rep): + """Convert a Python literal to the value it represents. + + Parameters + ---------- + rep : string + A Python literal. + + Returns + ------- + value : object + The value of the Python literal. + + Raises + ------ + ValueError + If `rep` is not a Python literal. + """ + if isinstance(rep, str): + orig_rep = rep + try: + return literal_eval(rep) + except SyntaxError as err: + raise ValueError(f"{orig_rep!r} is not a valid Python literal") from err + else: + raise ValueError(f"{rep!r} is not a string") + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_gml(path, label="label", destringizer=None): + """Read graph in GML format from `path`. + + Parameters + ---------- + path : filename or filehandle + The filename or filehandle to read from. + + label : string, optional + If not None, the parsed nodes will be renamed according to node + attributes indicated by `label`. Default value: 'label'. + + destringizer : callable, optional + A `destringizer` that recovers values stored as strings in GML. If it + cannot convert a string to a value, a `ValueError` is raised. Default + value : None. + + Returns + ------- + G : NetworkX graph + The parsed graph. + + Raises + ------ + NetworkXError + If the input cannot be parsed. + + See Also + -------- + write_gml, parse_gml + literal_destringizer + + Notes + ----- + GML files are stored using a 7-bit ASCII encoding with any extended + ASCII characters (iso8859-1) appearing as HTML character entities. + Without specifying a `stringizer`/`destringizer`, the code is capable of + writing `int`/`float`/`str`/`dict`/`list` data as required by the GML + specification. For writing other data types, and for reading data other + than `str` you need to explicitly supply a `stringizer`/`destringizer`. + + For additional documentation on the GML file format, please see the + `GML url <https://web.archive.org/web/20190207140002/http://www.fim.uni-passau.de/index.php?id=17297&L=1>`_. + + See the module docstring :mod:`networkx.readwrite.gml` for more details. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_gml(G, "test.gml") + + GML values are interpreted as strings by default: + + >>> H = nx.read_gml("test.gml") + >>> H.nodes + NodeView(('0', '1', '2', '3')) + + When a `destringizer` is provided, GML values are converted to the provided type. + For example, integer nodes can be recovered as shown below: + + >>> J = nx.read_gml("test.gml", destringizer=int) + >>> J.nodes + NodeView((0, 1, 2, 3)) + + """ + + def filter_lines(lines): + for line in lines: + try: + line = line.decode("ascii") + except UnicodeDecodeError as err: + raise NetworkXError("input is not ASCII-encoded") from err + if not isinstance(line, str): + lines = str(lines) + if line and line[-1] == "\n": + line = line[:-1] + yield line + + G = parse_gml_lines(filter_lines(path), label, destringizer) + return G + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_gml(lines, label="label", destringizer=None): + """Parse GML graph from a string or iterable. + + Parameters + ---------- + lines : string or iterable of strings + Data in GML format. + + label : string, optional + If not None, the parsed nodes will be renamed according to node + attributes indicated by `label`. Default value: 'label'. + + destringizer : callable, optional + A `destringizer` that recovers values stored as strings in GML. If it + cannot convert a string to a value, a `ValueError` is raised. Default + value : None. + + Returns + ------- + G : NetworkX graph + The parsed graph. + + Raises + ------ + NetworkXError + If the input cannot be parsed. + + See Also + -------- + write_gml, read_gml + + Notes + ----- + This stores nested GML attributes as dictionaries in the NetworkX graph, + node, and edge attribute structures. + + GML files are stored using a 7-bit ASCII encoding with any extended + ASCII characters (iso8859-1) appearing as HTML character entities. + Without specifying a `stringizer`/`destringizer`, the code is capable of + writing `int`/`float`/`str`/`dict`/`list` data as required by the GML + specification. For writing other data types, and for reading data other + than `str` you need to explicitly supply a `stringizer`/`destringizer`. + + For additional documentation on the GML file format, please see the + `GML url <https://web.archive.org/web/20190207140002/http://www.fim.uni-passau.de/index.php?id=17297&L=1>`_. + + See the module docstring :mod:`networkx.readwrite.gml` for more details. + """ + + def decode_line(line): + if isinstance(line, bytes): + try: + line.decode("ascii") + except UnicodeDecodeError as err: + raise NetworkXError("input is not ASCII-encoded") from err + if not isinstance(line, str): + line = str(line) + return line + + def filter_lines(lines): + if isinstance(lines, str): + lines = decode_line(lines) + lines = lines.splitlines() + yield from lines + else: + for line in lines: + line = decode_line(line) + if line and line[-1] == "\n": + line = line[:-1] + if line.find("\n") != -1: + raise NetworkXError("input line contains newline") + yield line + + G = parse_gml_lines(filter_lines(lines), label, destringizer) + return G + + +class Pattern(Enum): + """encodes the index of each token-matching pattern in `tokenize`.""" + + KEYS = 0 + REALS = 1 + INTS = 2 + STRINGS = 3 + DICT_START = 4 + DICT_END = 5 + COMMENT_WHITESPACE = 6 + + +class Token(NamedTuple): + category: Pattern + value: Any + line: int + position: int + + +LIST_START_VALUE = "_networkx_list_start" + + +def parse_gml_lines(lines, label, destringizer): + """Parse GML `lines` into a graph.""" + + def tokenize(): + patterns = [ + r"[A-Za-z][0-9A-Za-z_]*\b", # keys + # reals + r"[+-]?(?:[0-9]*\.[0-9]+|[0-9]+\.[0-9]*|INF)(?:[Ee][+-]?[0-9]+)?", + r"[+-]?[0-9]+", # ints + r'".*?"', # strings + r"\[", # dict start + r"\]", # dict end + r"#.*$|\s+", # comments and whitespaces + ] + tokens = re.compile("|".join(f"({pattern})" for pattern in patterns)) + lineno = 0 + multilines = [] # entries spread across multiple lines + for line in lines: + pos = 0 + + # deal with entries spread across multiple lines + # + # should we actually have to deal with escaped "s then do it here + if multilines: + multilines.append(line.strip()) + if line[-1] == '"': # closing multiline entry + # multiline entries will be joined by space. cannot + # reintroduce newlines as this will break the tokenizer + line = " ".join(multilines) + multilines = [] + else: # continued multiline entry + lineno += 1 + continue + else: + if line.count('"') == 1: # opening multiline entry + if line.strip()[0] != '"' and line.strip()[-1] != '"': + # since we expect something like key "value", the " should not be found at ends + # otherwise tokenizer will pick up the formatting mistake. + multilines = [line.rstrip()] + lineno += 1 + continue + + length = len(line) + + while pos < length: + match = tokens.match(line, pos) + if match is None: + m = f"cannot tokenize {line[pos:]} at ({lineno + 1}, {pos + 1})" + raise NetworkXError(m) + for i in range(len(patterns)): + group = match.group(i + 1) + if group is not None: + if i == 0: # keys + value = group.rstrip() + elif i == 1: # reals + value = float(group) + elif i == 2: # ints + value = int(group) + else: + value = group + if i != 6: # comments and whitespaces + yield Token(Pattern(i), value, lineno + 1, pos + 1) + pos += len(group) + break + lineno += 1 + yield Token(None, None, lineno + 1, 1) # EOF + + def unexpected(curr_token, expected): + category, value, lineno, pos = curr_token + value = repr(value) if value is not None else "EOF" + raise NetworkXError(f"expected {expected}, found {value} at ({lineno}, {pos})") + + def consume(curr_token, category, expected): + if curr_token.category == category: + return next(tokens) + unexpected(curr_token, expected) + + def parse_kv(curr_token): + dct = defaultdict(list) + while curr_token.category == Pattern.KEYS: + key = curr_token.value + curr_token = next(tokens) + category = curr_token.category + if category == Pattern.REALS or category == Pattern.INTS: + value = curr_token.value + curr_token = next(tokens) + elif category == Pattern.STRINGS: + value = unescape(curr_token.value[1:-1]) + if destringizer: + try: + value = destringizer(value) + except ValueError: + pass + # Special handling for empty lists and tuples + if value == "()": + value = () + if value == "[]": + value = [] + curr_token = next(tokens) + elif category == Pattern.DICT_START: + curr_token, value = parse_dict(curr_token) + else: + # Allow for string convertible id and label values + if key in ("id", "label", "source", "target"): + try: + # String convert the token value + value = unescape(str(curr_token.value)) + if destringizer: + try: + value = destringizer(value) + except ValueError: + pass + curr_token = next(tokens) + except Exception: + msg = ( + "an int, float, string, '[' or string" + + " convertible ASCII value for node id or label" + ) + unexpected(curr_token, msg) + # Special handling for nan and infinity. Since the gml language + # defines unquoted strings as keys, the numeric and string branches + # are skipped and we end up in this special branch, so we need to + # convert the current token value to a float for NAN and plain INF. + # +/-INF are handled in the pattern for 'reals' in tokenize(). This + # allows labels and values to be nan or infinity, but not keys. + elif curr_token.value in {"NAN", "INF"}: + value = float(curr_token.value) + curr_token = next(tokens) + else: # Otherwise error out + unexpected(curr_token, "an int, float, string or '['") + dct[key].append(value) + + def clean_dict_value(value): + if not isinstance(value, list): + return value + if len(value) == 1: + return value[0] + if value[0] == LIST_START_VALUE: + return value[1:] + return value + + dct = {key: clean_dict_value(value) for key, value in dct.items()} + return curr_token, dct + + def parse_dict(curr_token): + # dict start + curr_token = consume(curr_token, Pattern.DICT_START, "'['") + # dict contents + curr_token, dct = parse_kv(curr_token) + # dict end + curr_token = consume(curr_token, Pattern.DICT_END, "']'") + return curr_token, dct + + def parse_graph(): + curr_token, dct = parse_kv(next(tokens)) + if curr_token.category is not None: # EOF + unexpected(curr_token, "EOF") + if "graph" not in dct: + raise NetworkXError("input contains no graph") + graph = dct["graph"] + if isinstance(graph, list): + raise NetworkXError("input contains more than one graph") + return graph + + tokens = tokenize() + graph = parse_graph() + + directed = graph.pop("directed", False) + multigraph = graph.pop("multigraph", False) + if not multigraph: + G = nx.DiGraph() if directed else nx.Graph() + else: + G = nx.MultiDiGraph() if directed else nx.MultiGraph() + graph_attr = {k: v for k, v in graph.items() if k not in ("node", "edge")} + G.graph.update(graph_attr) + + def pop_attr(dct, category, attr, i): + try: + return dct.pop(attr) + except KeyError as err: + raise NetworkXError(f"{category} #{i} has no {attr!r} attribute") from err + + nodes = graph.get("node", []) + mapping = {} + node_labels = set() + for i, node in enumerate(nodes if isinstance(nodes, list) else [nodes]): + id = pop_attr(node, "node", "id", i) + if id in G: + raise NetworkXError(f"node id {id!r} is duplicated") + if label is not None and label != "id": + node_label = pop_attr(node, "node", label, i) + if node_label in node_labels: + raise NetworkXError(f"node label {node_label!r} is duplicated") + node_labels.add(node_label) + mapping[id] = node_label + G.add_node(id, **node) + + edges = graph.get("edge", []) + for i, edge in enumerate(edges if isinstance(edges, list) else [edges]): + source = pop_attr(edge, "edge", "source", i) + target = pop_attr(edge, "edge", "target", i) + if source not in G: + raise NetworkXError(f"edge #{i} has undefined source {source!r}") + if target not in G: + raise NetworkXError(f"edge #{i} has undefined target {target!r}") + if not multigraph: + if not G.has_edge(source, target): + G.add_edge(source, target, **edge) + else: + arrow = "->" if directed else "--" + msg = f"edge #{i} ({source!r}{arrow}{target!r}) is duplicated" + raise nx.NetworkXError(msg) + else: + key = edge.pop("key", None) + if key is not None and G.has_edge(source, target, key): + arrow = "->" if directed else "--" + msg = f"edge #{i} ({source!r}{arrow}{target!r}, {key!r})" + msg2 = 'Hint: If multigraph add "multigraph 1" to file header.' + raise nx.NetworkXError(msg + " is duplicated\n" + msg2) + G.add_edge(source, target, key, **edge) + + if label is not None and label != "id": + G = nx.relabel_nodes(G, mapping) + return G + + +def literal_stringizer(value): + """Convert a `value` to a Python literal in GML representation. + + Parameters + ---------- + value : object + The `value` to be converted to GML representation. + + Returns + ------- + rep : string + A double-quoted Python literal representing value. Unprintable + characters are replaced by XML character references. + + Raises + ------ + ValueError + If `value` cannot be converted to GML. + + Notes + ----- + The original value can be recovered using the + :func:`networkx.readwrite.gml.literal_destringizer` function. + """ + + def stringize(value): + if isinstance(value, int | bool) or value is None: + if value is True: # GML uses 1/0 for boolean values. + buf.write(str(1)) + elif value is False: + buf.write(str(0)) + else: + buf.write(str(value)) + elif isinstance(value, str): + text = repr(value) + if text[0] != "u": + try: + value.encode("latin1") + except UnicodeEncodeError: + text = "u" + text + buf.write(text) + elif isinstance(value, float | complex | str | bytes): + buf.write(repr(value)) + elif isinstance(value, list): + buf.write("[") + first = True + for item in value: + if not first: + buf.write(",") + else: + first = False + stringize(item) + buf.write("]") + elif isinstance(value, tuple): + if len(value) > 1: + buf.write("(") + first = True + for item in value: + if not first: + buf.write(",") + else: + first = False + stringize(item) + buf.write(")") + elif value: + buf.write("(") + stringize(value[0]) + buf.write(",)") + else: + buf.write("()") + elif isinstance(value, dict): + buf.write("{") + first = True + for key, value in value.items(): + if not first: + buf.write(",") + else: + first = False + stringize(key) + buf.write(":") + stringize(value) + buf.write("}") + elif isinstance(value, set): + buf.write("{") + first = True + for item in value: + if not first: + buf.write(",") + else: + first = False + stringize(item) + buf.write("}") + else: + msg = f"{value!r} cannot be converted into a Python literal" + raise ValueError(msg) + + buf = StringIO() + stringize(value) + return buf.getvalue() + + +def generate_gml(G, stringizer=None): + r"""Generate a single entry of the graph `G` in GML format. + + Parameters + ---------- + G : NetworkX graph + The graph to be converted to GML. + + stringizer : callable, optional + A `stringizer` which converts non-int/non-float/non-dict values into + strings. If it cannot convert a value into a string, it should raise a + `ValueError` to indicate that. Default value: None. + + Returns + ------- + lines: generator of strings + Lines of GML data. Newlines are not appended. + + Raises + ------ + NetworkXError + If `stringizer` cannot convert a value into a string, or the value to + convert is not a string while `stringizer` is None. + + See Also + -------- + literal_stringizer + + Notes + ----- + Graph attributes named 'directed', 'multigraph', 'node' or + 'edge', node attributes named 'id' or 'label', edge attributes + named 'source' or 'target' (or 'key' if `G` is a multigraph) + are ignored because these attribute names are used to encode the graph + structure. + + GML files are stored using a 7-bit ASCII encoding with any extended + ASCII characters (iso8859-1) appearing as HTML character entities. + Without specifying a `stringizer`/`destringizer`, the code is capable of + writing `int`/`float`/`str`/`dict`/`list` data as required by the GML + specification. For writing other data types, and for reading data other + than `str` you need to explicitly supply a `stringizer`/`destringizer`. + + For additional documentation on the GML file format, please see the + `GML url <https://web.archive.org/web/20190207140002/http://www.fim.uni-passau.de/index.php?id=17297&L=1>`_. + + See the module docstring :mod:`networkx.readwrite.gml` for more details. + + Examples + -------- + >>> G = nx.Graph() + >>> G.add_node("1") + >>> print("\n".join(nx.generate_gml(G))) + graph [ + node [ + id 0 + label "1" + ] + ] + >>> G = nx.MultiGraph([("a", "b"), ("a", "b")]) + >>> print("\n".join(nx.generate_gml(G))) + graph [ + multigraph 1 + node [ + id 0 + label "a" + ] + node [ + id 1 + label "b" + ] + edge [ + source 0 + target 1 + key 0 + ] + edge [ + source 0 + target 1 + key 1 + ] + ] + """ + valid_keys = re.compile("^[A-Za-z][0-9A-Za-z_]*$") + + def stringize(key, value, ignored_keys, indent, in_list=False): + if not isinstance(key, str): + raise NetworkXError(f"{key!r} is not a string") + if not valid_keys.match(key): + raise NetworkXError(f"{key!r} is not a valid key") + if not isinstance(key, str): + key = str(key) + if key not in ignored_keys: + if isinstance(value, int | bool): + if key == "label": + yield indent + key + ' "' + str(value) + '"' + elif value is True: + # python bool is an instance of int + yield indent + key + " 1" + elif value is False: + yield indent + key + " 0" + # GML only supports signed 32-bit integers + elif value < -(2**31) or value >= 2**31: + yield indent + key + ' "' + str(value) + '"' + else: + yield indent + key + " " + str(value) + elif isinstance(value, float): + text = repr(value).upper() + # GML matches INF to keys, so prepend + to INF. Use repr(float(*)) + # instead of string literal to future proof against changes to repr. + if text == repr(float("inf")).upper(): + text = "+" + text + else: + # GML requires that a real literal contain a decimal point, but + # repr may not output a decimal point when the mantissa is + # integral and hence needs fixing. + epos = text.rfind("E") + if epos != -1 and text.find(".", 0, epos) == -1: + text = text[:epos] + "." + text[epos:] + if key == "label": + yield indent + key + ' "' + text + '"' + else: + yield indent + key + " " + text + elif isinstance(value, dict): + yield indent + key + " [" + next_indent = indent + " " + for key, value in value.items(): + yield from stringize(key, value, (), next_indent) + yield indent + "]" + elif isinstance(value, tuple) and key == "label": + yield indent + key + f" \"({','.join(repr(v) for v in value)})\"" + elif isinstance(value, list | tuple) and key != "label" and not in_list: + if len(value) == 0: + yield indent + key + " " + f'"{value!r}"' + if len(value) == 1: + yield indent + key + " " + f'"{LIST_START_VALUE}"' + for val in value: + yield from stringize(key, val, (), indent, True) + else: + if stringizer: + try: + value = stringizer(value) + except ValueError as err: + raise NetworkXError( + f"{value!r} cannot be converted into a string" + ) from err + if not isinstance(value, str): + raise NetworkXError(f"{value!r} is not a string") + yield indent + key + ' "' + escape(value) + '"' + + multigraph = G.is_multigraph() + yield "graph [" + + # Output graph attributes + if G.is_directed(): + yield " directed 1" + if multigraph: + yield " multigraph 1" + ignored_keys = {"directed", "multigraph", "node", "edge"} + for attr, value in G.graph.items(): + yield from stringize(attr, value, ignored_keys, " ") + + # Output node data + node_id = dict(zip(G, range(len(G)))) + ignored_keys = {"id", "label"} + for node, attrs in G.nodes.items(): + yield " node [" + yield " id " + str(node_id[node]) + yield from stringize("label", node, (), " ") + for attr, value in attrs.items(): + yield from stringize(attr, value, ignored_keys, " ") + yield " ]" + + # Output edge data + ignored_keys = {"source", "target"} + kwargs = {"data": True} + if multigraph: + ignored_keys.add("key") + kwargs["keys"] = True + for e in G.edges(**kwargs): + yield " edge [" + yield " source " + str(node_id[e[0]]) + yield " target " + str(node_id[e[1]]) + if multigraph: + yield from stringize("key", e[2], (), " ") + for attr, value in e[-1].items(): + yield from stringize(attr, value, ignored_keys, " ") + yield " ]" + yield "]" + + +@open_file(1, mode="wb") +def write_gml(G, path, stringizer=None): + """Write a graph `G` in GML format to the file or file handle `path`. + + Parameters + ---------- + G : NetworkX graph + The graph to be converted to GML. + + path : filename or filehandle + The filename or filehandle to write. Files whose names end with .gz or + .bz2 will be compressed. + + stringizer : callable, optional + A `stringizer` which converts non-int/non-float/non-dict values into + strings. If it cannot convert a value into a string, it should raise a + `ValueError` to indicate that. Default value: None. + + Raises + ------ + NetworkXError + If `stringizer` cannot convert a value into a string, or the value to + convert is not a string while `stringizer` is None. + + See Also + -------- + read_gml, generate_gml + literal_stringizer + + Notes + ----- + Graph attributes named 'directed', 'multigraph', 'node' or + 'edge', node attributes named 'id' or 'label', edge attributes + named 'source' or 'target' (or 'key' if `G` is a multigraph) + are ignored because these attribute names are used to encode the graph + structure. + + GML files are stored using a 7-bit ASCII encoding with any extended + ASCII characters (iso8859-1) appearing as HTML character entities. + Without specifying a `stringizer`/`destringizer`, the code is capable of + writing `int`/`float`/`str`/`dict`/`list` data as required by the GML + specification. For writing other data types, and for reading data other + than `str` you need to explicitly supply a `stringizer`/`destringizer`. + + Note that while we allow non-standard GML to be read from a file, we make + sure to write GML format. In particular, underscores are not allowed in + attribute names. + For additional documentation on the GML file format, please see the + `GML url <https://web.archive.org/web/20190207140002/http://www.fim.uni-passau.de/index.php?id=17297&L=1>`_. + + See the module docstring :mod:`networkx.readwrite.gml` for more details. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_gml(G, "test.gml") + + Filenames ending in .gz or .bz2 will be compressed. + + >>> nx.write_gml(G, "test.gml.gz") + """ + for line in generate_gml(G, stringizer): + path.write((line + "\n").encode("ascii")) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/graph6.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/graph6.py new file mode 100644 index 00000000..4ff2f93c --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/graph6.py @@ -0,0 +1,417 @@ +# Original author: D. Eppstein, UC Irvine, August 12, 2003. +# The original code at http://www.ics.uci.edu/~eppstein/PADS/ is public domain. +"""Functions for reading and writing graphs in the *graph6* format. + +The *graph6* file format is suitable for small graphs or large dense +graphs. For large sparse graphs, use the *sparse6* format. + +For more information, see the `graph6`_ homepage. + +.. _graph6: http://users.cecs.anu.edu.au/~bdm/data/formats.html + +""" + +from itertools import islice + +import networkx as nx +from networkx.exception import NetworkXError +from networkx.utils import not_implemented_for, open_file + +__all__ = ["from_graph6_bytes", "read_graph6", "to_graph6_bytes", "write_graph6"] + + +def _generate_graph6_bytes(G, nodes, header): + """Yield bytes in the graph6 encoding of a graph. + + `G` is an undirected simple graph. `nodes` is the list of nodes for + which the node-induced subgraph will be encoded; if `nodes` is the + list of all nodes in the graph, the entire graph will be + encoded. `header` is a Boolean that specifies whether to generate + the header ``b'>>graph6<<'`` before the remaining data. + + This function generates `bytes` objects in the following order: + + 1. the header (if requested), + 2. the encoding of the number of nodes, + 3. each character, one-at-a-time, in the encoding of the requested + node-induced subgraph, + 4. a newline character. + + This function raises :exc:`ValueError` if the graph is too large for + the graph6 format (that is, greater than ``2 ** 36`` nodes). + + """ + n = len(G) + if n >= 2**36: + raise ValueError( + "graph6 is only defined if number of nodes is less than 2 ** 36" + ) + if header: + yield b">>graph6<<" + for d in n_to_data(n): + yield str.encode(chr(d + 63)) + # This generates the same as `(v in G[u] for u, v in combinations(G, 2))`, + # but in "column-major" order instead of "row-major" order. + bits = (nodes[j] in G[nodes[i]] for j in range(1, n) for i in range(j)) + chunk = list(islice(bits, 6)) + while chunk: + d = sum(b << 5 - i for i, b in enumerate(chunk)) + yield str.encode(chr(d + 63)) + chunk = list(islice(bits, 6)) + yield b"\n" + + +@nx._dispatchable(graphs=None, returns_graph=True) +def from_graph6_bytes(bytes_in): + """Read a simple undirected graph in graph6 format from bytes. + + Parameters + ---------- + bytes_in : bytes + Data in graph6 format, without a trailing newline. + + Returns + ------- + G : Graph + + Raises + ------ + NetworkXError + If bytes_in is unable to be parsed in graph6 format + + ValueError + If any character ``c`` in bytes_in does not satisfy + ``63 <= ord(c) < 127``. + + Examples + -------- + >>> G = nx.from_graph6_bytes(b"A_") + >>> sorted(G.edges()) + [(0, 1)] + + See Also + -------- + read_graph6, write_graph6 + + References + ---------- + .. [1] Graph6 specification + <http://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + + def bits(): + """Returns sequence of individual bits from 6-bit-per-value + list of data values.""" + for d in data: + for i in [5, 4, 3, 2, 1, 0]: + yield (d >> i) & 1 + + if bytes_in.startswith(b">>graph6<<"): + bytes_in = bytes_in[10:] + + data = [c - 63 for c in bytes_in] + if any(c > 63 for c in data): + raise ValueError("each input character must be in range(63, 127)") + + n, data = data_to_n(data) + nd = (n * (n - 1) // 2 + 5) // 6 + if len(data) != nd: + raise NetworkXError( + f"Expected {n * (n - 1) // 2} bits but got {len(data) * 6} in graph6" + ) + + G = nx.Graph() + G.add_nodes_from(range(n)) + for (i, j), b in zip(((i, j) for j in range(1, n) for i in range(j)), bits()): + if b: + G.add_edge(i, j) + + return G + + +@not_implemented_for("directed") +@not_implemented_for("multigraph") +def to_graph6_bytes(G, nodes=None, header=True): + """Convert a simple undirected graph to bytes in graph6 format. + + Parameters + ---------- + G : Graph (undirected) + + nodes: list or iterable + Nodes are labeled 0...n-1 in the order provided. If None the ordering + given by ``G.nodes()`` is used. + + header: bool + If True add '>>graph6<<' bytes to head of data. + + Raises + ------ + NetworkXNotImplemented + If the graph is directed or is a multigraph. + + ValueError + If the graph has at least ``2 ** 36`` nodes; the graph6 format + is only defined for graphs of order less than ``2 ** 36``. + + Examples + -------- + >>> nx.to_graph6_bytes(nx.path_graph(2)) + b'>>graph6<<A_\\n' + + See Also + -------- + from_graph6_bytes, read_graph6, write_graph6_bytes + + Notes + ----- + The returned bytes end with a newline character. + + The format does not support edge or node labels, parallel edges or + self loops. If self loops are present they are silently ignored. + + References + ---------- + .. [1] Graph6 specification + <http://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + if nodes is not None: + G = G.subgraph(nodes) + H = nx.convert_node_labels_to_integers(G) + nodes = sorted(H.nodes()) + return b"".join(_generate_graph6_bytes(H, nodes, header)) + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_graph6(path): + """Read simple undirected graphs in graph6 format from path. + + Parameters + ---------- + path : file or string + File or filename to write. + + Returns + ------- + G : Graph or list of Graphs + If the file contains multiple lines then a list of graphs is returned + + Raises + ------ + NetworkXError + If the string is unable to be parsed in graph6 format + + Examples + -------- + You can read a graph6 file by giving the path to the file:: + + >>> import tempfile + >>> with tempfile.NamedTemporaryFile(delete=False) as f: + ... _ = f.write(b">>graph6<<A_\\n") + ... _ = f.seek(0) + ... G = nx.read_graph6(f.name) + >>> list(G.edges()) + [(0, 1)] + + You can also read a graph6 file by giving an open file-like object:: + + >>> import tempfile + >>> with tempfile.NamedTemporaryFile() as f: + ... _ = f.write(b">>graph6<<A_\\n") + ... _ = f.seek(0) + ... G = nx.read_graph6(f) + >>> list(G.edges()) + [(0, 1)] + + See Also + -------- + from_graph6_bytes, write_graph6 + + References + ---------- + .. [1] Graph6 specification + <http://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + glist = [] + for line in path: + line = line.strip() + if not len(line): + continue + glist.append(from_graph6_bytes(line)) + if len(glist) == 1: + return glist[0] + else: + return glist + + +@not_implemented_for("directed") +@not_implemented_for("multigraph") +@open_file(1, mode="wb") +def write_graph6(G, path, nodes=None, header=True): + """Write a simple undirected graph to a path in graph6 format. + + Parameters + ---------- + G : Graph (undirected) + + path : str + The path naming the file to which to write the graph. + + nodes: list or iterable + Nodes are labeled 0...n-1 in the order provided. If None the ordering + given by ``G.nodes()`` is used. + + header: bool + If True add '>>graph6<<' string to head of data + + Raises + ------ + NetworkXNotImplemented + If the graph is directed or is a multigraph. + + ValueError + If the graph has at least ``2 ** 36`` nodes; the graph6 format + is only defined for graphs of order less than ``2 ** 36``. + + Examples + -------- + You can write a graph6 file by giving the path to a file:: + + >>> import tempfile + >>> with tempfile.NamedTemporaryFile(delete=False) as f: + ... nx.write_graph6(nx.path_graph(2), f.name) + ... _ = f.seek(0) + ... print(f.read()) + b'>>graph6<<A_\\n' + + See Also + -------- + from_graph6_bytes, read_graph6 + + Notes + ----- + The function writes a newline character after writing the encoding + of the graph. + + The format does not support edge or node labels, parallel edges or + self loops. If self loops are present they are silently ignored. + + References + ---------- + .. [1] Graph6 specification + <http://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + return write_graph6_file(G, path, nodes=nodes, header=header) + + +@not_implemented_for("directed") +@not_implemented_for("multigraph") +def write_graph6_file(G, f, nodes=None, header=True): + """Write a simple undirected graph to a file-like object in graph6 format. + + Parameters + ---------- + G : Graph (undirected) + + f : file-like object + The file to write. + + nodes: list or iterable + Nodes are labeled 0...n-1 in the order provided. If None the ordering + given by ``G.nodes()`` is used. + + header: bool + If True add '>>graph6<<' string to head of data + + Raises + ------ + NetworkXNotImplemented + If the graph is directed or is a multigraph. + + ValueError + If the graph has at least ``2 ** 36`` nodes; the graph6 format + is only defined for graphs of order less than ``2 ** 36``. + + Examples + -------- + You can write a graph6 file by giving an open file-like object:: + + >>> import tempfile + >>> with tempfile.NamedTemporaryFile() as f: + ... nx.write_graph6(nx.path_graph(2), f) + ... _ = f.seek(0) + ... print(f.read()) + b'>>graph6<<A_\\n' + + See Also + -------- + from_graph6_bytes, read_graph6 + + Notes + ----- + The function writes a newline character after writing the encoding + of the graph. + + The format does not support edge or node labels, parallel edges or + self loops. If self loops are present they are silently ignored. + + References + ---------- + .. [1] Graph6 specification + <http://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + if nodes is not None: + G = G.subgraph(nodes) + H = nx.convert_node_labels_to_integers(G) + nodes = sorted(H.nodes()) + for b in _generate_graph6_bytes(H, nodes, header): + f.write(b) + + +def data_to_n(data): + """Read initial one-, four- or eight-unit value from graph6 + integer sequence. + + Return (value, rest of seq.)""" + if data[0] <= 62: + return data[0], data[1:] + if data[1] <= 62: + return (data[1] << 12) + (data[2] << 6) + data[3], data[4:] + return ( + (data[2] << 30) + + (data[3] << 24) + + (data[4] << 18) + + (data[5] << 12) + + (data[6] << 6) + + data[7], + data[8:], + ) + + +def n_to_data(n): + """Convert an integer to one-, four- or eight-unit graph6 sequence. + + This function is undefined if `n` is not in ``range(2 ** 36)``. + + """ + if n <= 62: + return [n] + elif n <= 258047: + return [63, (n >> 12) & 0x3F, (n >> 6) & 0x3F, n & 0x3F] + else: # if n <= 68719476735: + return [ + 63, + 63, + (n >> 30) & 0x3F, + (n >> 24) & 0x3F, + (n >> 18) & 0x3F, + (n >> 12) & 0x3F, + (n >> 6) & 0x3F, + n & 0x3F, + ] diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/graphml.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/graphml.py new file mode 100644 index 00000000..7d0a1da0 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/graphml.py @@ -0,0 +1,1053 @@ +""" +******* +GraphML +******* +Read and write graphs in GraphML format. + +.. warning:: + + This parser uses the standard xml library present in Python, which is + insecure - see :external+python:mod:`xml` for additional information. + Only parse GraphML files you trust. + +This implementation does not support mixed graphs (directed and unidirected +edges together), hyperedges, nested graphs, or ports. + +"GraphML is a comprehensive and easy-to-use file format for graphs. It +consists of a language core to describe the structural properties of a +graph and a flexible extension mechanism to add application-specific +data. Its main features include support of + + * directed, undirected, and mixed graphs, + * hypergraphs, + * hierarchical graphs, + * graphical representations, + * references to external data, + * application-specific attribute data, and + * light-weight parsers. + +Unlike many other file formats for graphs, GraphML does not use a +custom syntax. Instead, it is based on XML and hence ideally suited as +a common denominator for all kinds of services generating, archiving, +or processing graphs." + +http://graphml.graphdrawing.org/ + +Format +------ +GraphML is an XML format. See +http://graphml.graphdrawing.org/specification.html for the specification and +http://graphml.graphdrawing.org/primer/graphml-primer.html +for examples. +""" + +import warnings +from collections import defaultdict + +import networkx as nx +from networkx.utils import open_file + +__all__ = [ + "write_graphml", + "read_graphml", + "generate_graphml", + "write_graphml_xml", + "write_graphml_lxml", + "parse_graphml", + "GraphMLWriter", + "GraphMLReader", +] + + +@open_file(1, mode="wb") +def write_graphml_xml( + G, + path, + encoding="utf-8", + prettyprint=True, + infer_numeric_types=False, + named_key_ids=False, + edge_id_from_attribute=None, +): + """Write G in GraphML XML format to path + + Parameters + ---------- + G : graph + A networkx graph + path : file or string + File or filename to write. + Filenames ending in .gz or .bz2 will be compressed. + encoding : string (optional) + Encoding for text data. + prettyprint : bool (optional) + If True use line breaks and indenting in output XML. + infer_numeric_types : boolean + Determine if numeric types should be generalized. + For example, if edges have both int and float 'weight' attributes, + we infer in GraphML that both are floats. + named_key_ids : bool (optional) + If True use attr.name as value for key elements' id attribute. + edge_id_from_attribute : dict key (optional) + If provided, the graphml edge id is set by looking up the corresponding + edge data attribute keyed by this parameter. If `None` or the key does not exist in edge data, + the edge id is set by the edge key if `G` is a MultiGraph, else the edge id is left unset. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_graphml(G, "test.graphml") + + Notes + ----- + This implementation does not support mixed graphs (directed + and unidirected edges together) hyperedges, nested graphs, or ports. + """ + writer = GraphMLWriter( + encoding=encoding, + prettyprint=prettyprint, + infer_numeric_types=infer_numeric_types, + named_key_ids=named_key_ids, + edge_id_from_attribute=edge_id_from_attribute, + ) + writer.add_graph_element(G) + writer.dump(path) + + +@open_file(1, mode="wb") +def write_graphml_lxml( + G, + path, + encoding="utf-8", + prettyprint=True, + infer_numeric_types=False, + named_key_ids=False, + edge_id_from_attribute=None, +): + """Write G in GraphML XML format to path + + This function uses the LXML framework and should be faster than + the version using the xml library. + + Parameters + ---------- + G : graph + A networkx graph + path : file or string + File or filename to write. + Filenames ending in .gz or .bz2 will be compressed. + encoding : string (optional) + Encoding for text data. + prettyprint : bool (optional) + If True use line breaks and indenting in output XML. + infer_numeric_types : boolean + Determine if numeric types should be generalized. + For example, if edges have both int and float 'weight' attributes, + we infer in GraphML that both are floats. + named_key_ids : bool (optional) + If True use attr.name as value for key elements' id attribute. + edge_id_from_attribute : dict key (optional) + If provided, the graphml edge id is set by looking up the corresponding + edge data attribute keyed by this parameter. If `None` or the key does not exist in edge data, + the edge id is set by the edge key if `G` is a MultiGraph, else the edge id is left unset. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_graphml_lxml(G, "fourpath.graphml") + + Notes + ----- + This implementation does not support mixed graphs (directed + and unidirected edges together) hyperedges, nested graphs, or ports. + """ + try: + import lxml.etree as lxmletree + except ImportError: + return write_graphml_xml( + G, + path, + encoding, + prettyprint, + infer_numeric_types, + named_key_ids, + edge_id_from_attribute, + ) + + writer = GraphMLWriterLxml( + path, + graph=G, + encoding=encoding, + prettyprint=prettyprint, + infer_numeric_types=infer_numeric_types, + named_key_ids=named_key_ids, + edge_id_from_attribute=edge_id_from_attribute, + ) + writer.dump() + + +def generate_graphml( + G, + encoding="utf-8", + prettyprint=True, + named_key_ids=False, + edge_id_from_attribute=None, +): + """Generate GraphML lines for G + + Parameters + ---------- + G : graph + A networkx graph + encoding : string (optional) + Encoding for text data. + prettyprint : bool (optional) + If True use line breaks and indenting in output XML. + named_key_ids : bool (optional) + If True use attr.name as value for key elements' id attribute. + edge_id_from_attribute : dict key (optional) + If provided, the graphml edge id is set by looking up the corresponding + edge data attribute keyed by this parameter. If `None` or the key does not exist in edge data, + the edge id is set by the edge key if `G` is a MultiGraph, else the edge id is left unset. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> linefeed = chr(10) # linefeed = \n + >>> s = linefeed.join(nx.generate_graphml(G)) + >>> for line in nx.generate_graphml(G): # doctest: +SKIP + ... print(line) + + Notes + ----- + This implementation does not support mixed graphs (directed and unidirected + edges together) hyperedges, nested graphs, or ports. + """ + writer = GraphMLWriter( + encoding=encoding, + prettyprint=prettyprint, + named_key_ids=named_key_ids, + edge_id_from_attribute=edge_id_from_attribute, + ) + writer.add_graph_element(G) + yield from str(writer).splitlines() + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_graphml(path, node_type=str, edge_key_type=int, force_multigraph=False): + """Read graph in GraphML format from path. + + Parameters + ---------- + path : file or string + File or filename to write. + Filenames ending in .gz or .bz2 will be compressed. + + node_type: Python type (default: str) + Convert node ids to this type + + edge_key_type: Python type (default: int) + Convert graphml edge ids to this type. Multigraphs use id as edge key. + Non-multigraphs add to edge attribute dict with name "id". + + force_multigraph : bool (default: False) + If True, return a multigraph with edge keys. If False (the default) + return a multigraph when multiedges are in the graph. + + Returns + ------- + graph: NetworkX graph + If parallel edges are present or `force_multigraph=True` then + a MultiGraph or MultiDiGraph is returned. Otherwise a Graph/DiGraph. + The returned graph is directed if the file indicates it should be. + + Notes + ----- + Default node and edge attributes are not propagated to each node and edge. + They can be obtained from `G.graph` and applied to node and edge attributes + if desired using something like this: + + >>> default_color = G.graph["node_default"]["color"] # doctest: +SKIP + >>> for node, data in G.nodes(data=True): # doctest: +SKIP + ... if "color" not in data: + ... data["color"] = default_color + >>> default_color = G.graph["edge_default"]["color"] # doctest: +SKIP + >>> for u, v, data in G.edges(data=True): # doctest: +SKIP + ... if "color" not in data: + ... data["color"] = default_color + + This implementation does not support mixed graphs (directed and unidirected + edges together), hypergraphs, nested graphs, or ports. + + For multigraphs the GraphML edge "id" will be used as the edge + key. If not specified then they "key" attribute will be used. If + there is no "key" attribute a default NetworkX multigraph edge key + will be provided. + + Files with the yEd "yfiles" extension can be read. The type of the node's + shape is preserved in the `shape_type` node attribute. + + yEd compressed files ("file.graphmlz" extension) can be read by renaming + the file to "file.graphml.gz". + + """ + reader = GraphMLReader(node_type, edge_key_type, force_multigraph) + # need to check for multiple graphs + glist = list(reader(path=path)) + if len(glist) == 0: + # If no graph comes back, try looking for an incomplete header + header = b'<graphml xmlns="http://graphml.graphdrawing.org/xmlns">' + path.seek(0) + old_bytes = path.read() + new_bytes = old_bytes.replace(b"<graphml>", header) + glist = list(reader(string=new_bytes)) + if len(glist) == 0: + raise nx.NetworkXError("file not successfully read as graphml") + return glist[0] + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_graphml( + graphml_string, node_type=str, edge_key_type=int, force_multigraph=False +): + """Read graph in GraphML format from string. + + Parameters + ---------- + graphml_string : string + String containing graphml information + (e.g., contents of a graphml file). + + node_type: Python type (default: str) + Convert node ids to this type + + edge_key_type: Python type (default: int) + Convert graphml edge ids to this type. Multigraphs use id as edge key. + Non-multigraphs add to edge attribute dict with name "id". + + force_multigraph : bool (default: False) + If True, return a multigraph with edge keys. If False (the default) + return a multigraph when multiedges are in the graph. + + + Returns + ------- + graph: NetworkX graph + If no parallel edges are found a Graph or DiGraph is returned. + Otherwise a MultiGraph or MultiDiGraph is returned. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> linefeed = chr(10) # linefeed = \n + >>> s = linefeed.join(nx.generate_graphml(G)) + >>> H = nx.parse_graphml(s) + + Notes + ----- + Default node and edge attributes are not propagated to each node and edge. + They can be obtained from `G.graph` and applied to node and edge attributes + if desired using something like this: + + >>> default_color = G.graph["node_default"]["color"] # doctest: +SKIP + >>> for node, data in G.nodes(data=True): # doctest: +SKIP + ... if "color" not in data: + ... data["color"] = default_color + >>> default_color = G.graph["edge_default"]["color"] # doctest: +SKIP + >>> for u, v, data in G.edges(data=True): # doctest: +SKIP + ... if "color" not in data: + ... data["color"] = default_color + + This implementation does not support mixed graphs (directed and unidirected + edges together), hypergraphs, nested graphs, or ports. + + For multigraphs the GraphML edge "id" will be used as the edge + key. If not specified then they "key" attribute will be used. If + there is no "key" attribute a default NetworkX multigraph edge key + will be provided. + + """ + reader = GraphMLReader(node_type, edge_key_type, force_multigraph) + # need to check for multiple graphs + glist = list(reader(string=graphml_string)) + if len(glist) == 0: + # If no graph comes back, try looking for an incomplete header + header = '<graphml xmlns="http://graphml.graphdrawing.org/xmlns">' + new_string = graphml_string.replace("<graphml>", header) + glist = list(reader(string=new_string)) + if len(glist) == 0: + raise nx.NetworkXError("file not successfully read as graphml") + return glist[0] + + +class GraphML: + NS_GRAPHML = "http://graphml.graphdrawing.org/xmlns" + NS_XSI = "http://www.w3.org/2001/XMLSchema-instance" + # xmlns:y="http://www.yworks.com/xml/graphml" + NS_Y = "http://www.yworks.com/xml/graphml" + SCHEMALOCATION = " ".join( + [ + "http://graphml.graphdrawing.org/xmlns", + "http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd", + ] + ) + + def construct_types(self): + types = [ + (int, "integer"), # for Gephi GraphML bug + (str, "yfiles"), + (str, "string"), + (int, "int"), + (int, "long"), + (float, "float"), + (float, "double"), + (bool, "boolean"), + ] + + # These additions to types allow writing numpy types + try: + import numpy as np + except: + pass + else: + # prepend so that python types are created upon read (last entry wins) + types = [ + (np.float64, "float"), + (np.float32, "float"), + (np.float16, "float"), + (np.int_, "int"), + (np.int8, "int"), + (np.int16, "int"), + (np.int32, "int"), + (np.int64, "int"), + (np.uint8, "int"), + (np.uint16, "int"), + (np.uint32, "int"), + (np.uint64, "int"), + (np.int_, "int"), + (np.intc, "int"), + (np.intp, "int"), + ] + types + + self.xml_type = dict(types) + self.python_type = dict(reversed(a) for a in types) + + # This page says that data types in GraphML follow Java(TM). + # http://graphml.graphdrawing.org/primer/graphml-primer.html#AttributesDefinition + # true and false are the only boolean literals: + # http://en.wikibooks.org/wiki/Java_Programming/Literals#Boolean_Literals + convert_bool = { + # We use data.lower() in actual use. + "true": True, + "false": False, + # Include integer strings for convenience. + "0": False, + 0: False, + "1": True, + 1: True, + } + + def get_xml_type(self, key): + """Wrapper around the xml_type dict that raises a more informative + exception message when a user attempts to use data of a type not + supported by GraphML.""" + try: + return self.xml_type[key] + except KeyError as err: + raise TypeError( + f"GraphML does not support type {key} as data values." + ) from err + + +class GraphMLWriter(GraphML): + def __init__( + self, + graph=None, + encoding="utf-8", + prettyprint=True, + infer_numeric_types=False, + named_key_ids=False, + edge_id_from_attribute=None, + ): + self.construct_types() + from xml.etree.ElementTree import Element + + self.myElement = Element + + self.infer_numeric_types = infer_numeric_types + self.prettyprint = prettyprint + self.named_key_ids = named_key_ids + self.edge_id_from_attribute = edge_id_from_attribute + self.encoding = encoding + self.xml = self.myElement( + "graphml", + { + "xmlns": self.NS_GRAPHML, + "xmlns:xsi": self.NS_XSI, + "xsi:schemaLocation": self.SCHEMALOCATION, + }, + ) + self.keys = {} + self.attributes = defaultdict(list) + self.attribute_types = defaultdict(set) + + if graph is not None: + self.add_graph_element(graph) + + def __str__(self): + from xml.etree.ElementTree import tostring + + if self.prettyprint: + self.indent(self.xml) + s = tostring(self.xml).decode(self.encoding) + return s + + def attr_type(self, name, scope, value): + """Infer the attribute type of data named name. Currently this only + supports inference of numeric types. + + If self.infer_numeric_types is false, type is used. Otherwise, pick the + most general of types found across all values with name and scope. This + means edges with data named 'weight' are treated separately from nodes + with data named 'weight'. + """ + if self.infer_numeric_types: + types = self.attribute_types[(name, scope)] + + if len(types) > 1: + types = {self.get_xml_type(t) for t in types} + if "string" in types: + return str + elif "float" in types or "double" in types: + return float + else: + return int + else: + return list(types)[0] + else: + return type(value) + + def get_key(self, name, attr_type, scope, default): + keys_key = (name, attr_type, scope) + try: + return self.keys[keys_key] + except KeyError: + if self.named_key_ids: + new_id = name + else: + new_id = f"d{len(list(self.keys))}" + + self.keys[keys_key] = new_id + key_kwargs = { + "id": new_id, + "for": scope, + "attr.name": name, + "attr.type": attr_type, + } + key_element = self.myElement("key", **key_kwargs) + # add subelement for data default value if present + if default is not None: + default_element = self.myElement("default") + default_element.text = str(default) + key_element.append(default_element) + self.xml.insert(0, key_element) + return new_id + + def add_data(self, name, element_type, value, scope="all", default=None): + """ + Make a data element for an edge or a node. Keep a log of the + type in the keys table. + """ + if element_type not in self.xml_type: + raise nx.NetworkXError( + f"GraphML writer does not support {element_type} as data values." + ) + keyid = self.get_key(name, self.get_xml_type(element_type), scope, default) + data_element = self.myElement("data", key=keyid) + data_element.text = str(value) + return data_element + + def add_attributes(self, scope, xml_obj, data, default): + """Appends attribute data to edges or nodes, and stores type information + to be added later. See add_graph_element. + """ + for k, v in data.items(): + self.attribute_types[(str(k), scope)].add(type(v)) + self.attributes[xml_obj].append([k, v, scope, default.get(k)]) + + def add_nodes(self, G, graph_element): + default = G.graph.get("node_default", {}) + for node, data in G.nodes(data=True): + node_element = self.myElement("node", id=str(node)) + self.add_attributes("node", node_element, data, default) + graph_element.append(node_element) + + def add_edges(self, G, graph_element): + if G.is_multigraph(): + for u, v, key, data in G.edges(data=True, keys=True): + edge_element = self.myElement( + "edge", + source=str(u), + target=str(v), + id=str(data.get(self.edge_id_from_attribute)) + if self.edge_id_from_attribute + and self.edge_id_from_attribute in data + else str(key), + ) + default = G.graph.get("edge_default", {}) + self.add_attributes("edge", edge_element, data, default) + graph_element.append(edge_element) + else: + for u, v, data in G.edges(data=True): + if self.edge_id_from_attribute and self.edge_id_from_attribute in data: + # select attribute to be edge id + edge_element = self.myElement( + "edge", + source=str(u), + target=str(v), + id=str(data.get(self.edge_id_from_attribute)), + ) + else: + # default: no edge id + edge_element = self.myElement("edge", source=str(u), target=str(v)) + default = G.graph.get("edge_default", {}) + self.add_attributes("edge", edge_element, data, default) + graph_element.append(edge_element) + + def add_graph_element(self, G): + """ + Serialize graph G in GraphML to the stream. + """ + if G.is_directed(): + default_edge_type = "directed" + else: + default_edge_type = "undirected" + + graphid = G.graph.pop("id", None) + if graphid is None: + graph_element = self.myElement("graph", edgedefault=default_edge_type) + else: + graph_element = self.myElement( + "graph", edgedefault=default_edge_type, id=graphid + ) + default = {} + data = { + k: v + for (k, v) in G.graph.items() + if k not in ["node_default", "edge_default"] + } + self.add_attributes("graph", graph_element, data, default) + self.add_nodes(G, graph_element) + self.add_edges(G, graph_element) + + # self.attributes contains a mapping from XML Objects to a list of + # data that needs to be added to them. + # We postpone processing in order to do type inference/generalization. + # See self.attr_type + for xml_obj, data in self.attributes.items(): + for k, v, scope, default in data: + xml_obj.append( + self.add_data( + str(k), self.attr_type(k, scope, v), str(v), scope, default + ) + ) + self.xml.append(graph_element) + + def add_graphs(self, graph_list): + """Add many graphs to this GraphML document.""" + for G in graph_list: + self.add_graph_element(G) + + def dump(self, stream): + from xml.etree.ElementTree import ElementTree + + if self.prettyprint: + self.indent(self.xml) + document = ElementTree(self.xml) + document.write(stream, encoding=self.encoding, xml_declaration=True) + + def indent(self, elem, level=0): + # in-place prettyprint formatter + i = "\n" + level * " " + if len(elem): + if not elem.text or not elem.text.strip(): + elem.text = i + " " + if not elem.tail or not elem.tail.strip(): + elem.tail = i + for elem in elem: + self.indent(elem, level + 1) + if not elem.tail or not elem.tail.strip(): + elem.tail = i + else: + if level and (not elem.tail or not elem.tail.strip()): + elem.tail = i + + +class IncrementalElement: + """Wrapper for _IncrementalWriter providing an Element like interface. + + This wrapper does not intend to be a complete implementation but rather to + deal with those calls used in GraphMLWriter. + """ + + def __init__(self, xml, prettyprint): + self.xml = xml + self.prettyprint = prettyprint + + def append(self, element): + self.xml.write(element, pretty_print=self.prettyprint) + + +class GraphMLWriterLxml(GraphMLWriter): + def __init__( + self, + path, + graph=None, + encoding="utf-8", + prettyprint=True, + infer_numeric_types=False, + named_key_ids=False, + edge_id_from_attribute=None, + ): + self.construct_types() + import lxml.etree as lxmletree + + self.myElement = lxmletree.Element + + self._encoding = encoding + self._prettyprint = prettyprint + self.named_key_ids = named_key_ids + self.edge_id_from_attribute = edge_id_from_attribute + self.infer_numeric_types = infer_numeric_types + + self._xml_base = lxmletree.xmlfile(path, encoding=encoding) + self._xml = self._xml_base.__enter__() + self._xml.write_declaration() + + # We need to have a xml variable that support insertion. This call is + # used for adding the keys to the document. + # We will store those keys in a plain list, and then after the graph + # element is closed we will add them to the main graphml element. + self.xml = [] + self._keys = self.xml + self._graphml = self._xml.element( + "graphml", + { + "xmlns": self.NS_GRAPHML, + "xmlns:xsi": self.NS_XSI, + "xsi:schemaLocation": self.SCHEMALOCATION, + }, + ) + self._graphml.__enter__() + self.keys = {} + self.attribute_types = defaultdict(set) + + if graph is not None: + self.add_graph_element(graph) + + def add_graph_element(self, G): + """ + Serialize graph G in GraphML to the stream. + """ + if G.is_directed(): + default_edge_type = "directed" + else: + default_edge_type = "undirected" + + graphid = G.graph.pop("id", None) + if graphid is None: + graph_element = self._xml.element("graph", edgedefault=default_edge_type) + else: + graph_element = self._xml.element( + "graph", edgedefault=default_edge_type, id=graphid + ) + + # gather attributes types for the whole graph + # to find the most general numeric format needed. + # Then pass through attributes to create key_id for each. + graphdata = { + k: v + for k, v in G.graph.items() + if k not in ("node_default", "edge_default") + } + node_default = G.graph.get("node_default", {}) + edge_default = G.graph.get("edge_default", {}) + # Graph attributes + for k, v in graphdata.items(): + self.attribute_types[(str(k), "graph")].add(type(v)) + for k, v in graphdata.items(): + element_type = self.get_xml_type(self.attr_type(k, "graph", v)) + self.get_key(str(k), element_type, "graph", None) + # Nodes and data + for node, d in G.nodes(data=True): + for k, v in d.items(): + self.attribute_types[(str(k), "node")].add(type(v)) + for node, d in G.nodes(data=True): + for k, v in d.items(): + T = self.get_xml_type(self.attr_type(k, "node", v)) + self.get_key(str(k), T, "node", node_default.get(k)) + # Edges and data + if G.is_multigraph(): + for u, v, ekey, d in G.edges(keys=True, data=True): + for k, v in d.items(): + self.attribute_types[(str(k), "edge")].add(type(v)) + for u, v, ekey, d in G.edges(keys=True, data=True): + for k, v in d.items(): + T = self.get_xml_type(self.attr_type(k, "edge", v)) + self.get_key(str(k), T, "edge", edge_default.get(k)) + else: + for u, v, d in G.edges(data=True): + for k, v in d.items(): + self.attribute_types[(str(k), "edge")].add(type(v)) + for u, v, d in G.edges(data=True): + for k, v in d.items(): + T = self.get_xml_type(self.attr_type(k, "edge", v)) + self.get_key(str(k), T, "edge", edge_default.get(k)) + + # Now add attribute keys to the xml file + for key in self.xml: + self._xml.write(key, pretty_print=self._prettyprint) + + # The incremental_writer writes each node/edge as it is created + incremental_writer = IncrementalElement(self._xml, self._prettyprint) + with graph_element: + self.add_attributes("graph", incremental_writer, graphdata, {}) + self.add_nodes(G, incremental_writer) # adds attributes too + self.add_edges(G, incremental_writer) # adds attributes too + + def add_attributes(self, scope, xml_obj, data, default): + """Appends attribute data.""" + for k, v in data.items(): + data_element = self.add_data( + str(k), self.attr_type(str(k), scope, v), str(v), scope, default.get(k) + ) + xml_obj.append(data_element) + + def __str__(self): + return object.__str__(self) + + def dump(self, stream=None): + self._graphml.__exit__(None, None, None) + self._xml_base.__exit__(None, None, None) + + +# default is lxml is present. +write_graphml = write_graphml_lxml + + +class GraphMLReader(GraphML): + """Read a GraphML document. Produces NetworkX graph objects.""" + + def __init__(self, node_type=str, edge_key_type=int, force_multigraph=False): + self.construct_types() + self.node_type = node_type + self.edge_key_type = edge_key_type + self.multigraph = force_multigraph # If False, test for multiedges + self.edge_ids = {} # dict mapping (u,v) tuples to edge id attributes + + def __call__(self, path=None, string=None): + from xml.etree.ElementTree import ElementTree, fromstring + + if path is not None: + self.xml = ElementTree(file=path) + elif string is not None: + self.xml = fromstring(string) + else: + raise ValueError("Must specify either 'path' or 'string' as kwarg") + (keys, defaults) = self.find_graphml_keys(self.xml) + for g in self.xml.findall(f"{{{self.NS_GRAPHML}}}graph"): + yield self.make_graph(g, keys, defaults) + + def make_graph(self, graph_xml, graphml_keys, defaults, G=None): + # set default graph type + edgedefault = graph_xml.get("edgedefault", None) + if G is None: + if edgedefault == "directed": + G = nx.MultiDiGraph() + else: + G = nx.MultiGraph() + # set defaults for graph attributes + G.graph["node_default"] = {} + G.graph["edge_default"] = {} + for key_id, value in defaults.items(): + key_for = graphml_keys[key_id]["for"] + name = graphml_keys[key_id]["name"] + python_type = graphml_keys[key_id]["type"] + if key_for == "node": + G.graph["node_default"].update({name: python_type(value)}) + if key_for == "edge": + G.graph["edge_default"].update({name: python_type(value)}) + # hyperedges are not supported + hyperedge = graph_xml.find(f"{{{self.NS_GRAPHML}}}hyperedge") + if hyperedge is not None: + raise nx.NetworkXError("GraphML reader doesn't support hyperedges") + # add nodes + for node_xml in graph_xml.findall(f"{{{self.NS_GRAPHML}}}node"): + self.add_node(G, node_xml, graphml_keys, defaults) + # add edges + for edge_xml in graph_xml.findall(f"{{{self.NS_GRAPHML}}}edge"): + self.add_edge(G, edge_xml, graphml_keys) + # add graph data + data = self.decode_data_elements(graphml_keys, graph_xml) + G.graph.update(data) + + # switch to Graph or DiGraph if no parallel edges were found + if self.multigraph: + return G + + G = nx.DiGraph(G) if G.is_directed() else nx.Graph(G) + # add explicit edge "id" from file as attribute in NX graph. + nx.set_edge_attributes(G, values=self.edge_ids, name="id") + return G + + def add_node(self, G, node_xml, graphml_keys, defaults): + """Add a node to the graph.""" + # warn on finding unsupported ports tag + ports = node_xml.find(f"{{{self.NS_GRAPHML}}}port") + if ports is not None: + warnings.warn("GraphML port tag not supported.") + # find the node by id and cast it to the appropriate type + node_id = self.node_type(node_xml.get("id")) + # get data/attributes for node + data = self.decode_data_elements(graphml_keys, node_xml) + G.add_node(node_id, **data) + # get child nodes + if node_xml.attrib.get("yfiles.foldertype") == "group": + graph_xml = node_xml.find(f"{{{self.NS_GRAPHML}}}graph") + self.make_graph(graph_xml, graphml_keys, defaults, G) + + def add_edge(self, G, edge_element, graphml_keys): + """Add an edge to the graph.""" + # warn on finding unsupported ports tag + ports = edge_element.find(f"{{{self.NS_GRAPHML}}}port") + if ports is not None: + warnings.warn("GraphML port tag not supported.") + + # raise error if we find mixed directed and undirected edges + directed = edge_element.get("directed") + if G.is_directed() and directed == "false": + msg = "directed=false edge found in directed graph." + raise nx.NetworkXError(msg) + if (not G.is_directed()) and directed == "true": + msg = "directed=true edge found in undirected graph." + raise nx.NetworkXError(msg) + + source = self.node_type(edge_element.get("source")) + target = self.node_type(edge_element.get("target")) + data = self.decode_data_elements(graphml_keys, edge_element) + # GraphML stores edge ids as an attribute + # NetworkX uses them as keys in multigraphs too if no key + # attribute is specified + edge_id = edge_element.get("id") + if edge_id: + # self.edge_ids is used by `make_graph` method for non-multigraphs + self.edge_ids[source, target] = edge_id + try: + edge_id = self.edge_key_type(edge_id) + except ValueError: # Could not convert. + pass + else: + edge_id = data.get("key") + + if G.has_edge(source, target): + # mark this as a multigraph + self.multigraph = True + + # Use add_edges_from to avoid error with add_edge when `'key' in data` + # Note there is only one edge here... + G.add_edges_from([(source, target, edge_id, data)]) + + def decode_data_elements(self, graphml_keys, obj_xml): + """Use the key information to decode the data XML if present.""" + data = {} + for data_element in obj_xml.findall(f"{{{self.NS_GRAPHML}}}data"): + key = data_element.get("key") + try: + data_name = graphml_keys[key]["name"] + data_type = graphml_keys[key]["type"] + except KeyError as err: + raise nx.NetworkXError(f"Bad GraphML data: no key {key}") from err + text = data_element.text + # assume anything with subelements is a yfiles extension + if text is not None and len(list(data_element)) == 0: + if data_type == bool: + # Ignore cases. + # http://docs.oracle.com/javase/6/docs/api/java/lang/ + # Boolean.html#parseBoolean%28java.lang.String%29 + data[data_name] = self.convert_bool[text.lower()] + else: + data[data_name] = data_type(text) + elif len(list(data_element)) > 0: + # Assume yfiles as subelements, try to extract node_label + node_label = None + # set GenericNode's configuration as shape type + gn = data_element.find(f"{{{self.NS_Y}}}GenericNode") + if gn is not None: + data["shape_type"] = gn.get("configuration") + for node_type in ["GenericNode", "ShapeNode", "SVGNode", "ImageNode"]: + pref = f"{{{self.NS_Y}}}{node_type}/{{{self.NS_Y}}}" + geometry = data_element.find(f"{pref}Geometry") + if geometry is not None: + data["x"] = geometry.get("x") + data["y"] = geometry.get("y") + if node_label is None: + node_label = data_element.find(f"{pref}NodeLabel") + shape = data_element.find(f"{pref}Shape") + if shape is not None: + data["shape_type"] = shape.get("type") + if node_label is not None: + data["label"] = node_label.text + + # check all the different types of edges available in yEd. + for edge_type in [ + "PolyLineEdge", + "SplineEdge", + "QuadCurveEdge", + "BezierEdge", + "ArcEdge", + ]: + pref = f"{{{self.NS_Y}}}{edge_type}/{{{self.NS_Y}}}" + edge_label = data_element.find(f"{pref}EdgeLabel") + if edge_label is not None: + break + if edge_label is not None: + data["label"] = edge_label.text + elif text is None: + data[data_name] = "" + return data + + def find_graphml_keys(self, graph_element): + """Extracts all the keys and key defaults from the xml.""" + graphml_keys = {} + graphml_key_defaults = {} + for k in graph_element.findall(f"{{{self.NS_GRAPHML}}}key"): + attr_id = k.get("id") + attr_type = k.get("attr.type") + attr_name = k.get("attr.name") + yfiles_type = k.get("yfiles.type") + if yfiles_type is not None: + attr_name = yfiles_type + attr_type = "yfiles" + if attr_type is None: + attr_type = "string" + warnings.warn(f"No key type for id {attr_id}. Using string") + if attr_name is None: + raise nx.NetworkXError(f"Unknown key for id {attr_id}.") + graphml_keys[attr_id] = { + "name": attr_name, + "type": self.python_type[attr_type], + "for": k.get("for"), + } + # check for "default" sub-element of key element + default = k.find(f"{{{self.NS_GRAPHML}}}default") + if default is not None: + # Handle default values identically to data element values + python_type = graphml_keys[attr_id]["type"] + if python_type == bool: + graphml_key_defaults[attr_id] = self.convert_bool[ + default.text.lower() + ] + else: + graphml_key_defaults[attr_id] = python_type(default.text) + return graphml_keys, graphml_key_defaults diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/__init__.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/__init__.py new file mode 100644 index 00000000..532c71d7 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/__init__.py @@ -0,0 +1,19 @@ +""" +********* +JSON data +********* +Generate and parse JSON serializable data for NetworkX graphs. + +These formats are suitable for use with the d3.js examples https://d3js.org/ + +The three formats that you can generate with NetworkX are: + + - node-link like in the d3.js example https://bl.ocks.org/mbostock/4062045 + - tree like in the d3.js example https://bl.ocks.org/mbostock/4063550 + - adjacency like in the d3.js example https://bost.ocks.org/mike/miserables/ +""" + +from networkx.readwrite.json_graph.node_link import * +from networkx.readwrite.json_graph.adjacency import * +from networkx.readwrite.json_graph.tree import * +from networkx.readwrite.json_graph.cytoscape import * diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/adjacency.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/adjacency.py new file mode 100644 index 00000000..3b057475 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/adjacency.py @@ -0,0 +1,156 @@ +import networkx as nx + +__all__ = ["adjacency_data", "adjacency_graph"] + +_attrs = {"id": "id", "key": "key"} + + +def adjacency_data(G, attrs=_attrs): + """Returns data in adjacency format that is suitable for JSON serialization + and use in JavaScript documents. + + Parameters + ---------- + G : NetworkX graph + + attrs : dict + A dictionary that contains two keys 'id' and 'key'. The corresponding + values provide the attribute names for storing NetworkX-internal graph + data. The values should be unique. Default value: + :samp:`dict(id='id', key='key')`. + + If some user-defined graph data use these attribute names as data keys, + they may be silently dropped. + + Returns + ------- + data : dict + A dictionary with adjacency formatted data. + + Raises + ------ + NetworkXError + If values in attrs are not unique. + + Examples + -------- + >>> from networkx.readwrite import json_graph + >>> G = nx.Graph([(1, 2)]) + >>> data = json_graph.adjacency_data(G) + + To serialize with json + + >>> import json + >>> s = json.dumps(data) + + Notes + ----- + Graph, node, and link attributes will be written when using this format + but attribute keys must be strings if you want to serialize the resulting + data with JSON. + + The default value of attrs will be changed in a future release of NetworkX. + + See Also + -------- + adjacency_graph, node_link_data, tree_data + """ + multigraph = G.is_multigraph() + id_ = attrs["id"] + # Allow 'key' to be omitted from attrs if the graph is not a multigraph. + key = None if not multigraph else attrs["key"] + if id_ == key: + raise nx.NetworkXError("Attribute names are not unique.") + data = {} + data["directed"] = G.is_directed() + data["multigraph"] = multigraph + data["graph"] = list(G.graph.items()) + data["nodes"] = [] + data["adjacency"] = [] + for n, nbrdict in G.adjacency(): + data["nodes"].append({**G.nodes[n], id_: n}) + adj = [] + if multigraph: + for nbr, keys in nbrdict.items(): + for k, d in keys.items(): + adj.append({**d, id_: nbr, key: k}) + else: + for nbr, d in nbrdict.items(): + adj.append({**d, id_: nbr}) + data["adjacency"].append(adj) + return data + + +@nx._dispatchable(graphs=None, returns_graph=True) +def adjacency_graph(data, directed=False, multigraph=True, attrs=_attrs): + """Returns graph from adjacency data format. + + Parameters + ---------- + data : dict + Adjacency list formatted graph data + + directed : bool + If True, and direction not specified in data, return a directed graph. + + multigraph : bool + If True, and multigraph not specified in data, return a multigraph. + + attrs : dict + A dictionary that contains two keys 'id' and 'key'. The corresponding + values provide the attribute names for storing NetworkX-internal graph + data. The values should be unique. Default value: + :samp:`dict(id='id', key='key')`. + + Returns + ------- + G : NetworkX graph + A NetworkX graph object + + Examples + -------- + >>> from networkx.readwrite import json_graph + >>> G = nx.Graph([(1, 2)]) + >>> data = json_graph.adjacency_data(G) + >>> H = json_graph.adjacency_graph(data) + + Notes + ----- + The default value of attrs will be changed in a future release of NetworkX. + + See Also + -------- + adjacency_graph, node_link_data, tree_data + """ + multigraph = data.get("multigraph", multigraph) + directed = data.get("directed", directed) + if multigraph: + graph = nx.MultiGraph() + else: + graph = nx.Graph() + if directed: + graph = graph.to_directed() + id_ = attrs["id"] + # Allow 'key' to be omitted from attrs if the graph is not a multigraph. + key = None if not multigraph else attrs["key"] + graph.graph = dict(data.get("graph", [])) + mapping = [] + for d in data["nodes"]: + node_data = d.copy() + node = node_data.pop(id_) + mapping.append(node) + graph.add_node(node) + graph.nodes[node].update(node_data) + for i, d in enumerate(data["adjacency"]): + source = mapping[i] + for tdata in d: + target_data = tdata.copy() + target = target_data.pop(id_) + if not multigraph: + graph.add_edge(source, target) + graph[source][target].update(target_data) + else: + ky = target_data.pop(key, None) + graph.add_edge(source, target, key=ky) + graph[source][target][ky].update(target_data) + return graph diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/cytoscape.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/cytoscape.py new file mode 100644 index 00000000..2f3b2176 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/cytoscape.py @@ -0,0 +1,178 @@ +import networkx as nx + +__all__ = ["cytoscape_data", "cytoscape_graph"] + + +def cytoscape_data(G, name="name", ident="id"): + """Returns data in Cytoscape JSON format (cyjs). + + Parameters + ---------- + G : NetworkX Graph + The graph to convert to cytoscape format + name : string + A string which is mapped to the 'name' node element in cyjs format. + Must not have the same value as `ident`. + ident : string + A string which is mapped to the 'id' node element in cyjs format. + Must not have the same value as `name`. + + Returns + ------- + data: dict + A dictionary with cyjs formatted data. + + Raises + ------ + NetworkXError + If the values for `name` and `ident` are identical. + + See Also + -------- + cytoscape_graph: convert a dictionary in cyjs format to a graph + + References + ---------- + .. [1] Cytoscape user's manual: + http://manual.cytoscape.org/en/stable/index.html + + Examples + -------- + >>> G = nx.path_graph(2) + >>> nx.cytoscape_data(G) # doctest: +SKIP + {'data': [], + 'directed': False, + 'multigraph': False, + 'elements': {'nodes': [{'data': {'id': '0', 'value': 0, 'name': '0'}}, + {'data': {'id': '1', 'value': 1, 'name': '1'}}], + 'edges': [{'data': {'source': 0, 'target': 1}}]}} + """ + if name == ident: + raise nx.NetworkXError("name and ident must be different.") + + jsondata = {"data": list(G.graph.items())} + jsondata["directed"] = G.is_directed() + jsondata["multigraph"] = G.is_multigraph() + jsondata["elements"] = {"nodes": [], "edges": []} + nodes = jsondata["elements"]["nodes"] + edges = jsondata["elements"]["edges"] + + for i, j in G.nodes.items(): + n = {"data": j.copy()} + n["data"]["id"] = j.get(ident) or str(i) + n["data"]["value"] = i + n["data"]["name"] = j.get(name) or str(i) + nodes.append(n) + + if G.is_multigraph(): + for e in G.edges(keys=True): + n = {"data": G.adj[e[0]][e[1]][e[2]].copy()} + n["data"]["source"] = e[0] + n["data"]["target"] = e[1] + n["data"]["key"] = e[2] + edges.append(n) + else: + for e in G.edges(): + n = {"data": G.adj[e[0]][e[1]].copy()} + n["data"]["source"] = e[0] + n["data"]["target"] = e[1] + edges.append(n) + return jsondata + + +@nx._dispatchable(graphs=None, returns_graph=True) +def cytoscape_graph(data, name="name", ident="id"): + """ + Create a NetworkX graph from a dictionary in cytoscape JSON format. + + Parameters + ---------- + data : dict + A dictionary of data conforming to cytoscape JSON format. + name : string + A string which is mapped to the 'name' node element in cyjs format. + Must not have the same value as `ident`. + ident : string + A string which is mapped to the 'id' node element in cyjs format. + Must not have the same value as `name`. + + Returns + ------- + graph : a NetworkX graph instance + The `graph` can be an instance of `Graph`, `DiGraph`, `MultiGraph`, or + `MultiDiGraph` depending on the input data. + + Raises + ------ + NetworkXError + If the `name` and `ident` attributes are identical. + + See Also + -------- + cytoscape_data: convert a NetworkX graph to a dict in cyjs format + + References + ---------- + .. [1] Cytoscape user's manual: + http://manual.cytoscape.org/en/stable/index.html + + Examples + -------- + >>> data_dict = { + ... "data": [], + ... "directed": False, + ... "multigraph": False, + ... "elements": { + ... "nodes": [ + ... {"data": {"id": "0", "value": 0, "name": "0"}}, + ... {"data": {"id": "1", "value": 1, "name": "1"}}, + ... ], + ... "edges": [{"data": {"source": 0, "target": 1}}], + ... }, + ... } + >>> G = nx.cytoscape_graph(data_dict) + >>> G.name + '' + >>> G.nodes() + NodeView((0, 1)) + >>> G.nodes(data=True)[0] + {'id': '0', 'value': 0, 'name': '0'} + >>> G.edges(data=True) + EdgeDataView([(0, 1, {'source': 0, 'target': 1})]) + """ + if name == ident: + raise nx.NetworkXError("name and ident must be different.") + + multigraph = data.get("multigraph") + directed = data.get("directed") + if multigraph: + graph = nx.MultiGraph() + else: + graph = nx.Graph() + if directed: + graph = graph.to_directed() + graph.graph = dict(data.get("data")) + for d in data["elements"]["nodes"]: + node_data = d["data"].copy() + node = d["data"]["value"] + + if d["data"].get(name): + node_data[name] = d["data"].get(name) + if d["data"].get(ident): + node_data[ident] = d["data"].get(ident) + + graph.add_node(node) + graph.nodes[node].update(node_data) + + for d in data["elements"]["edges"]: + edge_data = d["data"].copy() + sour = d["data"]["source"] + targ = d["data"]["target"] + if multigraph: + key = d["data"].get("key", 0) + graph.add_edge(sour, targ, key=key) + graph.edges[sour, targ, key].update(edge_data) + else: + graph.add_edge(sour, targ) + graph.edges[sour, targ].update(edge_data) + return graph diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/node_link.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/node_link.py new file mode 100644 index 00000000..63ca9789 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/node_link.py @@ -0,0 +1,330 @@ +import warnings +from itertools import count + +import networkx as nx + +__all__ = ["node_link_data", "node_link_graph"] + + +def _to_tuple(x): + """Converts lists to tuples, including nested lists. + + All other non-list inputs are passed through unmodified. This function is + intended to be used to convert potentially nested lists from json files + into valid nodes. + + Examples + -------- + >>> _to_tuple([1, 2, [3, 4]]) + (1, 2, (3, 4)) + """ + if not isinstance(x, tuple | list): + return x + return tuple(map(_to_tuple, x)) + + +def node_link_data( + G, + *, + source="source", + target="target", + name="id", + key="key", + edges=None, + nodes="nodes", + link=None, +): + """Returns data in node-link format that is suitable for JSON serialization + and use in JavaScript documents. + + Parameters + ---------- + G : NetworkX graph + source : string + A string that provides the 'source' attribute name for storing NetworkX-internal graph data. + target : string + A string that provides the 'target' attribute name for storing NetworkX-internal graph data. + name : string + A string that provides the 'name' attribute name for storing NetworkX-internal graph data. + key : string + A string that provides the 'key' attribute name for storing NetworkX-internal graph data. + edges : string + A string that provides the 'edges' attribute name for storing NetworkX-internal graph data. + nodes : string + A string that provides the 'nodes' attribute name for storing NetworkX-internal graph data. + link : string + .. deprecated:: 3.4 + + The `link` argument is deprecated and will be removed in version `3.6`. + Use the `edges` keyword instead. + + A string that provides the 'edges' attribute name for storing NetworkX-internal graph data. + + Returns + ------- + data : dict + A dictionary with node-link formatted data. + + Raises + ------ + NetworkXError + If the values of 'source', 'target' and 'key' are not unique. + + Examples + -------- + >>> from pprint import pprint + >>> G = nx.Graph([("A", "B")]) + >>> data1 = nx.node_link_data(G, edges="edges") + >>> pprint(data1) + {'directed': False, + 'edges': [{'source': 'A', 'target': 'B'}], + 'graph': {}, + 'multigraph': False, + 'nodes': [{'id': 'A'}, {'id': 'B'}]} + + To serialize with JSON + + >>> import json + >>> s1 = json.dumps(data1) + >>> s1 + '{"directed": false, "multigraph": false, "graph": {}, "nodes": [{"id": "A"}, {"id": "B"}], "edges": [{"source": "A", "target": "B"}]}' + + A graph can also be serialized by passing `node_link_data` as an encoder function. + + >>> s1 = json.dumps(G, default=nx.node_link_data) + >>> s1 + '{"directed": false, "multigraph": false, "graph": {}, "nodes": [{"id": "A"}, {"id": "B"}], "links": [{"source": "A", "target": "B"}]}' + + The attribute names for storing NetworkX-internal graph data can + be specified as keyword options. + + >>> H = nx.gn_graph(2) + >>> data2 = nx.node_link_data( + ... H, edges="links", source="from", target="to", nodes="vertices" + ... ) + >>> pprint(data2) + {'directed': True, + 'graph': {}, + 'links': [{'from': 1, 'to': 0}], + 'multigraph': False, + 'vertices': [{'id': 0}, {'id': 1}]} + + Notes + ----- + Graph, node, and link attributes are stored in this format. Note that + attribute keys will be converted to strings in order to comply with JSON. + + Attribute 'key' is only used for multigraphs. + + To use `node_link_data` in conjunction with `node_link_graph`, + the keyword names for the attributes must match. + + See Also + -------- + node_link_graph, adjacency_data, tree_data + """ + # TODO: Remove between the lines when `link` deprecation expires + # ------------------------------------------------------------- + if link is not None: + warnings.warn( + "Keyword argument 'link' is deprecated; use 'edges' instead", + DeprecationWarning, + stacklevel=2, + ) + if edges is not None: + raise ValueError( + "Both 'edges' and 'link' are specified. Use 'edges', 'link' will be remove in a future release" + ) + else: + edges = link + else: + if edges is None: + warnings.warn( + ( + '\nThe default value will be `edges="edges" in NetworkX 3.6.\n\n' + "To make this warning go away, explicitly set the edges kwarg, e.g.:\n\n" + ' nx.node_link_data(G, edges="links") to preserve current behavior, or\n' + ' nx.node_link_data(G, edges="edges") for forward compatibility.' + ), + FutureWarning, + ) + edges = "links" + # ------------------------------------------------------------ + + multigraph = G.is_multigraph() + + # Allow 'key' to be omitted from attrs if the graph is not a multigraph. + key = None if not multigraph else key + if len({source, target, key}) < 3: + raise nx.NetworkXError("Attribute names are not unique.") + data = { + "directed": G.is_directed(), + "multigraph": multigraph, + "graph": G.graph, + nodes: [{**G.nodes[n], name: n} for n in G], + } + if multigraph: + data[edges] = [ + {**d, source: u, target: v, key: k} + for u, v, k, d in G.edges(keys=True, data=True) + ] + else: + data[edges] = [{**d, source: u, target: v} for u, v, d in G.edges(data=True)] + return data + + +@nx._dispatchable(graphs=None, returns_graph=True) +def node_link_graph( + data, + directed=False, + multigraph=True, + *, + source="source", + target="target", + name="id", + key="key", + edges=None, + nodes="nodes", + link=None, +): + """Returns graph from node-link data format. + + Useful for de-serialization from JSON. + + Parameters + ---------- + data : dict + node-link formatted graph data + + directed : bool + If True, and direction not specified in data, return a directed graph. + + multigraph : bool + If True, and multigraph not specified in data, return a multigraph. + + source : string + A string that provides the 'source' attribute name for storing NetworkX-internal graph data. + target : string + A string that provides the 'target' attribute name for storing NetworkX-internal graph data. + name : string + A string that provides the 'name' attribute name for storing NetworkX-internal graph data. + key : string + A string that provides the 'key' attribute name for storing NetworkX-internal graph data. + edges : string + A string that provides the 'edges' attribute name for storing NetworkX-internal graph data. + nodes : string + A string that provides the 'nodes' attribute name for storing NetworkX-internal graph data. + link : string + .. deprecated:: 3.4 + + The `link` argument is deprecated and will be removed in version `3.6`. + Use the `edges` keyword instead. + + A string that provides the 'edges' attribute name for storing NetworkX-internal graph data. + + Returns + ------- + G : NetworkX graph + A NetworkX graph object + + Examples + -------- + + Create data in node-link format by converting a graph. + + >>> from pprint import pprint + >>> G = nx.Graph([("A", "B")]) + >>> data = nx.node_link_data(G, edges="edges") + >>> pprint(data) + {'directed': False, + 'edges': [{'source': 'A', 'target': 'B'}], + 'graph': {}, + 'multigraph': False, + 'nodes': [{'id': 'A'}, {'id': 'B'}]} + + Revert data in node-link format to a graph. + + >>> H = nx.node_link_graph(data, edges="edges") + >>> print(H.edges) + [('A', 'B')] + + To serialize and deserialize a graph with JSON, + + >>> import json + >>> d = json.dumps(nx.node_link_data(G, edges="edges")) + >>> H = nx.node_link_graph(json.loads(d), edges="edges") + >>> print(G.edges, H.edges) + [('A', 'B')] [('A', 'B')] + + + Notes + ----- + Attribute 'key' is only used for multigraphs. + + To use `node_link_data` in conjunction with `node_link_graph`, + the keyword names for the attributes must match. + + See Also + -------- + node_link_data, adjacency_data, tree_data + """ + # TODO: Remove between the lines when `link` deprecation expires + # ------------------------------------------------------------- + if link is not None: + warnings.warn( + "Keyword argument 'link' is deprecated; use 'edges' instead", + DeprecationWarning, + stacklevel=2, + ) + if edges is not None: + raise ValueError( + "Both 'edges' and 'link' are specified. Use 'edges', 'link' will be remove in a future release" + ) + else: + edges = link + else: + if edges is None: + warnings.warn( + ( + '\nThe default value will be changed to `edges="edges" in NetworkX 3.6.\n\n' + "To make this warning go away, explicitly set the edges kwarg, e.g.:\n\n" + ' nx.node_link_graph(data, edges="links") to preserve current behavior, or\n' + ' nx.node_link_graph(data, edges="edges") for forward compatibility.' + ), + FutureWarning, + ) + edges = "links" + # ------------------------------------------------------------- + + multigraph = data.get("multigraph", multigraph) + directed = data.get("directed", directed) + if multigraph: + graph = nx.MultiGraph() + else: + graph = nx.Graph() + if directed: + graph = graph.to_directed() + + # Allow 'key' to be omitted from attrs if the graph is not a multigraph. + key = None if not multigraph else key + graph.graph = data.get("graph", {}) + c = count() + for d in data[nodes]: + node = _to_tuple(d.get(name, next(c))) + nodedata = {str(k): v for k, v in d.items() if k != name} + graph.add_node(node, **nodedata) + for d in data[edges]: + src = tuple(d[source]) if isinstance(d[source], list) else d[source] + tgt = tuple(d[target]) if isinstance(d[target], list) else d[target] + if not multigraph: + edgedata = {str(k): v for k, v in d.items() if k != source and k != target} + graph.add_edge(src, tgt, **edgedata) + else: + ky = d.get(key, None) + edgedata = { + str(k): v + for k, v in d.items() + if k != source and k != target and k != key + } + graph.add_edge(src, tgt, ky, **edgedata) + return graph diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/__init__.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/__init__.py new file mode 100644 index 00000000..e69de29b --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/__init__.py diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_adjacency.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_adjacency.py new file mode 100644 index 00000000..37506382 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_adjacency.py @@ -0,0 +1,78 @@ +import copy +import json + +import pytest + +import networkx as nx +from networkx.readwrite.json_graph import adjacency_data, adjacency_graph +from networkx.utils import graphs_equal + + +class TestAdjacency: + def test_graph(self): + G = nx.path_graph(4) + H = adjacency_graph(adjacency_data(G)) + assert graphs_equal(G, H) + + def test_graph_attributes(self): + G = nx.path_graph(4) + G.add_node(1, color="red") + G.add_edge(1, 2, width=7) + G.graph["foo"] = "bar" + G.graph[1] = "one" + + H = adjacency_graph(adjacency_data(G)) + assert graphs_equal(G, H) + assert H.graph["foo"] == "bar" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + + d = json.dumps(adjacency_data(G)) + H = adjacency_graph(json.loads(d)) + assert graphs_equal(G, H) + assert H.graph["foo"] == "bar" + assert H.graph[1] == "one" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + + def test_digraph(self): + G = nx.DiGraph() + nx.add_path(G, [1, 2, 3]) + H = adjacency_graph(adjacency_data(G)) + assert H.is_directed() + assert graphs_equal(G, H) + + def test_multidigraph(self): + G = nx.MultiDiGraph() + nx.add_path(G, [1, 2, 3]) + H = adjacency_graph(adjacency_data(G)) + assert H.is_directed() + assert H.is_multigraph() + assert graphs_equal(G, H) + + def test_multigraph(self): + G = nx.MultiGraph() + G.add_edge(1, 2, key="first") + G.add_edge(1, 2, key="second", color="blue") + H = adjacency_graph(adjacency_data(G)) + assert graphs_equal(G, H) + assert H[1][2]["second"]["color"] == "blue" + + def test_input_data_is_not_modified_when_building_graph(self): + G = nx.path_graph(4) + input_data = adjacency_data(G) + orig_data = copy.deepcopy(input_data) + # Ensure input is unmodified by deserialisation + assert graphs_equal(G, adjacency_graph(input_data)) + assert input_data == orig_data + + def test_adjacency_form_json_serialisable(self): + G = nx.path_graph(4) + H = adjacency_graph(json.loads(json.dumps(adjacency_data(G)))) + assert graphs_equal(G, H) + + def test_exception(self): + with pytest.raises(nx.NetworkXError): + G = nx.MultiDiGraph() + attrs = {"id": "node", "key": "node"} + adjacency_data(G, attrs) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_cytoscape.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_cytoscape.py new file mode 100644 index 00000000..5d47f21f --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_cytoscape.py @@ -0,0 +1,78 @@ +import copy +import json + +import pytest + +import networkx as nx +from networkx.readwrite.json_graph import cytoscape_data, cytoscape_graph + + +def test_graph(): + G = nx.path_graph(4) + H = cytoscape_graph(cytoscape_data(G)) + assert nx.is_isomorphic(G, H) + + +def test_input_data_is_not_modified_when_building_graph(): + G = nx.path_graph(4) + input_data = cytoscape_data(G) + orig_data = copy.deepcopy(input_data) + # Ensure input is unmodified by cytoscape_graph (gh-4173) + cytoscape_graph(input_data) + assert input_data == orig_data + + +def test_graph_attributes(): + G = nx.path_graph(4) + G.add_node(1, color="red") + G.add_edge(1, 2, width=7) + G.graph["foo"] = "bar" + G.graph[1] = "one" + G.add_node(3, name="node", id="123") + + H = cytoscape_graph(cytoscape_data(G)) + assert H.graph["foo"] == "bar" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + assert H.nodes[3]["name"] == "node" + assert H.nodes[3]["id"] == "123" + + d = json.dumps(cytoscape_data(G)) + H = cytoscape_graph(json.loads(d)) + assert H.graph["foo"] == "bar" + assert H.graph[1] == "one" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + assert H.nodes[3]["name"] == "node" + assert H.nodes[3]["id"] == "123" + + +def test_digraph(): + G = nx.DiGraph() + nx.add_path(G, [1, 2, 3]) + H = cytoscape_graph(cytoscape_data(G)) + assert H.is_directed() + assert nx.is_isomorphic(G, H) + + +def test_multidigraph(): + G = nx.MultiDiGraph() + nx.add_path(G, [1, 2, 3]) + H = cytoscape_graph(cytoscape_data(G)) + assert H.is_directed() + assert H.is_multigraph() + + +def test_multigraph(): + G = nx.MultiGraph() + G.add_edge(1, 2, key="first") + G.add_edge(1, 2, key="second", color="blue") + H = cytoscape_graph(cytoscape_data(G)) + assert nx.is_isomorphic(G, H) + assert H[1][2]["second"]["color"] == "blue" + + +def test_exception(): + with pytest.raises(nx.NetworkXError): + G = nx.MultiDiGraph() + cytoscape_data(G, name="foo", ident="foo") diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_node_link.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_node_link.py new file mode 100644 index 00000000..f903f606 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_node_link.py @@ -0,0 +1,175 @@ +import json + +import pytest + +import networkx as nx +from networkx.readwrite.json_graph import node_link_data, node_link_graph + + +def test_node_link_edges_default_future_warning(): + "Test FutureWarning is raised when `edges=None` in node_link_data and node_link_graph" + G = nx.Graph([(1, 2)]) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + data = nx.node_link_data(G) # edges=None, the default + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = nx.node_link_graph(data) # edges=None, the default + + +def test_node_link_deprecated_link_param(): + G = nx.Graph([(1, 2)]) + with pytest.warns(DeprecationWarning, match="Keyword argument 'link'"): + data = nx.node_link_data(G, link="links") + with pytest.warns(DeprecationWarning, match="Keyword argument 'link'"): + H = nx.node_link_graph(data, link="links") + + +class TestNodeLink: + # TODO: To be removed when signature change complete + def test_custom_attrs_dep(self): + G = nx.path_graph(4) + G.add_node(1, color="red") + G.add_edge(1, 2, width=7) + G.graph[1] = "one" + G.graph["foo"] = "bar" + + attrs = { + "source": "c_source", + "target": "c_target", + "name": "c_id", + "key": "c_key", + "link": "c_links", + } + + H = node_link_graph(node_link_data(G, **attrs), multigraph=False, **attrs) + assert nx.is_isomorphic(G, H) + assert H.graph["foo"] == "bar" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + + # provide only a partial dictionary of keywords. + # This is similar to an example in the doc string + attrs = { + "link": "c_links", + "source": "c_source", + "target": "c_target", + } + H = node_link_graph(node_link_data(G, **attrs), multigraph=False, **attrs) + assert nx.is_isomorphic(G, H) + assert H.graph["foo"] == "bar" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + + def test_exception_dep(self): + G = nx.MultiDiGraph() + with pytest.raises(nx.NetworkXError): + with pytest.warns(FutureWarning, match="\nThe default value will be"): + node_link_data(G, name="node", source="node", target="node", key="node") + + def test_graph(self): + G = nx.path_graph(4) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(node_link_data(G)) + assert nx.is_isomorphic(G, H) + + def test_graph_attributes(self): + G = nx.path_graph(4) + G.add_node(1, color="red") + G.add_edge(1, 2, width=7) + G.graph[1] = "one" + G.graph["foo"] = "bar" + + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(node_link_data(G)) + assert H.graph["foo"] == "bar" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + + with pytest.warns(FutureWarning, match="\nThe default value will be"): + d = json.dumps(node_link_data(G)) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(json.loads(d)) + assert H.graph["foo"] == "bar" + assert H.graph["1"] == "one" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 + + def test_digraph(self): + G = nx.DiGraph() + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(node_link_data(G)) + assert H.is_directed() + + def test_multigraph(self): + G = nx.MultiGraph() + G.add_edge(1, 2, key="first") + G.add_edge(1, 2, key="second", color="blue") + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(node_link_data(G)) + assert nx.is_isomorphic(G, H) + assert H[1][2]["second"]["color"] == "blue" + + def test_graph_with_tuple_nodes(self): + G = nx.Graph() + G.add_edge((0, 0), (1, 0), color=[255, 255, 0]) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + d = node_link_data(G) + dumped_d = json.dumps(d) + dd = json.loads(dumped_d) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(dd) + assert H.nodes[(0, 0)] == G.nodes[(0, 0)] + assert H[(0, 0)][(1, 0)]["color"] == [255, 255, 0] + + def test_unicode_keys(self): + q = "qualité" + G = nx.Graph() + G.add_node(1, **{q: q}) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + s = node_link_data(G) + output = json.dumps(s, ensure_ascii=False) + data = json.loads(output) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(data) + assert H.nodes[1][q] == q + + def test_exception(self): + G = nx.MultiDiGraph() + attrs = {"name": "node", "source": "node", "target": "node", "key": "node"} + with pytest.raises(nx.NetworkXError): + with pytest.warns(FutureWarning, match="\nThe default value will be"): + node_link_data(G, **attrs) + + def test_string_ids(self): + q = "qualité" + G = nx.DiGraph() + G.add_node("A") + G.add_node(q) + G.add_edge("A", q) + with pytest.warns(FutureWarning, match="\nThe default value will be"): + data = node_link_data(G) + assert data["links"][0]["source"] == "A" + assert data["links"][0]["target"] == q + with pytest.warns(FutureWarning, match="\nThe default value will be"): + H = node_link_graph(data) + assert nx.is_isomorphic(G, H) + + def test_custom_attrs(self): + G = nx.path_graph(4) + G.add_node(1, color="red") + G.add_edge(1, 2, width=7) + G.graph[1] = "one" + G.graph["foo"] = "bar" + + attrs = { + "source": "c_source", + "target": "c_target", + "name": "c_id", + "key": "c_key", + "link": "c_links", + } + + H = node_link_graph(node_link_data(G, **attrs), multigraph=False, **attrs) + assert nx.is_isomorphic(G, H) + assert H.graph["foo"] == "bar" + assert H.nodes[1]["color"] == "red" + assert H[1][2]["width"] == 7 diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_tree.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_tree.py new file mode 100644 index 00000000..643a14d8 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tests/test_tree.py @@ -0,0 +1,48 @@ +import json + +import pytest + +import networkx as nx +from networkx.readwrite.json_graph import tree_data, tree_graph + + +def test_graph(): + G = nx.DiGraph() + G.add_nodes_from([1, 2, 3], color="red") + G.add_edge(1, 2, foo=7) + G.add_edge(1, 3, foo=10) + G.add_edge(3, 4, foo=10) + H = tree_graph(tree_data(G, 1)) + assert nx.is_isomorphic(G, H) + + +def test_graph_attributes(): + G = nx.DiGraph() + G.add_nodes_from([1, 2, 3], color="red") + G.add_edge(1, 2, foo=7) + G.add_edge(1, 3, foo=10) + G.add_edge(3, 4, foo=10) + H = tree_graph(tree_data(G, 1)) + assert H.nodes[1]["color"] == "red" + + d = json.dumps(tree_data(G, 1)) + H = tree_graph(json.loads(d)) + assert H.nodes[1]["color"] == "red" + + +def test_exceptions(): + with pytest.raises(TypeError, match="is not a tree."): + G = nx.complete_graph(3) + tree_data(G, 0) + with pytest.raises(TypeError, match="is not directed."): + G = nx.path_graph(3) + tree_data(G, 0) + with pytest.raises(TypeError, match="is not weakly connected."): + G = nx.path_graph(3, create_using=nx.DiGraph) + G.add_edge(2, 0) + G.add_node(3) + tree_data(G, 0) + with pytest.raises(nx.NetworkXError, match="must be different."): + G = nx.MultiDiGraph() + G.add_node(0) + tree_data(G, 0, ident="node", children="node") diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tree.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tree.py new file mode 100644 index 00000000..22b07b09 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/json_graph/tree.py @@ -0,0 +1,137 @@ +from itertools import chain + +import networkx as nx + +__all__ = ["tree_data", "tree_graph"] + + +def tree_data(G, root, ident="id", children="children"): + """Returns data in tree format that is suitable for JSON serialization + and use in JavaScript documents. + + Parameters + ---------- + G : NetworkX graph + G must be an oriented tree + + root : node + The root of the tree + + ident : string + Attribute name for storing NetworkX-internal graph data. `ident` must + have a different value than `children`. The default is 'id'. + + children : string + Attribute name for storing NetworkX-internal graph data. `children` + must have a different value than `ident`. The default is 'children'. + + Returns + ------- + data : dict + A dictionary with node-link formatted data. + + Raises + ------ + NetworkXError + If `children` and `ident` attributes are identical. + + Examples + -------- + >>> from networkx.readwrite import json_graph + >>> G = nx.DiGraph([(1, 2)]) + >>> data = json_graph.tree_data(G, root=1) + + To serialize with json + + >>> import json + >>> s = json.dumps(data) + + Notes + ----- + Node attributes are stored in this format but keys + for attributes must be strings if you want to serialize with JSON. + + Graph and edge attributes are not stored. + + See Also + -------- + tree_graph, node_link_data, adjacency_data + """ + if G.number_of_nodes() != G.number_of_edges() + 1: + raise TypeError("G is not a tree.") + if not G.is_directed(): + raise TypeError("G is not directed.") + if not nx.is_weakly_connected(G): + raise TypeError("G is not weakly connected.") + + if ident == children: + raise nx.NetworkXError("The values for `id` and `children` must be different.") + + def add_children(n, G): + nbrs = G[n] + if len(nbrs) == 0: + return [] + children_ = [] + for child in nbrs: + d = {**G.nodes[child], ident: child} + c = add_children(child, G) + if c: + d[children] = c + children_.append(d) + return children_ + + return {**G.nodes[root], ident: root, children: add_children(root, G)} + + +@nx._dispatchable(graphs=None, returns_graph=True) +def tree_graph(data, ident="id", children="children"): + """Returns graph from tree data format. + + Parameters + ---------- + data : dict + Tree formatted graph data + + ident : string + Attribute name for storing NetworkX-internal graph data. `ident` must + have a different value than `children`. The default is 'id'. + + children : string + Attribute name for storing NetworkX-internal graph data. `children` + must have a different value than `ident`. The default is 'children'. + + Returns + ------- + G : NetworkX DiGraph + + Examples + -------- + >>> from networkx.readwrite import json_graph + >>> G = nx.DiGraph([(1, 2)]) + >>> data = json_graph.tree_data(G, root=1) + >>> H = json_graph.tree_graph(data) + + See Also + -------- + tree_data, node_link_data, adjacency_data + """ + graph = nx.DiGraph() + + def add_children(parent, children_): + for data in children_: + child = data[ident] + graph.add_edge(parent, child) + grandchildren = data.get(children, []) + if grandchildren: + add_children(child, grandchildren) + nodedata = { + str(k): v for k, v in data.items() if k != ident and k != children + } + graph.add_node(child, **nodedata) + + root = data[ident] + children_ = data.get(children, []) + nodedata = {str(k): v for k, v in data.items() if k != ident and k != children} + graph.add_node(root, **nodedata) + add_children(root, children_) + return graph diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/leda.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/leda.py new file mode 100644 index 00000000..9fb57db1 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/leda.py @@ -0,0 +1,108 @@ +""" +Read graphs in LEDA format. + +LEDA is a C++ class library for efficient data types and algorithms. + +Format +------ +See http://www.algorithmic-solutions.info/leda_guide/graphs/leda_native_graph_fileformat.html + +""" +# Original author: D. Eppstein, UC Irvine, August 12, 2003. +# The original code at http://www.ics.uci.edu/~eppstein/PADS/ is public domain. + +__all__ = ["read_leda", "parse_leda"] + +import networkx as nx +from networkx.exception import NetworkXError +from networkx.utils import open_file + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_leda(path, encoding="UTF-8"): + """Read graph in LEDA format from path. + + Parameters + ---------- + path : file or string + File or filename to read. Filenames ending in .gz or .bz2 will be + uncompressed. + + Returns + ------- + G : NetworkX graph + + Examples + -------- + G=nx.read_leda('file.leda') + + References + ---------- + .. [1] http://www.algorithmic-solutions.info/leda_guide/graphs/leda_native_graph_fileformat.html + """ + lines = (line.decode(encoding) for line in path) + G = parse_leda(lines) + return G + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_leda(lines): + """Read graph in LEDA format from string or iterable. + + Parameters + ---------- + lines : string or iterable + Data in LEDA format. + + Returns + ------- + G : NetworkX graph + + Examples + -------- + G=nx.parse_leda(string) + + References + ---------- + .. [1] http://www.algorithmic-solutions.info/leda_guide/graphs/leda_native_graph_fileformat.html + """ + if isinstance(lines, str): + lines = iter(lines.split("\n")) + lines = iter( + [ + line.rstrip("\n") + for line in lines + if not (line.startswith(("#", "\n")) or line == "") + ] + ) + for i in range(3): + next(lines) + # Graph + du = int(next(lines)) # -1=directed, -2=undirected + if du == -1: + G = nx.DiGraph() + else: + G = nx.Graph() + + # Nodes + n = int(next(lines)) # number of nodes + node = {} + for i in range(1, n + 1): # LEDA counts from 1 to n + symbol = next(lines).rstrip().strip("|{}| ") + if symbol == "": + symbol = str(i) # use int if no label - could be trouble + node[i] = symbol + + G.add_nodes_from([s for i, s in node.items()]) + + # Edges + m = int(next(lines)) # number of edges + for i in range(m): + try: + s, t, reversal, label = next(lines).split() + except BaseException as err: + raise NetworkXError(f"Too few fields in LEDA.GRAPH edge {i+1}") from err + # BEWARE: no handling of reversal edges + G.add_edge(node[int(s)], node[int(t)], label=label[2:-2]) + return G diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/multiline_adjlist.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/multiline_adjlist.py new file mode 100644 index 00000000..808445db --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/multiline_adjlist.py @@ -0,0 +1,393 @@ +""" +************************* +Multi-line Adjacency List +************************* +Read and write NetworkX graphs as multi-line adjacency lists. + +The multi-line adjacency list format is useful for graphs with +nodes that can be meaningfully represented as strings. With this format +simple edge data can be stored but node or graph data is not. + +Format +------ +The first label in a line is the source node label followed by the node degree +d. The next d lines are target node labels and optional edge data. +That pattern repeats for all nodes in the graph. + +The graph with edges a-b, a-c, d-e can be represented as the following +adjacency list (anything following the # in a line is a comment):: + + # example.multiline-adjlist + a 2 + b + c + d 1 + e +""" + +__all__ = [ + "generate_multiline_adjlist", + "write_multiline_adjlist", + "parse_multiline_adjlist", + "read_multiline_adjlist", +] + +import networkx as nx +from networkx.utils import open_file + + +def generate_multiline_adjlist(G, delimiter=" "): + """Generate a single line of the graph G in multiline adjacency list format. + + Parameters + ---------- + G : NetworkX graph + + delimiter : string, optional + Separator for node labels + + Returns + ------- + lines : string + Lines of data in multiline adjlist format. + + Examples + -------- + >>> G = nx.lollipop_graph(4, 3) + >>> for line in nx.generate_multiline_adjlist(G): + ... print(line) + 0 3 + 1 {} + 2 {} + 3 {} + 1 2 + 2 {} + 3 {} + 2 1 + 3 {} + 3 1 + 4 {} + 4 1 + 5 {} + 5 1 + 6 {} + 6 0 + + See Also + -------- + write_multiline_adjlist, read_multiline_adjlist + """ + if G.is_directed(): + if G.is_multigraph(): + for s, nbrs in G.adjacency(): + nbr_edges = [ + (u, data) + for u, datadict in nbrs.items() + for key, data in datadict.items() + ] + deg = len(nbr_edges) + yield str(s) + delimiter + str(deg) + for u, d in nbr_edges: + if d is None: + yield str(u) + else: + yield str(u) + delimiter + str(d) + else: # directed single edges + for s, nbrs in G.adjacency(): + deg = len(nbrs) + yield str(s) + delimiter + str(deg) + for u, d in nbrs.items(): + if d is None: + yield str(u) + else: + yield str(u) + delimiter + str(d) + else: # undirected + if G.is_multigraph(): + seen = set() # helper dict used to avoid duplicate edges + for s, nbrs in G.adjacency(): + nbr_edges = [ + (u, data) + for u, datadict in nbrs.items() + if u not in seen + for key, data in datadict.items() + ] + deg = len(nbr_edges) + yield str(s) + delimiter + str(deg) + for u, d in nbr_edges: + if d is None: + yield str(u) + else: + yield str(u) + delimiter + str(d) + seen.add(s) + else: # undirected single edges + seen = set() # helper dict used to avoid duplicate edges + for s, nbrs in G.adjacency(): + nbr_edges = [(u, d) for u, d in nbrs.items() if u not in seen] + deg = len(nbr_edges) + yield str(s) + delimiter + str(deg) + for u, d in nbr_edges: + if d is None: + yield str(u) + else: + yield str(u) + delimiter + str(d) + seen.add(s) + + +@open_file(1, mode="wb") +def write_multiline_adjlist(G, path, delimiter=" ", comments="#", encoding="utf-8"): + """Write the graph G in multiline adjacency list format to path + + Parameters + ---------- + G : NetworkX graph + + path : string or file + Filename or file handle to write to. + Filenames ending in .gz or .bz2 will be compressed. + + comments : string, optional + Marker for comment lines + + delimiter : string, optional + Separator for node labels + + encoding : string, optional + Text encoding. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_multiline_adjlist(G, "test.adjlist") + + The path can be a file handle or a string with the name of the file. If a + file handle is provided, it has to be opened in 'wb' mode. + + >>> fh = open("test.adjlist", "wb") + >>> nx.write_multiline_adjlist(G, fh) + + Filenames ending in .gz or .bz2 will be compressed. + + >>> nx.write_multiline_adjlist(G, "test.adjlist.gz") + + See Also + -------- + read_multiline_adjlist + """ + import sys + import time + + pargs = comments + " ".join(sys.argv) + header = ( + f"{pargs}\n" + + comments + + f" GMT {time.asctime(time.gmtime())}\n" + + comments + + f" {G.name}\n" + ) + path.write(header.encode(encoding)) + + for multiline in generate_multiline_adjlist(G, delimiter): + multiline += "\n" + path.write(multiline.encode(encoding)) + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_multiline_adjlist( + lines, comments="#", delimiter=None, create_using=None, nodetype=None, edgetype=None +): + """Parse lines of a multiline adjacency list representation of a graph. + + Parameters + ---------- + lines : list or iterator of strings + Input data in multiline adjlist format + + create_using : NetworkX graph constructor, optional (default=nx.Graph) + Graph type to create. If graph instance, then cleared before populated. + + nodetype : Python type, optional + Convert nodes to this type. + + edgetype : Python type, optional + Convert edges to this type. + + comments : string, optional + Marker for comment lines + + delimiter : string, optional + Separator for node labels. The default is whitespace. + + Returns + ------- + G: NetworkX graph + The graph corresponding to the lines in multiline adjacency list format. + + Examples + -------- + >>> lines = [ + ... "1 2", + ... "2 {'weight':3, 'name': 'Frodo'}", + ... "3 {}", + ... "2 1", + ... "5 {'weight':6, 'name': 'Saruman'}", + ... ] + >>> G = nx.parse_multiline_adjlist(iter(lines), nodetype=int) + >>> list(G) + [1, 2, 3, 5] + + """ + from ast import literal_eval + + G = nx.empty_graph(0, create_using) + for line in lines: + p = line.find(comments) + if p >= 0: + line = line[:p] + if not line: + continue + try: + (u, deg) = line.rstrip("\n").split(delimiter) + deg = int(deg) + except BaseException as err: + raise TypeError(f"Failed to read node and degree on line ({line})") from err + if nodetype is not None: + try: + u = nodetype(u) + except BaseException as err: + raise TypeError( + f"Failed to convert node ({u}) to type {nodetype}" + ) from err + G.add_node(u) + for i in range(deg): + while True: + try: + line = next(lines) + except StopIteration as err: + msg = f"Failed to find neighbor for node ({u})" + raise TypeError(msg) from err + p = line.find(comments) + if p >= 0: + line = line[:p] + if line: + break + vlist = line.rstrip("\n").split(delimiter) + numb = len(vlist) + if numb < 1: + continue # isolated node + v = vlist.pop(0) + data = "".join(vlist) + if nodetype is not None: + try: + v = nodetype(v) + except BaseException as err: + raise TypeError( + f"Failed to convert node ({v}) to type {nodetype}" + ) from err + if edgetype is not None: + try: + edgedata = {"weight": edgetype(data)} + except BaseException as err: + raise TypeError( + f"Failed to convert edge data ({data}) to type {edgetype}" + ) from err + else: + try: # try to evaluate + edgedata = literal_eval(data) + except: + edgedata = {} + G.add_edge(u, v, **edgedata) + + return G + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_multiline_adjlist( + path, + comments="#", + delimiter=None, + create_using=None, + nodetype=None, + edgetype=None, + encoding="utf-8", +): + """Read graph in multi-line adjacency list format from path. + + Parameters + ---------- + path : string or file + Filename or file handle to read. + Filenames ending in .gz or .bz2 will be uncompressed. + + create_using : NetworkX graph constructor, optional (default=nx.Graph) + Graph type to create. If graph instance, then cleared before populated. + + nodetype : Python type, optional + Convert nodes to this type. + + edgetype : Python type, optional + Convert edge data to this type. + + comments : string, optional + Marker for comment lines + + delimiter : string, optional + Separator for node labels. The default is whitespace. + + Returns + ------- + G: NetworkX graph + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_multiline_adjlist(G, "test.adjlist") + >>> G = nx.read_multiline_adjlist("test.adjlist") + + The path can be a file or a string with the name of the file. If a + file s provided, it has to be opened in 'rb' mode. + + >>> fh = open("test.adjlist", "rb") + >>> G = nx.read_multiline_adjlist(fh) + + Filenames ending in .gz or .bz2 will be compressed. + + >>> nx.write_multiline_adjlist(G, "test.adjlist.gz") + >>> G = nx.read_multiline_adjlist("test.adjlist.gz") + + The optional nodetype is a function to convert node strings to nodetype. + + For example + + >>> G = nx.read_multiline_adjlist("test.adjlist", nodetype=int) + + will attempt to convert all nodes to integer type. + + The optional edgetype is a function to convert edge data strings to + edgetype. + + >>> G = nx.read_multiline_adjlist("test.adjlist") + + The optional create_using parameter is a NetworkX graph container. + The default is Graph(), an undirected graph. To read the data as + a directed graph use + + >>> G = nx.read_multiline_adjlist("test.adjlist", create_using=nx.DiGraph) + + Notes + ----- + This format does not store graph, node, or edge data. + + See Also + -------- + write_multiline_adjlist + """ + lines = (line.decode(encoding) for line in path) + return parse_multiline_adjlist( + lines, + comments=comments, + delimiter=delimiter, + create_using=create_using, + nodetype=nodetype, + edgetype=edgetype, + ) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/p2g.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/p2g.py new file mode 100644 index 00000000..804adb23 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/p2g.py @@ -0,0 +1,105 @@ +""" +This module provides the following: read and write of p2g format +used in metabolic pathway studies. + +See https://web.archive.org/web/20080626113807/http://www.cs.purdue.edu/homes/koyuturk/pathway/ for a description. + +The summary is included here: + +A file that describes a uniquely labeled graph (with extension ".gr") +format looks like the following: + + +name +3 4 +a +1 2 +b + +c +0 2 + +"name" is simply a description of what the graph corresponds to. The +second line displays the number of nodes and number of edges, +respectively. This sample graph contains three nodes labeled "a", "b", +and "c". The rest of the graph contains two lines for each node. The +first line for a node contains the node label. After the declaration +of the node label, the out-edges of that node in the graph are +provided. For instance, "a" is linked to nodes 1 and 2, which are +labeled "b" and "c", while the node labeled "b" has no outgoing +edges. Observe that node labeled "c" has an outgoing edge to +itself. Indeed, self-loops are allowed. Node index starts from 0. + +""" + +import networkx as nx +from networkx.utils import open_file + + +@open_file(1, mode="w") +def write_p2g(G, path, encoding="utf-8"): + """Write NetworkX graph in p2g format. + + Notes + ----- + This format is meant to be used with directed graphs with + possible self loops. + """ + path.write((f"{G.name}\n").encode(encoding)) + path.write((f"{G.order()} {G.size()}\n").encode(encoding)) + nodes = list(G) + # make dictionary mapping nodes to integers + nodenumber = dict(zip(nodes, range(len(nodes)))) + for n in nodes: + path.write((f"{n}\n").encode(encoding)) + for nbr in G.neighbors(n): + path.write((f"{nodenumber[nbr]} ").encode(encoding)) + path.write("\n".encode(encoding)) + + +@open_file(0, mode="r") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_p2g(path, encoding="utf-8"): + """Read graph in p2g format from path. + + Returns + ------- + MultiDiGraph + + Notes + ----- + If you want a DiGraph (with no self loops allowed and no edge data) + use D=nx.DiGraph(read_p2g(path)) + """ + lines = (line.decode(encoding) for line in path) + G = parse_p2g(lines) + return G + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_p2g(lines): + """Parse p2g format graph from string or iterable. + + Returns + ------- + MultiDiGraph + """ + description = next(lines).strip() + # are multiedges (parallel edges) allowed? + G = nx.MultiDiGraph(name=description, selfloops=True) + nnodes, nedges = map(int, next(lines).split()) + nodelabel = {} + nbrs = {} + # loop over the nodes keeping track of node labels and out neighbors + # defer adding edges until all node labels are known + for i in range(nnodes): + n = next(lines).strip() + nodelabel[i] = n + G.add_node(n) + nbrs[n] = map(int, next(lines).split()) + # now we know all of the node labels so we can add the edges + # with the correct labels + for n in G: + for nbr in nbrs[n]: + G.add_edge(n, nodelabel[nbr]) + return G diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/pajek.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/pajek.py new file mode 100644 index 00000000..f148f162 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/pajek.py @@ -0,0 +1,286 @@ +""" +***** +Pajek +***** +Read graphs in Pajek format. + +This implementation handles directed and undirected graphs including +those with self loops and parallel edges. + +Format +------ +See http://vlado.fmf.uni-lj.si/pub/networks/pajek/doc/draweps.htm +for format information. + +""" + +import warnings + +import networkx as nx +from networkx.utils import open_file + +__all__ = ["read_pajek", "parse_pajek", "generate_pajek", "write_pajek"] + + +def generate_pajek(G): + """Generate lines in Pajek graph format. + + Parameters + ---------- + G : graph + A Networkx graph + + References + ---------- + See http://vlado.fmf.uni-lj.si/pub/networks/pajek/doc/draweps.htm + for format information. + """ + if G.name == "": + name = "NetworkX" + else: + name = G.name + # Apparently many Pajek format readers can't process this line + # So we'll leave it out for now. + # yield '*network %s'%name + + # write nodes with attributes + yield f"*vertices {G.order()}" + nodes = list(G) + # make dictionary mapping nodes to integers + nodenumber = dict(zip(nodes, range(1, len(nodes) + 1))) + for n in nodes: + # copy node attributes and pop mandatory attributes + # to avoid duplication. + na = G.nodes.get(n, {}).copy() + x = na.pop("x", 0.0) + y = na.pop("y", 0.0) + try: + id = int(na.pop("id", nodenumber[n])) + except ValueError as err: + err.args += ( + ( + "Pajek format requires 'id' to be an int()." + " Refer to the 'Relabeling nodes' section." + ), + ) + raise + nodenumber[n] = id + shape = na.pop("shape", "ellipse") + s = " ".join(map(make_qstr, (id, n, x, y, shape))) + # only optional attributes are left in na. + for k, v in na.items(): + if isinstance(v, str) and v.strip() != "": + s += f" {make_qstr(k)} {make_qstr(v)}" + else: + warnings.warn( + f"Node attribute {k} is not processed. {('Empty attribute' if isinstance(v, str) else 'Non-string attribute')}." + ) + yield s + + # write edges with attributes + if G.is_directed(): + yield "*arcs" + else: + yield "*edges" + for u, v, edgedata in G.edges(data=True): + d = edgedata.copy() + value = d.pop("weight", 1.0) # use 1 as default edge value + s = " ".join(map(make_qstr, (nodenumber[u], nodenumber[v], value))) + for k, v in d.items(): + if isinstance(v, str) and v.strip() != "": + s += f" {make_qstr(k)} {make_qstr(v)}" + else: + warnings.warn( + f"Edge attribute {k} is not processed. {('Empty attribute' if isinstance(v, str) else 'Non-string attribute')}." + ) + yield s + + +@open_file(1, mode="wb") +def write_pajek(G, path, encoding="UTF-8"): + """Write graph in Pajek format to path. + + Parameters + ---------- + G : graph + A Networkx graph + path : file or string + File or filename to write. + Filenames ending in .gz or .bz2 will be compressed. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_pajek(G, "test.net") + + Warnings + -------- + Optional node attributes and edge attributes must be non-empty strings. + Otherwise it will not be written into the file. You will need to + convert those attributes to strings if you want to keep them. + + References + ---------- + See http://vlado.fmf.uni-lj.si/pub/networks/pajek/doc/draweps.htm + for format information. + """ + for line in generate_pajek(G): + line += "\n" + path.write(line.encode(encoding)) + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_pajek(path, encoding="UTF-8"): + """Read graph in Pajek format from path. + + Parameters + ---------- + path : file or string + File or filename to write. + Filenames ending in .gz or .bz2 will be uncompressed. + + Returns + ------- + G : NetworkX MultiGraph or MultiDiGraph. + + Examples + -------- + >>> G = nx.path_graph(4) + >>> nx.write_pajek(G, "test.net") + >>> G = nx.read_pajek("test.net") + + To create a Graph instead of a MultiGraph use + + >>> G1 = nx.Graph(G) + + References + ---------- + See http://vlado.fmf.uni-lj.si/pub/networks/pajek/doc/draweps.htm + for format information. + """ + lines = (line.decode(encoding) for line in path) + return parse_pajek(lines) + + +@nx._dispatchable(graphs=None, returns_graph=True) +def parse_pajek(lines): + """Parse Pajek format graph from string or iterable. + + Parameters + ---------- + lines : string or iterable + Data in Pajek format. + + Returns + ------- + G : NetworkX graph + + See Also + -------- + read_pajek + + """ + import shlex + + # multigraph=False + if isinstance(lines, str): + lines = iter(lines.split("\n")) + lines = iter([line.rstrip("\n") for line in lines]) + G = nx.MultiDiGraph() # are multiedges allowed in Pajek? assume yes + labels = [] # in the order of the file, needed for matrix + while lines: + try: + l = next(lines) + except: # EOF + break + if l.lower().startswith("*network"): + try: + label, name = l.split(None, 1) + except ValueError: + # Line was not of the form: *network NAME + pass + else: + G.graph["name"] = name + elif l.lower().startswith("*vertices"): + nodelabels = {} + l, nnodes = l.split() + for i in range(int(nnodes)): + l = next(lines) + try: + splitline = [ + x.decode("utf-8") for x in shlex.split(str(l).encode("utf-8")) + ] + except AttributeError: + splitline = shlex.split(str(l)) + id, label = splitline[0:2] + labels.append(label) + G.add_node(label) + nodelabels[id] = label + G.nodes[label]["id"] = id + try: + x, y, shape = splitline[2:5] + G.nodes[label].update( + {"x": float(x), "y": float(y), "shape": shape} + ) + except: + pass + extra_attr = zip(splitline[5::2], splitline[6::2]) + G.nodes[label].update(extra_attr) + elif l.lower().startswith("*edges") or l.lower().startswith("*arcs"): + if l.lower().startswith("*edge"): + # switch from multidigraph to multigraph + G = nx.MultiGraph(G) + if l.lower().startswith("*arcs"): + # switch to directed with multiple arcs for each existing edge + G = G.to_directed() + for l in lines: + try: + splitline = [ + x.decode("utf-8") for x in shlex.split(str(l).encode("utf-8")) + ] + except AttributeError: + splitline = shlex.split(str(l)) + + if len(splitline) < 2: + continue + ui, vi = splitline[0:2] + u = nodelabels.get(ui, ui) + v = nodelabels.get(vi, vi) + # parse the data attached to this edge and put in a dictionary + edge_data = {} + try: + # there should always be a single value on the edge? + w = splitline[2:3] + edge_data.update({"weight": float(w[0])}) + except: + pass + # if there isn't, just assign a 1 + # edge_data.update({'value':1}) + extra_attr = zip(splitline[3::2], splitline[4::2]) + edge_data.update(extra_attr) + # if G.has_edge(u,v): + # multigraph=True + G.add_edge(u, v, **edge_data) + elif l.lower().startswith("*matrix"): + G = nx.DiGraph(G) + adj_list = ( + (labels[row], labels[col], {"weight": int(data)}) + for (row, line) in enumerate(lines) + for (col, data) in enumerate(line.split()) + if int(data) != 0 + ) + G.add_edges_from(adj_list) + + return G + + +def make_qstr(t): + """Returns the string representation of t. + Add outer double-quotes if the string has a space. + """ + if not isinstance(t, str): + t = str(t) + if " " in t: + t = f'"{t}"' + return t diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/sparse6.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/sparse6.py new file mode 100644 index 00000000..74d16dbc --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/sparse6.py @@ -0,0 +1,377 @@ +# Original author: D. Eppstein, UC Irvine, August 12, 2003. +# The original code at https://www.ics.uci.edu/~eppstein/PADS/ is public domain. +"""Functions for reading and writing graphs in the *sparse6* format. + +The *sparse6* file format is a space-efficient format for large sparse +graphs. For small graphs or large dense graphs, use the *graph6* file +format. + +For more information, see the `sparse6`_ homepage. + +.. _sparse6: https://users.cecs.anu.edu.au/~bdm/data/formats.html + +""" + +import networkx as nx +from networkx.exception import NetworkXError +from networkx.readwrite.graph6 import data_to_n, n_to_data +from networkx.utils import not_implemented_for, open_file + +__all__ = ["from_sparse6_bytes", "read_sparse6", "to_sparse6_bytes", "write_sparse6"] + + +def _generate_sparse6_bytes(G, nodes, header): + """Yield bytes in the sparse6 encoding of a graph. + + `G` is an undirected simple graph. `nodes` is the list of nodes for + which the node-induced subgraph will be encoded; if `nodes` is the + list of all nodes in the graph, the entire graph will be + encoded. `header` is a Boolean that specifies whether to generate + the header ``b'>>sparse6<<'`` before the remaining data. + + This function generates `bytes` objects in the following order: + + 1. the header (if requested), + 2. the encoding of the number of nodes, + 3. each character, one-at-a-time, in the encoding of the requested + node-induced subgraph, + 4. a newline character. + + This function raises :exc:`ValueError` if the graph is too large for + the graph6 format (that is, greater than ``2 ** 36`` nodes). + + """ + n = len(G) + if n >= 2**36: + raise ValueError( + "sparse6 is only defined if number of nodes is less than 2 ** 36" + ) + if header: + yield b">>sparse6<<" + yield b":" + for d in n_to_data(n): + yield str.encode(chr(d + 63)) + + k = 1 + while 1 << k < n: + k += 1 + + def enc(x): + """Big endian k-bit encoding of x""" + return [1 if (x & 1 << (k - 1 - i)) else 0 for i in range(k)] + + edges = sorted((max(u, v), min(u, v)) for u, v in G.edges()) + bits = [] + curv = 0 + for v, u in edges: + if v == curv: # current vertex edge + bits.append(0) + bits.extend(enc(u)) + elif v == curv + 1: # next vertex edge + curv += 1 + bits.append(1) + bits.extend(enc(u)) + else: # skip to vertex v and then add edge to u + curv = v + bits.append(1) + bits.extend(enc(v)) + bits.append(0) + bits.extend(enc(u)) + if k < 6 and n == (1 << k) and ((-len(bits)) % 6) >= k and curv < (n - 1): + # Padding special case: small k, n=2^k, + # more than k bits of padding needed, + # current vertex is not (n-1) -- + # appending 1111... would add a loop on (n-1) + bits.append(0) + bits.extend([1] * ((-len(bits)) % 6)) + else: + bits.extend([1] * ((-len(bits)) % 6)) + + data = [ + (bits[i + 0] << 5) + + (bits[i + 1] << 4) + + (bits[i + 2] << 3) + + (bits[i + 3] << 2) + + (bits[i + 4] << 1) + + (bits[i + 5] << 0) + for i in range(0, len(bits), 6) + ] + + for d in data: + yield str.encode(chr(d + 63)) + yield b"\n" + + +@nx._dispatchable(graphs=None, returns_graph=True) +def from_sparse6_bytes(string): + """Read an undirected graph in sparse6 format from string. + + Parameters + ---------- + string : string + Data in sparse6 format + + Returns + ------- + G : Graph + + Raises + ------ + NetworkXError + If the string is unable to be parsed in sparse6 format + + Examples + -------- + >>> G = nx.from_sparse6_bytes(b":A_") + >>> sorted(G.edges()) + [(0, 1), (0, 1), (0, 1)] + + See Also + -------- + read_sparse6, write_sparse6 + + References + ---------- + .. [1] Sparse6 specification + <https://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + if string.startswith(b">>sparse6<<"): + string = string[11:] + if not string.startswith(b":"): + raise NetworkXError("Expected leading colon in sparse6") + + chars = [c - 63 for c in string[1:]] + n, data = data_to_n(chars) + k = 1 + while 1 << k < n: + k += 1 + + def parseData(): + """Returns stream of pairs b[i], x[i] for sparse6 format.""" + chunks = iter(data) + d = None # partial data word + dLen = 0 # how many unparsed bits are left in d + + while 1: + if dLen < 1: + try: + d = next(chunks) + except StopIteration: + return + dLen = 6 + dLen -= 1 + b = (d >> dLen) & 1 # grab top remaining bit + + x = d & ((1 << dLen) - 1) # partially built up value of x + xLen = dLen # how many bits included so far in x + while xLen < k: # now grab full chunks until we have enough + try: + d = next(chunks) + except StopIteration: + return + dLen = 6 + x = (x << 6) + d + xLen += 6 + x = x >> (xLen - k) # shift back the extra bits + dLen = xLen - k + yield b, x + + v = 0 + + G = nx.MultiGraph() + G.add_nodes_from(range(n)) + + multigraph = False + for b, x in parseData(): + if b == 1: + v += 1 + # padding with ones can cause overlarge number here + if x >= n or v >= n: + break + elif x > v: + v = x + else: + if G.has_edge(x, v): + multigraph = True + G.add_edge(x, v) + if not multigraph: + G = nx.Graph(G) + return G + + +def to_sparse6_bytes(G, nodes=None, header=True): + """Convert an undirected graph to bytes in sparse6 format. + + Parameters + ---------- + G : Graph (undirected) + + nodes: list or iterable + Nodes are labeled 0...n-1 in the order provided. If None the ordering + given by ``G.nodes()`` is used. + + header: bool + If True add '>>sparse6<<' bytes to head of data. + + Raises + ------ + NetworkXNotImplemented + If the graph is directed. + + ValueError + If the graph has at least ``2 ** 36`` nodes; the sparse6 format + is only defined for graphs of order less than ``2 ** 36``. + + Examples + -------- + >>> nx.to_sparse6_bytes(nx.path_graph(2)) + b'>>sparse6<<:An\\n' + + See Also + -------- + to_sparse6_bytes, read_sparse6, write_sparse6_bytes + + Notes + ----- + The returned bytes end with a newline character. + + The format does not support edge or node labels. + + References + ---------- + .. [1] Graph6 specification + <https://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + if nodes is not None: + G = G.subgraph(nodes) + G = nx.convert_node_labels_to_integers(G, ordering="sorted") + return b"".join(_generate_sparse6_bytes(G, nodes, header)) + + +@open_file(0, mode="rb") +@nx._dispatchable(graphs=None, returns_graph=True) +def read_sparse6(path): + """Read an undirected graph in sparse6 format from path. + + Parameters + ---------- + path : file or string + File or filename to write. + + Returns + ------- + G : Graph/Multigraph or list of Graphs/MultiGraphs + If the file contains multiple lines then a list of graphs is returned + + Raises + ------ + NetworkXError + If the string is unable to be parsed in sparse6 format + + Examples + -------- + You can read a sparse6 file by giving the path to the file:: + + >>> import tempfile + >>> with tempfile.NamedTemporaryFile(delete=False) as f: + ... _ = f.write(b">>sparse6<<:An\\n") + ... _ = f.seek(0) + ... G = nx.read_sparse6(f.name) + >>> list(G.edges()) + [(0, 1)] + + You can also read a sparse6 file by giving an open file-like object:: + + >>> import tempfile + >>> with tempfile.NamedTemporaryFile() as f: + ... _ = f.write(b">>sparse6<<:An\\n") + ... _ = f.seek(0) + ... G = nx.read_sparse6(f) + >>> list(G.edges()) + [(0, 1)] + + See Also + -------- + read_sparse6, from_sparse6_bytes + + References + ---------- + .. [1] Sparse6 specification + <https://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + glist = [] + for line in path: + line = line.strip() + if not len(line): + continue + glist.append(from_sparse6_bytes(line)) + if len(glist) == 1: + return glist[0] + else: + return glist + + +@not_implemented_for("directed") +@open_file(1, mode="wb") +def write_sparse6(G, path, nodes=None, header=True): + """Write graph G to given path in sparse6 format. + + Parameters + ---------- + G : Graph (undirected) + + path : file or string + File or filename to write + + nodes: list or iterable + Nodes are labeled 0...n-1 in the order provided. If None the ordering + given by G.nodes() is used. + + header: bool + If True add '>>sparse6<<' string to head of data + + Raises + ------ + NetworkXError + If the graph is directed + + Examples + -------- + You can write a sparse6 file by giving the path to the file:: + + >>> import tempfile + >>> with tempfile.NamedTemporaryFile(delete=False) as f: + ... nx.write_sparse6(nx.path_graph(2), f.name) + ... print(f.read()) + b'>>sparse6<<:An\\n' + + You can also write a sparse6 file by giving an open file-like object:: + + >>> with tempfile.NamedTemporaryFile() as f: + ... nx.write_sparse6(nx.path_graph(2), f) + ... _ = f.seek(0) + ... print(f.read()) + b'>>sparse6<<:An\\n' + + See Also + -------- + read_sparse6, from_sparse6_bytes + + Notes + ----- + The format does not support edge or node labels. + + References + ---------- + .. [1] Sparse6 specification + <https://users.cecs.anu.edu.au/~bdm/data/formats.html> + + """ + if nodes is not None: + G = G.subgraph(nodes) + G = nx.convert_node_labels_to_integers(G, ordering="sorted") + for b in _generate_sparse6_bytes(G, nodes, header): + path.write(b) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/__init__.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/__init__.py new file mode 100644 index 00000000..e69de29b --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/__init__.py diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_adjlist.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_adjlist.py new file mode 100644 index 00000000..f2218eba --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_adjlist.py @@ -0,0 +1,262 @@ +""" +Unit tests for adjlist. +""" + +import io + +import pytest + +import networkx as nx +from networkx.utils import edges_equal, graphs_equal, nodes_equal + + +class TestAdjlist: + @classmethod + def setup_class(cls): + cls.G = nx.Graph(name="test") + e = [("a", "b"), ("b", "c"), ("c", "d"), ("d", "e"), ("e", "f"), ("a", "f")] + cls.G.add_edges_from(e) + cls.G.add_node("g") + cls.DG = nx.DiGraph(cls.G) + cls.XG = nx.MultiGraph() + cls.XG.add_weighted_edges_from([(1, 2, 5), (1, 2, 5), (1, 2, 1), (3, 3, 42)]) + cls.XDG = nx.MultiDiGraph(cls.XG) + + def test_read_multiline_adjlist_1(self): + # Unit test for https://networkx.lanl.gov/trac/ticket/252 + s = b"""# comment line +1 2 +# comment line +2 +3 +""" + bytesIO = io.BytesIO(s) + G = nx.read_multiline_adjlist(bytesIO) + adj = {"1": {"3": {}, "2": {}}, "3": {"1": {}}, "2": {"1": {}}} + assert graphs_equal(G, nx.Graph(adj)) + + def test_unicode(self, tmp_path): + G = nx.Graph() + name1 = chr(2344) + chr(123) + chr(6543) + name2 = chr(5543) + chr(1543) + chr(324) + G.add_edge(name1, "Radiohead", **{name2: 3}) + + fname = tmp_path / "adjlist.txt" + nx.write_multiline_adjlist(G, fname) + H = nx.read_multiline_adjlist(fname) + assert graphs_equal(G, H) + + def test_latin1_err(self, tmp_path): + G = nx.Graph() + name1 = chr(2344) + chr(123) + chr(6543) + name2 = chr(5543) + chr(1543) + chr(324) + G.add_edge(name1, "Radiohead", **{name2: 3}) + fname = tmp_path / "adjlist.txt" + with pytest.raises(UnicodeEncodeError): + nx.write_multiline_adjlist(G, fname, encoding="latin-1") + + def test_latin1(self, tmp_path): + G = nx.Graph() + name1 = "Bj" + chr(246) + "rk" + name2 = chr(220) + "ber" + G.add_edge(name1, "Radiohead", **{name2: 3}) + fname = tmp_path / "adjlist.txt" + nx.write_multiline_adjlist(G, fname, encoding="latin-1") + H = nx.read_multiline_adjlist(fname, encoding="latin-1") + assert graphs_equal(G, H) + + def test_parse_adjlist(self): + lines = ["1 2 5", "2 3 4", "3 5", "4", "5"] + nx.parse_adjlist(lines, nodetype=int) # smoke test + with pytest.raises(TypeError): + nx.parse_adjlist(lines, nodetype="int") + lines = ["1 2 5", "2 b", "c"] + with pytest.raises(TypeError): + nx.parse_adjlist(lines, nodetype=int) + + def test_adjlist_graph(self, tmp_path): + G = self.G + fname = tmp_path / "adjlist.txt" + nx.write_adjlist(G, fname) + H = nx.read_adjlist(fname) + H2 = nx.read_adjlist(fname) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_adjlist_digraph(self, tmp_path): + G = self.DG + fname = tmp_path / "adjlist.txt" + nx.write_adjlist(G, fname) + H = nx.read_adjlist(fname, create_using=nx.DiGraph()) + H2 = nx.read_adjlist(fname, create_using=nx.DiGraph()) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_adjlist_integers(self, tmp_path): + fname = tmp_path / "adjlist.txt" + G = nx.convert_node_labels_to_integers(self.G) + nx.write_adjlist(G, fname) + H = nx.read_adjlist(fname, nodetype=int) + H2 = nx.read_adjlist(fname, nodetype=int) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_adjlist_multigraph(self, tmp_path): + G = self.XG + fname = tmp_path / "adjlist.txt" + nx.write_adjlist(G, fname) + H = nx.read_adjlist(fname, nodetype=int, create_using=nx.MultiGraph()) + H2 = nx.read_adjlist(fname, nodetype=int, create_using=nx.MultiGraph()) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_adjlist_multidigraph(self, tmp_path): + G = self.XDG + fname = tmp_path / "adjlist.txt" + nx.write_adjlist(G, fname) + H = nx.read_adjlist(fname, nodetype=int, create_using=nx.MultiDiGraph()) + H2 = nx.read_adjlist(fname, nodetype=int, create_using=nx.MultiDiGraph()) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_adjlist_delimiter(self): + fh = io.BytesIO() + G = nx.path_graph(3) + nx.write_adjlist(G, fh, delimiter=":") + fh.seek(0) + H = nx.read_adjlist(fh, nodetype=int, delimiter=":") + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + +class TestMultilineAdjlist: + @classmethod + def setup_class(cls): + cls.G = nx.Graph(name="test") + e = [("a", "b"), ("b", "c"), ("c", "d"), ("d", "e"), ("e", "f"), ("a", "f")] + cls.G.add_edges_from(e) + cls.G.add_node("g") + cls.DG = nx.DiGraph(cls.G) + cls.DG.remove_edge("b", "a") + cls.DG.remove_edge("b", "c") + cls.XG = nx.MultiGraph() + cls.XG.add_weighted_edges_from([(1, 2, 5), (1, 2, 5), (1, 2, 1), (3, 3, 42)]) + cls.XDG = nx.MultiDiGraph(cls.XG) + + def test_parse_multiline_adjlist(self): + lines = [ + "1 2", + "b {'weight':3, 'name': 'Frodo'}", + "c {}", + "d 1", + "e {'weight':6, 'name': 'Saruman'}", + ] + nx.parse_multiline_adjlist(iter(lines)) # smoke test + with pytest.raises(TypeError): + nx.parse_multiline_adjlist(iter(lines), nodetype=int) + nx.parse_multiline_adjlist(iter(lines), edgetype=str) # smoke test + with pytest.raises(TypeError): + nx.parse_multiline_adjlist(iter(lines), nodetype=int) + lines = ["1 a"] + with pytest.raises(TypeError): + nx.parse_multiline_adjlist(iter(lines)) + lines = ["a 2"] + with pytest.raises(TypeError): + nx.parse_multiline_adjlist(iter(lines), nodetype=int) + lines = ["1 2"] + with pytest.raises(TypeError): + nx.parse_multiline_adjlist(iter(lines)) + lines = ["1 2", "2 {}"] + with pytest.raises(TypeError): + nx.parse_multiline_adjlist(iter(lines)) + + def test_multiline_adjlist_graph(self, tmp_path): + G = self.G + fname = tmp_path / "adjlist.txt" + nx.write_multiline_adjlist(G, fname) + H = nx.read_multiline_adjlist(fname) + H2 = nx.read_multiline_adjlist(fname) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_multiline_adjlist_digraph(self, tmp_path): + G = self.DG + fname = tmp_path / "adjlist.txt" + nx.write_multiline_adjlist(G, fname) + H = nx.read_multiline_adjlist(fname, create_using=nx.DiGraph()) + H2 = nx.read_multiline_adjlist(fname, create_using=nx.DiGraph()) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_multiline_adjlist_integers(self, tmp_path): + fname = tmp_path / "adjlist.txt" + G = nx.convert_node_labels_to_integers(self.G) + nx.write_multiline_adjlist(G, fname) + H = nx.read_multiline_adjlist(fname, nodetype=int) + H2 = nx.read_multiline_adjlist(fname, nodetype=int) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_multiline_adjlist_multigraph(self, tmp_path): + G = self.XG + fname = tmp_path / "adjlist.txt" + nx.write_multiline_adjlist(G, fname) + H = nx.read_multiline_adjlist(fname, nodetype=int, create_using=nx.MultiGraph()) + H2 = nx.read_multiline_adjlist( + fname, nodetype=int, create_using=nx.MultiGraph() + ) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_multiline_adjlist_multidigraph(self, tmp_path): + G = self.XDG + fname = tmp_path / "adjlist.txt" + nx.write_multiline_adjlist(G, fname) + H = nx.read_multiline_adjlist( + fname, nodetype=int, create_using=nx.MultiDiGraph() + ) + H2 = nx.read_multiline_adjlist( + fname, nodetype=int, create_using=nx.MultiDiGraph() + ) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_multiline_adjlist_delimiter(self): + fh = io.BytesIO() + G = nx.path_graph(3) + nx.write_multiline_adjlist(G, fh, delimiter=":") + fh.seek(0) + H = nx.read_multiline_adjlist(fh, nodetype=int, delimiter=":") + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + +@pytest.mark.parametrize( + ("lines", "delim"), + ( + (["1 2 5", "2 3 4", "3 5", "4", "5"], None), # No extra whitespace + (["1\t2\t5", "2\t3\t4", "3\t5", "4", "5"], "\t"), # tab-delimited + ( + ["1\t2\t5", "2\t3\t4", "3\t5\t", "4\t", "5"], + "\t", + ), # tab-delimited, extra delims + ( + ["1\t2\t5", "2\t3\t4", "3\t5\t\t\n", "4\t", "5"], + "\t", + ), # extra delim+newlines + ), +) +def test_adjlist_rstrip_parsing(lines, delim): + """Regression test related to gh-7465""" + expected = nx.Graph([(1, 2), (1, 5), (2, 3), (2, 4), (3, 5)]) + nx.utils.graphs_equal(nx.parse_adjlist(lines, delimiter=delim), expected) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_edgelist.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_edgelist.py new file mode 100644 index 00000000..fe58b3b7 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_edgelist.py @@ -0,0 +1,314 @@ +""" +Unit tests for edgelists. +""" + +import io +import textwrap + +import pytest + +import networkx as nx +from networkx.utils import edges_equal, graphs_equal, nodes_equal + +edges_no_data = textwrap.dedent( + """ + # comment line + 1 2 + # comment line + 2 3 + """ +) + + +edges_with_values = textwrap.dedent( + """ + # comment line + 1 2 2.0 + # comment line + 2 3 3.0 + """ +) + + +edges_with_weight = textwrap.dedent( + """ + # comment line + 1 2 {'weight':2.0} + # comment line + 2 3 {'weight':3.0} + """ +) + + +edges_with_multiple_attrs = textwrap.dedent( + """ + # comment line + 1 2 {'weight':2.0, 'color':'green'} + # comment line + 2 3 {'weight':3.0, 'color':'red'} + """ +) + + +edges_with_multiple_attrs_csv = textwrap.dedent( + """ + # comment line + 1, 2, {'weight':2.0, 'color':'green'} + # comment line + 2, 3, {'weight':3.0, 'color':'red'} + """ +) + + +_expected_edges_weights = [(1, 2, {"weight": 2.0}), (2, 3, {"weight": 3.0})] +_expected_edges_multiattr = [ + (1, 2, {"weight": 2.0, "color": "green"}), + (2, 3, {"weight": 3.0, "color": "red"}), +] + + +@pytest.mark.parametrize( + ("data", "extra_kwargs"), + ( + (edges_no_data, {}), + (edges_with_values, {}), + (edges_with_weight, {}), + (edges_with_multiple_attrs, {}), + (edges_with_multiple_attrs_csv, {"delimiter": ","}), + ), +) +def test_read_edgelist_no_data(data, extra_kwargs): + bytesIO = io.BytesIO(data.encode("utf-8")) + G = nx.read_edgelist(bytesIO, nodetype=int, data=False, **extra_kwargs) + assert edges_equal(G.edges(), [(1, 2), (2, 3)]) + + +def test_read_weighted_edgelist(): + bytesIO = io.BytesIO(edges_with_values.encode("utf-8")) + G = nx.read_weighted_edgelist(bytesIO, nodetype=int) + assert edges_equal(G.edges(data=True), _expected_edges_weights) + + +@pytest.mark.parametrize( + ("data", "extra_kwargs", "expected"), + ( + (edges_with_weight, {}, _expected_edges_weights), + (edges_with_multiple_attrs, {}, _expected_edges_multiattr), + (edges_with_multiple_attrs_csv, {"delimiter": ","}, _expected_edges_multiattr), + ), +) +def test_read_edgelist_with_data(data, extra_kwargs, expected): + bytesIO = io.BytesIO(data.encode("utf-8")) + G = nx.read_edgelist(bytesIO, nodetype=int, **extra_kwargs) + assert edges_equal(G.edges(data=True), expected) + + +@pytest.fixture +def example_graph(): + G = nx.Graph() + G.add_weighted_edges_from([(1, 2, 3.0), (2, 3, 27.0), (3, 4, 3.0)]) + return G + + +def test_parse_edgelist_no_data(example_graph): + G = example_graph + H = nx.parse_edgelist(["1 2", "2 3", "3 4"], nodetype=int) + assert nodes_equal(G.nodes, H.nodes) + assert edges_equal(G.edges, H.edges) + + +def test_parse_edgelist_with_data_dict(example_graph): + G = example_graph + H = nx.parse_edgelist( + ["1 2 {'weight': 3}", "2 3 {'weight': 27}", "3 4 {'weight': 3.0}"], nodetype=int + ) + assert nodes_equal(G.nodes, H.nodes) + assert edges_equal(G.edges(data=True), H.edges(data=True)) + + +def test_parse_edgelist_with_data_list(example_graph): + G = example_graph + H = nx.parse_edgelist( + ["1 2 3", "2 3 27", "3 4 3.0"], nodetype=int, data=(("weight", float),) + ) + assert nodes_equal(G.nodes, H.nodes) + assert edges_equal(G.edges(data=True), H.edges(data=True)) + + +def test_parse_edgelist(): + # ignore lines with less than 2 nodes + lines = ["1;2", "2 3", "3 4"] + G = nx.parse_edgelist(lines, nodetype=int) + assert list(G.edges()) == [(2, 3), (3, 4)] + # unknown nodetype + with pytest.raises(TypeError, match="Failed to convert nodes"): + lines = ["1 2", "2 3", "3 4"] + nx.parse_edgelist(lines, nodetype="nope") + # lines have invalid edge format + with pytest.raises(TypeError, match="Failed to convert edge data"): + lines = ["1 2 3", "2 3", "3 4"] + nx.parse_edgelist(lines, nodetype=int) + # edge data and data_keys not the same length + with pytest.raises(IndexError, match="not the same length"): + lines = ["1 2 3", "2 3 27", "3 4 3.0"] + nx.parse_edgelist( + lines, nodetype=int, data=(("weight", float), ("capacity", int)) + ) + # edge data can't be converted to edge type + with pytest.raises(TypeError, match="Failed to convert"): + lines = ["1 2 't1'", "2 3 't3'", "3 4 't3'"] + nx.parse_edgelist(lines, nodetype=int, data=(("weight", float),)) + + +def test_comments_None(): + edgelist = ["node#1 node#2", "node#2 node#3"] + # comments=None supported to ignore all comment characters + G = nx.parse_edgelist(edgelist, comments=None) + H = nx.Graph([e.split(" ") for e in edgelist]) + assert edges_equal(G.edges, H.edges) + + +class TestEdgelist: + @classmethod + def setup_class(cls): + cls.G = nx.Graph(name="test") + e = [("a", "b"), ("b", "c"), ("c", "d"), ("d", "e"), ("e", "f"), ("a", "f")] + cls.G.add_edges_from(e) + cls.G.add_node("g") + cls.DG = nx.DiGraph(cls.G) + cls.XG = nx.MultiGraph() + cls.XG.add_weighted_edges_from([(1, 2, 5), (1, 2, 5), (1, 2, 1), (3, 3, 42)]) + cls.XDG = nx.MultiDiGraph(cls.XG) + + def test_write_edgelist_1(self): + fh = io.BytesIO() + G = nx.Graph() + G.add_edges_from([(1, 2), (2, 3)]) + nx.write_edgelist(G, fh, data=False) + fh.seek(0) + assert fh.read() == b"1 2\n2 3\n" + + def test_write_edgelist_2(self): + fh = io.BytesIO() + G = nx.Graph() + G.add_edges_from([(1, 2), (2, 3)]) + nx.write_edgelist(G, fh, data=True) + fh.seek(0) + assert fh.read() == b"1 2 {}\n2 3 {}\n" + + def test_write_edgelist_3(self): + fh = io.BytesIO() + G = nx.Graph() + G.add_edge(1, 2, weight=2.0) + G.add_edge(2, 3, weight=3.0) + nx.write_edgelist(G, fh, data=True) + fh.seek(0) + assert fh.read() == b"1 2 {'weight': 2.0}\n2 3 {'weight': 3.0}\n" + + def test_write_edgelist_4(self): + fh = io.BytesIO() + G = nx.Graph() + G.add_edge(1, 2, weight=2.0) + G.add_edge(2, 3, weight=3.0) + nx.write_edgelist(G, fh, data=[("weight")]) + fh.seek(0) + assert fh.read() == b"1 2 2.0\n2 3 3.0\n" + + def test_unicode(self, tmp_path): + G = nx.Graph() + name1 = chr(2344) + chr(123) + chr(6543) + name2 = chr(5543) + chr(1543) + chr(324) + G.add_edge(name1, "Radiohead", **{name2: 3}) + fname = tmp_path / "el.txt" + nx.write_edgelist(G, fname) + H = nx.read_edgelist(fname) + assert graphs_equal(G, H) + + def test_latin1_issue(self, tmp_path): + G = nx.Graph() + name1 = chr(2344) + chr(123) + chr(6543) + name2 = chr(5543) + chr(1543) + chr(324) + G.add_edge(name1, "Radiohead", **{name2: 3}) + fname = tmp_path / "el.txt" + with pytest.raises(UnicodeEncodeError): + nx.write_edgelist(G, fname, encoding="latin-1") + + def test_latin1(self, tmp_path): + G = nx.Graph() + name1 = "Bj" + chr(246) + "rk" + name2 = chr(220) + "ber" + G.add_edge(name1, "Radiohead", **{name2: 3}) + fname = tmp_path / "el.txt" + + nx.write_edgelist(G, fname, encoding="latin-1") + H = nx.read_edgelist(fname, encoding="latin-1") + assert graphs_equal(G, H) + + def test_edgelist_graph(self, tmp_path): + G = self.G + fname = tmp_path / "el.txt" + nx.write_edgelist(G, fname) + H = nx.read_edgelist(fname) + H2 = nx.read_edgelist(fname) + assert H is not H2 # they should be different graphs + G.remove_node("g") # isolated nodes are not written in edgelist + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_edgelist_digraph(self, tmp_path): + G = self.DG + fname = tmp_path / "el.txt" + nx.write_edgelist(G, fname) + H = nx.read_edgelist(fname, create_using=nx.DiGraph()) + H2 = nx.read_edgelist(fname, create_using=nx.DiGraph()) + assert H is not H2 # they should be different graphs + G.remove_node("g") # isolated nodes are not written in edgelist + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_edgelist_integers(self, tmp_path): + G = nx.convert_node_labels_to_integers(self.G) + fname = tmp_path / "el.txt" + nx.write_edgelist(G, fname) + H = nx.read_edgelist(fname, nodetype=int) + # isolated nodes are not written in edgelist + G.remove_nodes_from(list(nx.isolates(G))) + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_edgelist_multigraph(self, tmp_path): + G = self.XG + fname = tmp_path / "el.txt" + nx.write_edgelist(G, fname) + H = nx.read_edgelist(fname, nodetype=int, create_using=nx.MultiGraph()) + H2 = nx.read_edgelist(fname, nodetype=int, create_using=nx.MultiGraph()) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + def test_edgelist_multidigraph(self, tmp_path): + G = self.XDG + fname = tmp_path / "el.txt" + nx.write_edgelist(G, fname) + H = nx.read_edgelist(fname, nodetype=int, create_using=nx.MultiDiGraph()) + H2 = nx.read_edgelist(fname, nodetype=int, create_using=nx.MultiDiGraph()) + assert H is not H2 # they should be different graphs + assert nodes_equal(list(H), list(G)) + assert edges_equal(list(H.edges()), list(G.edges())) + + +def test_edgelist_consistent_strip_handling(): + """See gh-7462 + + Input when printed looks like:: + + 1 2 3 + 2 3 + 3 4 3.0 + + Note the trailing \\t after the `3` in the second row, indicating an empty + data value. + """ + s = io.StringIO("1\t2\t3\n2\t3\t\n3\t4\t3.0") + G = nx.parse_edgelist(s, delimiter="\t", nodetype=int, data=[("value", str)]) + assert sorted(G.edges(data="value")) == [(1, 2, "3"), (2, 3, ""), (3, 4, "3.0")] diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_gexf.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_gexf.py new file mode 100644 index 00000000..6ff14c99 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_gexf.py @@ -0,0 +1,557 @@ +import io +import time + +import pytest + +import networkx as nx + + +class TestGEXF: + @classmethod + def setup_class(cls): + cls.simple_directed_data = """<?xml version="1.0" encoding="UTF-8"?> +<gexf xmlns="http://www.gexf.net/1.2draft" version="1.2"> + <graph mode="static" defaultedgetype="directed"> + <nodes> + <node id="0" label="Hello" /> + <node id="1" label="Word" /> + </nodes> + <edges> + <edge id="0" source="0" target="1" /> + </edges> + </graph> +</gexf> +""" + cls.simple_directed_graph = nx.DiGraph() + cls.simple_directed_graph.add_node("0", label="Hello") + cls.simple_directed_graph.add_node("1", label="World") + cls.simple_directed_graph.add_edge("0", "1", id="0") + + cls.simple_directed_fh = io.BytesIO(cls.simple_directed_data.encode("UTF-8")) + + cls.attribute_data = """<?xml version="1.0" encoding="UTF-8"?>\ +<gexf xmlns="http://www.gexf.net/1.2draft" xmlns:xsi="http://www.w3.\ +org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.gexf.net/\ +1.2draft http://www.gexf.net/1.2draft/gexf.xsd" version="1.2"> + <meta lastmodifieddate="2009-03-20"> + <creator>Gephi.org</creator> + <description>A Web network</description> + </meta> + <graph defaultedgetype="directed"> + <attributes class="node"> + <attribute id="0" title="url" type="string"/> + <attribute id="1" title="indegree" type="integer"/> + <attribute id="2" title="frog" type="boolean"> + <default>true</default> + </attribute> + </attributes> + <nodes> + <node id="0" label="Gephi"> + <attvalues> + <attvalue for="0" value="https://gephi.org"/> + <attvalue for="1" value="1"/> + <attvalue for="2" value="false"/> + </attvalues> + </node> + <node id="1" label="Webatlas"> + <attvalues> + <attvalue for="0" value="http://webatlas.fr"/> + <attvalue for="1" value="2"/> + <attvalue for="2" value="false"/> + </attvalues> + </node> + <node id="2" label="RTGI"> + <attvalues> + <attvalue for="0" value="http://rtgi.fr"/> + <attvalue for="1" value="1"/> + <attvalue for="2" value="true"/> + </attvalues> + </node> + <node id="3" label="BarabasiLab"> + <attvalues> + <attvalue for="0" value="http://barabasilab.com"/> + <attvalue for="1" value="1"/> + <attvalue for="2" value="true"/> + </attvalues> + </node> + </nodes> + <edges> + <edge id="0" source="0" target="1" label="foo"/> + <edge id="1" source="0" target="2"/> + <edge id="2" source="1" target="0"/> + <edge id="3" source="2" target="1"/> + <edge id="4" source="0" target="3"/> + </edges> + </graph> +</gexf> +""" + cls.attribute_graph = nx.DiGraph() + cls.attribute_graph.graph["node_default"] = {"frog": True} + cls.attribute_graph.add_node( + "0", label="Gephi", url="https://gephi.org", indegree=1, frog=False + ) + cls.attribute_graph.add_node( + "1", label="Webatlas", url="http://webatlas.fr", indegree=2, frog=False + ) + cls.attribute_graph.add_node( + "2", label="RTGI", url="http://rtgi.fr", indegree=1, frog=True + ) + cls.attribute_graph.add_node( + "3", + label="BarabasiLab", + url="http://barabasilab.com", + indegree=1, + frog=True, + ) + cls.attribute_graph.add_edge("0", "1", id="0", label="foo") + cls.attribute_graph.add_edge("0", "2", id="1") + cls.attribute_graph.add_edge("1", "0", id="2") + cls.attribute_graph.add_edge("2", "1", id="3") + cls.attribute_graph.add_edge("0", "3", id="4") + cls.attribute_fh = io.BytesIO(cls.attribute_data.encode("UTF-8")) + + cls.simple_undirected_data = """<?xml version="1.0" encoding="UTF-8"?> +<gexf xmlns="http://www.gexf.net/1.2draft" version="1.2"> + <graph mode="static" defaultedgetype="undirected"> + <nodes> + <node id="0" label="Hello" /> + <node id="1" label="Word" /> + </nodes> + <edges> + <edge id="0" source="0" target="1" /> + </edges> + </graph> +</gexf> +""" + cls.simple_undirected_graph = nx.Graph() + cls.simple_undirected_graph.add_node("0", label="Hello") + cls.simple_undirected_graph.add_node("1", label="World") + cls.simple_undirected_graph.add_edge("0", "1", id="0") + + cls.simple_undirected_fh = io.BytesIO( + cls.simple_undirected_data.encode("UTF-8") + ) + + def test_read_simple_directed_graphml(self): + G = self.simple_directed_graph + H = nx.read_gexf(self.simple_directed_fh) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(G.edges()) == sorted(H.edges()) + assert sorted(G.edges(data=True)) == sorted(H.edges(data=True)) + self.simple_directed_fh.seek(0) + + def test_write_read_simple_directed_graphml(self): + G = self.simple_directed_graph + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(G.edges()) == sorted(H.edges()) + assert sorted(G.edges(data=True)) == sorted(H.edges(data=True)) + self.simple_directed_fh.seek(0) + + def test_read_simple_undirected_graphml(self): + G = self.simple_undirected_graph + H = nx.read_gexf(self.simple_undirected_fh) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + self.simple_undirected_fh.seek(0) + + def test_read_attribute_graphml(self): + G = self.attribute_graph + H = nx.read_gexf(self.attribute_fh) + assert sorted(G.nodes(True)) == sorted(H.nodes(data=True)) + ge = sorted(G.edges(data=True)) + he = sorted(H.edges(data=True)) + for a, b in zip(ge, he): + assert a == b + self.attribute_fh.seek(0) + + def test_directed_edge_in_undirected(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<gexf xmlns="http://www.gexf.net/1.2draft" version='1.2'> + <graph mode="static" defaultedgetype="undirected" name=""> + <nodes> + <node id="0" label="Hello" /> + <node id="1" label="Word" /> + </nodes> + <edges> + <edge id="0" source="0" target="1" type="directed"/> + </edges> + </graph> +</gexf> +""" + fh = io.BytesIO(s.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_gexf, fh) + + def test_undirected_edge_in_directed(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<gexf xmlns="http://www.gexf.net/1.2draft" version='1.2'> + <graph mode="static" defaultedgetype="directed" name=""> + <nodes> + <node id="0" label="Hello" /> + <node id="1" label="Word" /> + </nodes> + <edges> + <edge id="0" source="0" target="1" type="undirected"/> + </edges> + </graph> +</gexf> +""" + fh = io.BytesIO(s.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_gexf, fh) + + def test_key_raises(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<gexf xmlns="http://www.gexf.net/1.2draft" version='1.2'> + <graph mode="static" defaultedgetype="directed" name=""> + <nodes> + <node id="0" label="Hello"> + <attvalues> + <attvalue for='0' value='1'/> + </attvalues> + </node> + <node id="1" label="Word" /> + </nodes> + <edges> + <edge id="0" source="0" target="1" type="undirected"/> + </edges> + </graph> +</gexf> +""" + fh = io.BytesIO(s.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_gexf, fh) + + def test_relabel(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<gexf xmlns="http://www.gexf.net/1.2draft" version='1.2'> + <graph mode="static" defaultedgetype="directed" name=""> + <nodes> + <node id="0" label="Hello" /> + <node id="1" label="Word" /> + </nodes> + <edges> + <edge id="0" source="0" target="1"/> + </edges> + </graph> +</gexf> +""" + fh = io.BytesIO(s.encode("UTF-8")) + G = nx.read_gexf(fh, relabel=True) + assert sorted(G.nodes()) == ["Hello", "Word"] + + def test_default_attribute(self): + G = nx.Graph() + G.add_node(1, label="1", color="green") + nx.add_path(G, [0, 1, 2, 3]) + G.add_edge(1, 2, foo=3) + G.graph["node_default"] = {"color": "yellow"} + G.graph["edge_default"] = {"foo": 7} + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + # Reading a gexf graph always sets mode attribute to either + # 'static' or 'dynamic'. Remove the mode attribute from the + # read graph for the sake of comparing remaining attributes. + del H.graph["mode"] + assert G.graph == H.graph + + def test_serialize_ints_to_strings(self): + G = nx.Graph() + G.add_node(1, id=7, label=77) + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert list(H) == [7] + assert H.nodes[7]["label"] == "77" + + def test_write_with_node_attributes(self): + # Addresses #673. + G = nx.Graph() + G.add_edges_from([(0, 1), (1, 2), (2, 3)]) + for i in range(4): + G.nodes[i]["id"] = i + G.nodes[i]["label"] = i + G.nodes[i]["pid"] = i + G.nodes[i]["start"] = i + G.nodes[i]["end"] = i + 1 + + expected = f"""<gexf xmlns="http://www.gexf.net/1.2draft" xmlns:xsi\ +="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation=\ +"http://www.gexf.net/1.2draft http://www.gexf.net/1.2draft/\ +gexf.xsd" version="1.2"> + <meta lastmodifieddate="{time.strftime('%Y-%m-%d')}"> + <creator>NetworkX {nx.__version__}</creator> + </meta> + <graph defaultedgetype="undirected" mode="dynamic" name="" timeformat="long"> + <nodes> + <node id="0" label="0" pid="0" start="0" end="1" /> + <node id="1" label="1" pid="1" start="1" end="2" /> + <node id="2" label="2" pid="2" start="2" end="3" /> + <node id="3" label="3" pid="3" start="3" end="4" /> + </nodes> + <edges> + <edge source="0" target="1" id="0" /> + <edge source="1" target="2" id="1" /> + <edge source="2" target="3" id="2" /> + </edges> + </graph> +</gexf>""" + obtained = "\n".join(nx.generate_gexf(G)) + assert expected == obtained + + def test_edge_id_construct(self): + G = nx.Graph() + G.add_edges_from([(0, 1, {"id": 0}), (1, 2, {"id": 2}), (2, 3)]) + + expected = f"""<gexf xmlns="http://www.gexf.net/1.2draft" xmlns:xsi\ +="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.\ +gexf.net/1.2draft http://www.gexf.net/1.2draft/gexf.xsd" version="1.2"> + <meta lastmodifieddate="{time.strftime('%Y-%m-%d')}"> + <creator>NetworkX {nx.__version__}</creator> + </meta> + <graph defaultedgetype="undirected" mode="static" name=""> + <nodes> + <node id="0" label="0" /> + <node id="1" label="1" /> + <node id="2" label="2" /> + <node id="3" label="3" /> + </nodes> + <edges> + <edge source="0" target="1" id="0" /> + <edge source="1" target="2" id="2" /> + <edge source="2" target="3" id="1" /> + </edges> + </graph> +</gexf>""" + + obtained = "\n".join(nx.generate_gexf(G)) + assert expected == obtained + + def test_numpy_type(self): + np = pytest.importorskip("numpy") + G = nx.path_graph(4) + nx.set_node_attributes(G, {n: n for n in np.arange(4)}, "number") + G[0][1]["edge-number"] = np.float64(1.1) + + expected = f"""<gexf xmlns="http://www.gexf.net/1.2draft"\ + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation\ +="http://www.gexf.net/1.2draft http://www.gexf.net/1.2draft/gexf.xsd"\ + version="1.2"> + <meta lastmodifieddate="{time.strftime('%Y-%m-%d')}"> + <creator>NetworkX {nx.__version__}</creator> + </meta> + <graph defaultedgetype="undirected" mode="static" name=""> + <attributes mode="static" class="edge"> + <attribute id="1" title="edge-number" type="float" /> + </attributes> + <attributes mode="static" class="node"> + <attribute id="0" title="number" type="int" /> + </attributes> + <nodes> + <node id="0" label="0"> + <attvalues> + <attvalue for="0" value="0" /> + </attvalues> + </node> + <node id="1" label="1"> + <attvalues> + <attvalue for="0" value="1" /> + </attvalues> + </node> + <node id="2" label="2"> + <attvalues> + <attvalue for="0" value="2" /> + </attvalues> + </node> + <node id="3" label="3"> + <attvalues> + <attvalue for="0" value="3" /> + </attvalues> + </node> + </nodes> + <edges> + <edge source="0" target="1" id="0"> + <attvalues> + <attvalue for="1" value="1.1" /> + </attvalues> + </edge> + <edge source="1" target="2" id="1" /> + <edge source="2" target="3" id="2" /> + </edges> + </graph> +</gexf>""" + obtained = "\n".join(nx.generate_gexf(G)) + assert expected == obtained + + def test_bool(self): + G = nx.Graph() + G.add_node(1, testattr=True) + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert H.nodes[1]["testattr"] + + # Test for NaN, INF and -INF + def test_specials(self): + from math import isnan + + inf, nan = float("inf"), float("nan") + G = nx.Graph() + G.add_node(1, testattr=inf, strdata="inf", key="a") + G.add_node(2, testattr=nan, strdata="nan", key="b") + G.add_node(3, testattr=-inf, strdata="-inf", key="c") + + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + filetext = fh.read() + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + + assert b"INF" in filetext + assert b"NaN" in filetext + assert b"-INF" in filetext + + assert H.nodes[1]["testattr"] == inf + assert isnan(H.nodes[2]["testattr"]) + assert H.nodes[3]["testattr"] == -inf + + assert H.nodes[1]["strdata"] == "inf" + assert H.nodes[2]["strdata"] == "nan" + assert H.nodes[3]["strdata"] == "-inf" + + assert H.nodes[1]["networkx_key"] == "a" + assert H.nodes[2]["networkx_key"] == "b" + assert H.nodes[3]["networkx_key"] == "c" + + def test_simple_list(self): + G = nx.Graph() + list_value = [(1, 2, 3), (9, 1, 2)] + G.add_node(1, key=list_value) + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert H.nodes[1]["networkx_key"] == list_value + + def test_dynamic_mode(self): + G = nx.Graph() + G.add_node(1, label="1", color="green") + G.graph["mode"] = "dynamic" + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + + def test_multigraph_with_missing_attributes(self): + G = nx.MultiGraph() + G.add_node(0, label="1", color="green") + G.add_node(1, label="2", color="green") + G.add_edge(0, 1, id="0", weight=3, type="undirected", start=0, end=1) + G.add_edge(0, 1, id="1", label="foo", start=0, end=1) + G.add_edge(0, 1) + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + + def test_missing_viz_attributes(self): + G = nx.Graph() + G.add_node(0, label="1", color="green") + G.nodes[0]["viz"] = {"size": 54} + G.nodes[0]["viz"]["position"] = {"x": 0, "y": 1, "z": 0} + G.nodes[0]["viz"]["color"] = {"r": 0, "g": 0, "b": 256} + G.nodes[0]["viz"]["shape"] = "http://random.url" + G.nodes[0]["viz"]["thickness"] = 2 + fh = io.BytesIO() + nx.write_gexf(G, fh, version="1.1draft") + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + + # Test missing alpha value for version >draft1.1 - set default alpha value + # to 1.0 instead of `None` when writing for better general compatibility + fh = io.BytesIO() + # G.nodes[0]["viz"]["color"] does not have an alpha value explicitly defined + # so the default is used instead + nx.write_gexf(G, fh, version="1.2draft") + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert H.nodes[0]["viz"]["color"]["a"] == 1.0 + + # Second graph for the other branch + G = nx.Graph() + G.add_node(0, label="1", color="green") + G.nodes[0]["viz"] = {"size": 54} + G.nodes[0]["viz"]["position"] = {"x": 0, "y": 1, "z": 0} + G.nodes[0]["viz"]["color"] = {"r": 0, "g": 0, "b": 256, "a": 0.5} + G.nodes[0]["viz"]["shape"] = "ftp://random.url" + G.nodes[0]["viz"]["thickness"] = 2 + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + + def test_slice_and_spell(self): + # Test spell first, so version = 1.2 + G = nx.Graph() + G.add_node(0, label="1", color="green") + G.nodes[0]["spells"] = [(1, 2)] + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + + G = nx.Graph() + G.add_node(0, label="1", color="green") + G.nodes[0]["slices"] = [(1, 2)] + fh = io.BytesIO() + nx.write_gexf(G, fh, version="1.1draft") + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) + + def test_add_parent(self): + G = nx.Graph() + G.add_node(0, label="1", color="green", parents=[1, 2]) + fh = io.BytesIO() + nx.write_gexf(G, fh) + fh.seek(0) + H = nx.read_gexf(fh, node_type=int) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(sorted(e) for e in G.edges()) == sorted( + sorted(e) for e in H.edges() + ) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_gml.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_gml.py new file mode 100644 index 00000000..f575ad26 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_gml.py @@ -0,0 +1,744 @@ +import codecs +import io +import math +from ast import literal_eval +from contextlib import contextmanager +from textwrap import dedent + +import pytest + +import networkx as nx +from networkx.readwrite.gml import literal_destringizer, literal_stringizer + + +class TestGraph: + @classmethod + def setup_class(cls): + cls.simple_data = """Creator "me" +Version "xx" +graph [ + comment "This is a sample graph" + directed 1 + IsPlanar 1 + pos [ x 0 y 1 ] + node [ + id 1 + label "Node 1" + pos [ x 1 y 1 ] + ] + node [ + id 2 + pos [ x 1 y 2 ] + label "Node 2" + ] + node [ + id 3 + label "Node 3" + pos [ x 1 y 3 ] + ] + edge [ + source 1 + target 2 + label "Edge from node 1 to node 2" + color [line "blue" thickness 3] + + ] + edge [ + source 2 + target 3 + label "Edge from node 2 to node 3" + ] + edge [ + source 3 + target 1 + label "Edge from node 3 to node 1" + ] +] +""" + + def test_parse_gml_cytoscape_bug(self): + # example from issue #321, originally #324 in trac + cytoscape_example = """ +Creator "Cytoscape" +Version 1.0 +graph [ + node [ + root_index -3 + id -3 + graphics [ + x -96.0 + y -67.0 + w 40.0 + h 40.0 + fill "#ff9999" + type "ellipse" + outline "#666666" + outline_width 1.5 + ] + label "node2" + ] + node [ + root_index -2 + id -2 + graphics [ + x 63.0 + y 37.0 + w 40.0 + h 40.0 + fill "#ff9999" + type "ellipse" + outline "#666666" + outline_width 1.5 + ] + label "node1" + ] + node [ + root_index -1 + id -1 + graphics [ + x -31.0 + y -17.0 + w 40.0 + h 40.0 + fill "#ff9999" + type "ellipse" + outline "#666666" + outline_width 1.5 + ] + label "node0" + ] + edge [ + root_index -2 + target -2 + source -1 + graphics [ + width 1.5 + fill "#0000ff" + type "line" + Line [ + ] + source_arrow 0 + target_arrow 3 + ] + label "DirectedEdge" + ] + edge [ + root_index -1 + target -1 + source -3 + graphics [ + width 1.5 + fill "#0000ff" + type "line" + Line [ + ] + source_arrow 0 + target_arrow 3 + ] + label "DirectedEdge" + ] +] +""" + nx.parse_gml(cytoscape_example) + + def test_parse_gml(self): + G = nx.parse_gml(self.simple_data, label="label") + assert sorted(G.nodes()) == ["Node 1", "Node 2", "Node 3"] + assert sorted(G.edges()) == [ + ("Node 1", "Node 2"), + ("Node 2", "Node 3"), + ("Node 3", "Node 1"), + ] + + assert sorted(G.edges(data=True)) == [ + ( + "Node 1", + "Node 2", + { + "color": {"line": "blue", "thickness": 3}, + "label": "Edge from node 1 to node 2", + }, + ), + ("Node 2", "Node 3", {"label": "Edge from node 2 to node 3"}), + ("Node 3", "Node 1", {"label": "Edge from node 3 to node 1"}), + ] + + def test_read_gml(self, tmp_path): + fname = tmp_path / "test.gml" + with open(fname, "w") as fh: + fh.write(self.simple_data) + Gin = nx.read_gml(fname, label="label") + G = nx.parse_gml(self.simple_data, label="label") + assert sorted(G.nodes(data=True)) == sorted(Gin.nodes(data=True)) + assert sorted(G.edges(data=True)) == sorted(Gin.edges(data=True)) + + def test_labels_are_strings(self): + # GML requires labels to be strings (i.e., in quotes) + answer = """graph [ + node [ + id 0 + label "1203" + ] +]""" + G = nx.Graph() + G.add_node(1203) + data = "\n".join(nx.generate_gml(G, stringizer=literal_stringizer)) + assert data == answer + + def test_relabel_duplicate(self): + data = """ +graph +[ + label "" + directed 1 + node + [ + id 0 + label "same" + ] + node + [ + id 1 + label "same" + ] +] +""" + fh = io.BytesIO(data.encode("UTF-8")) + fh.seek(0) + pytest.raises(nx.NetworkXError, nx.read_gml, fh, label="label") + + @pytest.mark.parametrize("stringizer", (None, literal_stringizer)) + def test_tuplelabels(self, stringizer): + # https://github.com/networkx/networkx/pull/1048 + # Writing tuple labels to GML failed. + G = nx.Graph() + G.add_edge((0, 1), (1, 0)) + data = "\n".join(nx.generate_gml(G, stringizer=stringizer)) + answer = """graph [ + node [ + id 0 + label "(0,1)" + ] + node [ + id 1 + label "(1,0)" + ] + edge [ + source 0 + target 1 + ] +]""" + assert data == answer + + def test_quotes(self, tmp_path): + # https://github.com/networkx/networkx/issues/1061 + # Encoding quotes as HTML entities. + G = nx.path_graph(1) + G.name = "path_graph(1)" + attr = 'This is "quoted" and this is a copyright: ' + chr(169) + G.nodes[0]["demo"] = attr + with open(tmp_path / "test.gml", "w+b") as fobj: + nx.write_gml(G, fobj) + fobj.seek(0) + # Should be bytes in 2.x and 3.x + data = fobj.read().strip().decode("ascii") + answer = """graph [ + name "path_graph(1)" + node [ + id 0 + label "0" + demo "This is "quoted" and this is a copyright: ©" + ] +]""" + assert data == answer + + def test_unicode_node(self, tmp_path): + node = "node" + chr(169) + G = nx.Graph() + G.add_node(node) + with open(tmp_path / "test.gml", "w+b") as fobj: + nx.write_gml(G, fobj) + fobj.seek(0) + # Should be bytes in 2.x and 3.x + data = fobj.read().strip().decode("ascii") + answer = """graph [ + node [ + id 0 + label "node©" + ] +]""" + assert data == answer + + def test_float_label(self, tmp_path): + node = 1.0 + G = nx.Graph() + G.add_node(node) + with open(tmp_path / "test.gml", "w+b") as fobj: + nx.write_gml(G, fobj) + fobj.seek(0) + # Should be bytes in 2.x and 3.x + data = fobj.read().strip().decode("ascii") + answer = """graph [ + node [ + id 0 + label "1.0" + ] +]""" + assert data == answer + + def test_special_float_label(self, tmp_path): + special_floats = [float("nan"), float("+inf"), float("-inf")] + try: + import numpy as np + + special_floats += [np.nan, np.inf, np.inf * -1] + except ImportError: + special_floats += special_floats + + G = nx.cycle_graph(len(special_floats)) + attrs = dict(enumerate(special_floats)) + nx.set_node_attributes(G, attrs, "nodefloat") + edges = list(G.edges) + attrs = {edges[i]: value for i, value in enumerate(special_floats)} + nx.set_edge_attributes(G, attrs, "edgefloat") + + with open(tmp_path / "test.gml", "w+b") as fobj: + nx.write_gml(G, fobj) + fobj.seek(0) + # Should be bytes in 2.x and 3.x + data = fobj.read().strip().decode("ascii") + answer = """graph [ + node [ + id 0 + label "0" + nodefloat NAN + ] + node [ + id 1 + label "1" + nodefloat +INF + ] + node [ + id 2 + label "2" + nodefloat -INF + ] + node [ + id 3 + label "3" + nodefloat NAN + ] + node [ + id 4 + label "4" + nodefloat +INF + ] + node [ + id 5 + label "5" + nodefloat -INF + ] + edge [ + source 0 + target 1 + edgefloat NAN + ] + edge [ + source 0 + target 5 + edgefloat +INF + ] + edge [ + source 1 + target 2 + edgefloat -INF + ] + edge [ + source 2 + target 3 + edgefloat NAN + ] + edge [ + source 3 + target 4 + edgefloat +INF + ] + edge [ + source 4 + target 5 + edgefloat -INF + ] +]""" + assert data == answer + + fobj.seek(0) + graph = nx.read_gml(fobj) + for indx, value in enumerate(special_floats): + node_value = graph.nodes[str(indx)]["nodefloat"] + if math.isnan(value): + assert math.isnan(node_value) + else: + assert node_value == value + + edge = edges[indx] + string_edge = (str(edge[0]), str(edge[1])) + edge_value = graph.edges[string_edge]["edgefloat"] + if math.isnan(value): + assert math.isnan(edge_value) + else: + assert edge_value == value + + def test_name(self): + G = nx.parse_gml('graph [ name "x" node [ id 0 label "x" ] ]') + assert "x" == G.graph["name"] + G = nx.parse_gml('graph [ node [ id 0 label "x" ] ]') + assert "" == G.name + assert "name" not in G.graph + + def test_graph_types(self): + for directed in [None, False, True]: + for multigraph in [None, False, True]: + gml = "graph [" + if directed is not None: + gml += " directed " + str(int(directed)) + if multigraph is not None: + gml += " multigraph " + str(int(multigraph)) + gml += ' node [ id 0 label "0" ]' + gml += " edge [ source 0 target 0 ]" + gml += " ]" + G = nx.parse_gml(gml) + assert bool(directed) == G.is_directed() + assert bool(multigraph) == G.is_multigraph() + gml = "graph [\n" + if directed is True: + gml += " directed 1\n" + if multigraph is True: + gml += " multigraph 1\n" + gml += """ node [ + id 0 + label "0" + ] + edge [ + source 0 + target 0 +""" + if multigraph: + gml += " key 0\n" + gml += " ]\n]" + assert gml == "\n".join(nx.generate_gml(G)) + + def test_data_types(self): + data = [ + True, + False, + 10**20, + -2e33, + "'", + '"&&&""', + [{(b"\xfd",): "\x7f", chr(0x4444): (1, 2)}, (2, "3")], + ] + data.append(chr(0x14444)) + data.append(literal_eval("{2.3j, 1 - 2.3j, ()}")) + G = nx.Graph() + G.name = data + G.graph["data"] = data + G.add_node(0, int=-1, data={"data": data}) + G.add_edge(0, 0, float=-2.5, data=data) + gml = "\n".join(nx.generate_gml(G, stringizer=literal_stringizer)) + G = nx.parse_gml(gml, destringizer=literal_destringizer) + assert data == G.name + assert {"name": data, "data": data} == G.graph + assert list(G.nodes(data=True)) == [(0, {"int": -1, "data": {"data": data}})] + assert list(G.edges(data=True)) == [(0, 0, {"float": -2.5, "data": data})] + G = nx.Graph() + G.graph["data"] = "frozenset([1, 2, 3])" + G = nx.parse_gml(nx.generate_gml(G), destringizer=literal_eval) + assert G.graph["data"] == "frozenset([1, 2, 3])" + + def test_escape_unescape(self): + gml = """graph [ + name "&"䑄��&unknown;" +]""" + G = nx.parse_gml(gml) + assert ( + '&"\x0f' + chr(0x4444) + "��&unknown;" + == G.name + ) + gml = "\n".join(nx.generate_gml(G)) + alnu = "#1234567890;&#x1234567890abcdef" + answer = ( + """graph [ + name "&"䑄&""" + + alnu + + """;&unknown;" +]""" + ) + assert answer == gml + + def test_exceptions(self, tmp_path): + pytest.raises(ValueError, literal_destringizer, "(") + pytest.raises(ValueError, literal_destringizer, "frozenset([1, 2, 3])") + pytest.raises(ValueError, literal_destringizer, literal_destringizer) + pytest.raises(ValueError, literal_stringizer, frozenset([1, 2, 3])) + pytest.raises(ValueError, literal_stringizer, literal_stringizer) + with open(tmp_path / "test.gml", "w+b") as f: + f.write(codecs.BOM_UTF8 + b"graph[]") + f.seek(0) + pytest.raises(nx.NetworkXError, nx.read_gml, f) + + def assert_parse_error(gml): + pytest.raises(nx.NetworkXError, nx.parse_gml, gml) + + assert_parse_error(["graph [\n\n", "]"]) + assert_parse_error("") + assert_parse_error('Creator ""') + assert_parse_error("0") + assert_parse_error("graph ]") + assert_parse_error("graph [ 1 ]") + assert_parse_error("graph [ 1.E+2 ]") + assert_parse_error('graph [ "A" ]') + assert_parse_error("graph [ ] graph ]") + assert_parse_error("graph [ ] graph [ ]") + assert_parse_error("graph [ data [1, 2, 3] ]") + assert_parse_error("graph [ node [ ] ]") + assert_parse_error("graph [ node [ id 0 ] ]") + nx.parse_gml('graph [ node [ id "a" ] ]', label="id") + assert_parse_error("graph [ node [ id 0 label 0 ] node [ id 0 label 1 ] ]") + assert_parse_error("graph [ node [ id 0 label 0 ] node [ id 1 label 0 ] ]") + assert_parse_error("graph [ node [ id 0 label 0 ] edge [ ] ]") + assert_parse_error("graph [ node [ id 0 label 0 ] edge [ source 0 ] ]") + nx.parse_gml("graph [edge [ source 0 target 0 ] node [ id 0 label 0 ] ]") + assert_parse_error("graph [ node [ id 0 label 0 ] edge [ source 1 target 0 ] ]") + assert_parse_error("graph [ node [ id 0 label 0 ] edge [ source 0 target 1 ] ]") + assert_parse_error( + "graph [ node [ id 0 label 0 ] node [ id 1 label 1 ] " + "edge [ source 0 target 1 ] edge [ source 1 target 0 ] ]" + ) + nx.parse_gml( + "graph [ node [ id 0 label 0 ] node [ id 1 label 1 ] " + "edge [ source 0 target 1 ] edge [ source 1 target 0 ] " + "directed 1 ]" + ) + nx.parse_gml( + "graph [ node [ id 0 label 0 ] node [ id 1 label 1 ] " + "edge [ source 0 target 1 ] edge [ source 0 target 1 ]" + "multigraph 1 ]" + ) + nx.parse_gml( + "graph [ node [ id 0 label 0 ] node [ id 1 label 1 ] " + "edge [ source 0 target 1 key 0 ] edge [ source 0 target 1 ]" + "multigraph 1 ]" + ) + assert_parse_error( + "graph [ node [ id 0 label 0 ] node [ id 1 label 1 ] " + "edge [ source 0 target 1 key 0 ] edge [ source 0 target 1 key 0 ]" + "multigraph 1 ]" + ) + nx.parse_gml( + "graph [ node [ id 0 label 0 ] node [ id 1 label 1 ] " + "edge [ source 0 target 1 key 0 ] edge [ source 1 target 0 key 0 ]" + "directed 1 multigraph 1 ]" + ) + + # Tests for string convertible alphanumeric id and label values + nx.parse_gml("graph [edge [ source a target a ] node [ id a label b ] ]") + nx.parse_gml( + "graph [ node [ id n42 label 0 ] node [ id x43 label 1 ]" + "edge [ source n42 target x43 key 0 ]" + "edge [ source x43 target n42 key 0 ]" + "directed 1 multigraph 1 ]" + ) + assert_parse_error( + "graph [edge [ source '\u4200' target '\u4200' ] " + + "node [ id '\u4200' label b ] ]" + ) + + def assert_generate_error(*args, **kwargs): + pytest.raises( + nx.NetworkXError, lambda: list(nx.generate_gml(*args, **kwargs)) + ) + + G = nx.Graph() + G.graph[3] = 3 + assert_generate_error(G) + G = nx.Graph() + G.graph["3"] = 3 + assert_generate_error(G) + G = nx.Graph() + G.graph["data"] = frozenset([1, 2, 3]) + assert_generate_error(G, stringizer=literal_stringizer) + + def test_label_kwarg(self): + G = nx.parse_gml(self.simple_data, label="id") + assert sorted(G.nodes) == [1, 2, 3] + labels = [G.nodes[n]["label"] for n in sorted(G.nodes)] + assert labels == ["Node 1", "Node 2", "Node 3"] + + G = nx.parse_gml(self.simple_data, label=None) + assert sorted(G.nodes) == [1, 2, 3] + labels = [G.nodes[n]["label"] for n in sorted(G.nodes)] + assert labels == ["Node 1", "Node 2", "Node 3"] + + def test_outofrange_integers(self, tmp_path): + # GML restricts integers to 32 signed bits. + # Check that we honor this restriction on export + G = nx.Graph() + # Test export for numbers that barely fit or don't fit into 32 bits, + # and 3 numbers in the middle + numbers = { + "toosmall": (-(2**31)) - 1, + "small": -(2**31), + "med1": -4, + "med2": 0, + "med3": 17, + "big": (2**31) - 1, + "toobig": 2**31, + } + G.add_node("Node", **numbers) + + fname = tmp_path / "test.gml" + nx.write_gml(G, fname) + # Check that the export wrote the nonfitting numbers as strings + G2 = nx.read_gml(fname) + for attr, value in G2.nodes["Node"].items(): + if attr == "toosmall" or attr == "toobig": + assert type(value) == str + else: + assert type(value) == int + + def test_multiline(self): + # example from issue #6836 + multiline_example = """ +graph +[ + node + [ + id 0 + label "multiline node" + label2 "multiline1 + multiline2 + multiline3" + alt_name "id 0" + ] +] +""" + G = nx.parse_gml(multiline_example) + assert G.nodes["multiline node"] == { + "label2": "multiline1 multiline2 multiline3", + "alt_name": "id 0", + } + + +@contextmanager +def byte_file(): + _file_handle = io.BytesIO() + yield _file_handle + _file_handle.seek(0) + + +class TestPropertyLists: + def test_writing_graph_with_multi_element_property_list(self): + g = nx.Graph() + g.add_node("n1", properties=["element", 0, 1, 2.5, True, False]) + with byte_file() as f: + nx.write_gml(g, f) + result = f.read().decode() + + assert result == dedent( + """\ + graph [ + node [ + id 0 + label "n1" + properties "element" + properties 0 + properties 1 + properties 2.5 + properties 1 + properties 0 + ] + ] + """ + ) + + def test_writing_graph_with_one_element_property_list(self): + g = nx.Graph() + g.add_node("n1", properties=["element"]) + with byte_file() as f: + nx.write_gml(g, f) + result = f.read().decode() + + assert result == dedent( + """\ + graph [ + node [ + id 0 + label "n1" + properties "_networkx_list_start" + properties "element" + ] + ] + """ + ) + + def test_reading_graph_with_list_property(self): + with byte_file() as f: + f.write( + dedent( + """ + graph [ + node [ + id 0 + label "n1" + properties "element" + properties 0 + properties 1 + properties 2.5 + ] + ] + """ + ).encode("ascii") + ) + f.seek(0) + graph = nx.read_gml(f) + assert graph.nodes(data=True)["n1"] == {"properties": ["element", 0, 1, 2.5]} + + def test_reading_graph_with_single_element_list_property(self): + with byte_file() as f: + f.write( + dedent( + """ + graph [ + node [ + id 0 + label "n1" + properties "_networkx_list_start" + properties "element" + ] + ] + """ + ).encode("ascii") + ) + f.seek(0) + graph = nx.read_gml(f) + assert graph.nodes(data=True)["n1"] == {"properties": ["element"]} + + +@pytest.mark.parametrize("coll", ([], ())) +def test_stringize_empty_list_tuple(coll): + G = nx.path_graph(2) + G.nodes[0]["test"] = coll # test serializing an empty collection + f = io.BytesIO() + nx.write_gml(G, f) # Smoke test - should not raise + f.seek(0) + H = nx.read_gml(f) + assert H.nodes["0"]["test"] == coll # Check empty list round-trips properly + # Check full round-tripping. Note that nodes are loaded as strings by + # default, so there needs to be some remapping prior to comparison + H = nx.relabel_nodes(H, {"0": 0, "1": 1}) + assert nx.utils.graphs_equal(G, H) + # Same as above, but use destringizer for node remapping. Should have no + # effect on node attr + f.seek(0) + H = nx.read_gml(f, destringizer=int) + assert nx.utils.graphs_equal(G, H) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_graph6.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_graph6.py new file mode 100644 index 00000000..a8032694 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_graph6.py @@ -0,0 +1,168 @@ +from io import BytesIO + +import pytest + +import networkx as nx +import networkx.readwrite.graph6 as g6 +from networkx.utils import edges_equal, nodes_equal + + +class TestGraph6Utils: + def test_n_data_n_conversion(self): + for i in [0, 1, 42, 62, 63, 64, 258047, 258048, 7744773, 68719476735]: + assert g6.data_to_n(g6.n_to_data(i))[0] == i + assert g6.data_to_n(g6.n_to_data(i))[1] == [] + assert g6.data_to_n(g6.n_to_data(i) + [42, 43])[1] == [42, 43] + + +class TestFromGraph6Bytes: + def test_from_graph6_bytes(self): + data = b"DF{" + G = nx.from_graph6_bytes(data) + assert nodes_equal(G.nodes(), [0, 1, 2, 3, 4]) + assert edges_equal( + G.edges(), [(0, 3), (0, 4), (1, 3), (1, 4), (2, 3), (2, 4), (3, 4)] + ) + + def test_read_equals_from_bytes(self): + data = b"DF{" + G = nx.from_graph6_bytes(data) + fh = BytesIO(data) + Gin = nx.read_graph6(fh) + assert nodes_equal(G.nodes(), Gin.nodes()) + assert edges_equal(G.edges(), Gin.edges()) + + +class TestReadGraph6: + def test_read_many_graph6(self): + """Test for reading many graphs from a file into a list.""" + data = b"DF{\nD`{\nDqK\nD~{\n" + fh = BytesIO(data) + glist = nx.read_graph6(fh) + assert len(glist) == 4 + for G in glist: + assert sorted(G) == list(range(5)) + + +class TestWriteGraph6: + """Unit tests for writing a graph to a file in graph6 format.""" + + def test_null_graph(self): + result = BytesIO() + nx.write_graph6(nx.null_graph(), result) + assert result.getvalue() == b">>graph6<<?\n" + + def test_trivial_graph(self): + result = BytesIO() + nx.write_graph6(nx.trivial_graph(), result) + assert result.getvalue() == b">>graph6<<@\n" + + def test_complete_graph(self): + result = BytesIO() + nx.write_graph6(nx.complete_graph(4), result) + assert result.getvalue() == b">>graph6<<C~\n" + + def test_large_complete_graph(self): + result = BytesIO() + nx.write_graph6(nx.complete_graph(67), result, header=False) + assert result.getvalue() == b"~?@B" + b"~" * 368 + b"w\n" + + def test_no_header(self): + result = BytesIO() + nx.write_graph6(nx.complete_graph(4), result, header=False) + assert result.getvalue() == b"C~\n" + + def test_complete_bipartite_graph(self): + result = BytesIO() + G = nx.complete_bipartite_graph(6, 9) + nx.write_graph6(G, result, header=False) + # The expected encoding here was verified by Sage. + assert result.getvalue() == b"N??F~z{~Fw^_~?~?^_?\n" + + @pytest.mark.parametrize("G", (nx.MultiGraph(), nx.DiGraph())) + def test_no_directed_or_multi_graphs(self, G): + with pytest.raises(nx.NetworkXNotImplemented): + nx.write_graph6(G, BytesIO()) + + def test_length(self): + for i in list(range(13)) + [31, 47, 62, 63, 64, 72]: + g = nx.random_graphs.gnm_random_graph(i, i * i // 4, seed=i) + gstr = BytesIO() + nx.write_graph6(g, gstr, header=False) + # Strip the trailing newline. + gstr = gstr.getvalue().rstrip() + assert len(gstr) == ((i - 1) * i // 2 + 5) // 6 + (1 if i < 63 else 4) + + def test_roundtrip(self): + for i in list(range(13)) + [31, 47, 62, 63, 64, 72]: + G = nx.random_graphs.gnm_random_graph(i, i * i // 4, seed=i) + f = BytesIO() + nx.write_graph6(G, f) + f.seek(0) + H = nx.read_graph6(f) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + + def test_write_path(self, tmp_path): + with open(tmp_path / "test.g6", "w+b") as f: + g6.write_graph6_file(nx.null_graph(), f) + f.seek(0) + assert f.read() == b">>graph6<<?\n" + + @pytest.mark.parametrize("edge", ((0, 1), (1, 2), (1, 42))) + def test_relabeling(self, edge): + G = nx.Graph([edge]) + f = BytesIO() + nx.write_graph6(G, f) + f.seek(0) + assert f.read() == b">>graph6<<A_\n" + + +class TestToGraph6Bytes: + def test_null_graph(self): + G = nx.null_graph() + assert g6.to_graph6_bytes(G) == b">>graph6<<?\n" + + def test_trivial_graph(self): + G = nx.trivial_graph() + assert g6.to_graph6_bytes(G) == b">>graph6<<@\n" + + def test_complete_graph(self): + assert g6.to_graph6_bytes(nx.complete_graph(4)) == b">>graph6<<C~\n" + + def test_large_complete_graph(self): + G = nx.complete_graph(67) + assert g6.to_graph6_bytes(G, header=False) == b"~?@B" + b"~" * 368 + b"w\n" + + def test_no_header(self): + G = nx.complete_graph(4) + assert g6.to_graph6_bytes(G, header=False) == b"C~\n" + + def test_complete_bipartite_graph(self): + G = nx.complete_bipartite_graph(6, 9) + assert g6.to_graph6_bytes(G, header=False) == b"N??F~z{~Fw^_~?~?^_?\n" + + @pytest.mark.parametrize("G", (nx.MultiGraph(), nx.DiGraph())) + def test_no_directed_or_multi_graphs(self, G): + with pytest.raises(nx.NetworkXNotImplemented): + g6.to_graph6_bytes(G) + + def test_length(self): + for i in list(range(13)) + [31, 47, 62, 63, 64, 72]: + G = nx.random_graphs.gnm_random_graph(i, i * i // 4, seed=i) + # Strip the trailing newline. + gstr = g6.to_graph6_bytes(G, header=False).rstrip() + assert len(gstr) == ((i - 1) * i // 2 + 5) // 6 + (1 if i < 63 else 4) + + def test_roundtrip(self): + for i in list(range(13)) + [31, 47, 62, 63, 64, 72]: + G = nx.random_graphs.gnm_random_graph(i, i * i // 4, seed=i) + data = g6.to_graph6_bytes(G) + H = nx.from_graph6_bytes(data.rstrip()) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + + @pytest.mark.parametrize("edge", ((0, 1), (1, 2), (1, 42))) + def test_relabeling(self, edge): + G = nx.Graph([edge]) + assert g6.to_graph6_bytes(G) == b">>graph6<<A_\n" diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_graphml.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_graphml.py new file mode 100644 index 00000000..5ffa837e --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_graphml.py @@ -0,0 +1,1531 @@ +import io + +import pytest + +import networkx as nx +from networkx.readwrite.graphml import GraphMLWriter +from networkx.utils import edges_equal, nodes_equal + + +class BaseGraphML: + @classmethod + def setup_class(cls): + cls.simple_directed_data = """<?xml version="1.0" encoding="UTF-8"?> +<!-- This file was written by the JAVA GraphML Library.--> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G" edgedefault="directed"> + <node id="n0"/> + <node id="n1"/> + <node id="n2"/> + <node id="n3"/> + <node id="n4"/> + <node id="n5"/> + <node id="n6"/> + <node id="n7"/> + <node id="n8"/> + <node id="n9"/> + <node id="n10"/> + <edge id="foo" source="n0" target="n2"/> + <edge source="n1" target="n2"/> + <edge source="n2" target="n3"/> + <edge source="n3" target="n5"/> + <edge source="n3" target="n4"/> + <edge source="n4" target="n6"/> + <edge source="n6" target="n5"/> + <edge source="n5" target="n7"/> + <edge source="n6" target="n8"/> + <edge source="n8" target="n7"/> + <edge source="n8" target="n9"/> + </graph> +</graphml>""" + cls.simple_directed_graph = nx.DiGraph() + cls.simple_directed_graph.add_node("n10") + cls.simple_directed_graph.add_edge("n0", "n2", id="foo") + cls.simple_directed_graph.add_edge("n0", "n2") + cls.simple_directed_graph.add_edges_from( + [ + ("n1", "n2"), + ("n2", "n3"), + ("n3", "n5"), + ("n3", "n4"), + ("n4", "n6"), + ("n6", "n5"), + ("n5", "n7"), + ("n6", "n8"), + ("n8", "n7"), + ("n8", "n9"), + ] + ) + cls.simple_directed_fh = io.BytesIO(cls.simple_directed_data.encode("UTF-8")) + + cls.attribute_data = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key id="d0" for="node" attr.name="color" attr.type="string"> + <default>yellow</default> + </key> + <key id="d1" for="edge" attr.name="weight" attr.type="double"/> + <graph id="G" edgedefault="directed"> + <node id="n0"> + <data key="d0">green</data> + </node> + <node id="n1"/> + <node id="n2"> + <data key="d0">blue</data> + </node> + <node id="n3"> + <data key="d0">red</data> + </node> + <node id="n4"/> + <node id="n5"> + <data key="d0">turquoise</data> + </node> + <edge id="e0" source="n0" target="n2"> + <data key="d1">1.0</data> + </edge> + <edge id="e1" source="n0" target="n1"> + <data key="d1">1.0</data> + </edge> + <edge id="e2" source="n1" target="n3"> + <data key="d1">2.0</data> + </edge> + <edge id="e3" source="n3" target="n2"/> + <edge id="e4" source="n2" target="n4"/> + <edge id="e5" source="n3" target="n5"/> + <edge id="e6" source="n5" target="n4"> + <data key="d1">1.1</data> + </edge> + </graph> +</graphml> +""" + cls.attribute_graph = nx.DiGraph(id="G") + cls.attribute_graph.graph["node_default"] = {"color": "yellow"} + cls.attribute_graph.add_node("n0", color="green") + cls.attribute_graph.add_node("n2", color="blue") + cls.attribute_graph.add_node("n3", color="red") + cls.attribute_graph.add_node("n4") + cls.attribute_graph.add_node("n5", color="turquoise") + cls.attribute_graph.add_edge("n0", "n2", id="e0", weight=1.0) + cls.attribute_graph.add_edge("n0", "n1", id="e1", weight=1.0) + cls.attribute_graph.add_edge("n1", "n3", id="e2", weight=2.0) + cls.attribute_graph.add_edge("n3", "n2", id="e3") + cls.attribute_graph.add_edge("n2", "n4", id="e4") + cls.attribute_graph.add_edge("n3", "n5", id="e5") + cls.attribute_graph.add_edge("n5", "n4", id="e6", weight=1.1) + cls.attribute_fh = io.BytesIO(cls.attribute_data.encode("UTF-8")) + + cls.node_attribute_default_data = """<?xml version="1.0" encoding="UTF-8"?> + <graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key id="d0" for="node" attr.name="boolean_attribute" attr.type="boolean"><default>false</default></key> + <key id="d1" for="node" attr.name="int_attribute" attr.type="int"><default>0</default></key> + <key id="d2" for="node" attr.name="long_attribute" attr.type="long"><default>0</default></key> + <key id="d3" for="node" attr.name="float_attribute" attr.type="float"><default>0.0</default></key> + <key id="d4" for="node" attr.name="double_attribute" attr.type="double"><default>0.0</default></key> + <key id="d5" for="node" attr.name="string_attribute" attr.type="string"><default>Foo</default></key> + <graph id="G" edgedefault="directed"> + <node id="n0"/> + <node id="n1"/> + <edge id="e0" source="n0" target="n1"/> + </graph> + </graphml> + """ + cls.node_attribute_default_graph = nx.DiGraph(id="G") + cls.node_attribute_default_graph.graph["node_default"] = { + "boolean_attribute": False, + "int_attribute": 0, + "long_attribute": 0, + "float_attribute": 0.0, + "double_attribute": 0.0, + "string_attribute": "Foo", + } + cls.node_attribute_default_graph.add_node("n0") + cls.node_attribute_default_graph.add_node("n1") + cls.node_attribute_default_graph.add_edge("n0", "n1", id="e0") + cls.node_attribute_default_fh = io.BytesIO( + cls.node_attribute_default_data.encode("UTF-8") + ) + + cls.attribute_named_key_ids_data = """<?xml version='1.0' encoding='utf-8'?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key id="edge_prop" for="edge" attr.name="edge_prop" attr.type="string"/> + <key id="prop2" for="node" attr.name="prop2" attr.type="string"/> + <key id="prop1" for="node" attr.name="prop1" attr.type="string"/> + <graph edgedefault="directed"> + <node id="0"> + <data key="prop1">val1</data> + <data key="prop2">val2</data> + </node> + <node id="1"> + <data key="prop1">val_one</data> + <data key="prop2">val2</data> + </node> + <edge source="0" target="1"> + <data key="edge_prop">edge_value</data> + </edge> + </graph> +</graphml> +""" + cls.attribute_named_key_ids_graph = nx.DiGraph() + cls.attribute_named_key_ids_graph.add_node("0", prop1="val1", prop2="val2") + cls.attribute_named_key_ids_graph.add_node("1", prop1="val_one", prop2="val2") + cls.attribute_named_key_ids_graph.add_edge("0", "1", edge_prop="edge_value") + fh = io.BytesIO(cls.attribute_named_key_ids_data.encode("UTF-8")) + cls.attribute_named_key_ids_fh = fh + + cls.attribute_numeric_type_data = """<?xml version='1.0' encoding='utf-8'?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key attr.name="weight" attr.type="double" for="node" id="d1" /> + <key attr.name="weight" attr.type="double" for="edge" id="d0" /> + <graph edgedefault="directed"> + <node id="n0"> + <data key="d1">1</data> + </node> + <node id="n1"> + <data key="d1">2.0</data> + </node> + <edge source="n0" target="n1"> + <data key="d0">1</data> + </edge> + <edge source="n1" target="n0"> + <data key="d0">k</data> + </edge> + <edge source="n1" target="n1"> + <data key="d0">1.0</data> + </edge> + </graph> +</graphml> +""" + cls.attribute_numeric_type_graph = nx.DiGraph() + cls.attribute_numeric_type_graph.add_node("n0", weight=1) + cls.attribute_numeric_type_graph.add_node("n1", weight=2.0) + cls.attribute_numeric_type_graph.add_edge("n0", "n1", weight=1) + cls.attribute_numeric_type_graph.add_edge("n1", "n1", weight=1.0) + fh = io.BytesIO(cls.attribute_numeric_type_data.encode("UTF-8")) + cls.attribute_numeric_type_fh = fh + + cls.simple_undirected_data = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G"> + <node id="n0"/> + <node id="n1"/> + <node id="n2"/> + <node id="n10"/> + <edge id="foo" source="n0" target="n2"/> + <edge source="n1" target="n2"/> + <edge source="n2" target="n3"/> + </graph> +</graphml>""" + # <edge source="n8" target="n10" directed="false"/> + cls.simple_undirected_graph = nx.Graph() + cls.simple_undirected_graph.add_node("n10") + cls.simple_undirected_graph.add_edge("n0", "n2", id="foo") + cls.simple_undirected_graph.add_edges_from([("n1", "n2"), ("n2", "n3")]) + fh = io.BytesIO(cls.simple_undirected_data.encode("UTF-8")) + cls.simple_undirected_fh = fh + + cls.undirected_multigraph_data = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G"> + <node id="n0"/> + <node id="n1"/> + <node id="n2"/> + <node id="n10"/> + <edge id="e0" source="n0" target="n2"/> + <edge id="e1" source="n1" target="n2"/> + <edge id="e2" source="n2" target="n1"/> + </graph> +</graphml>""" + cls.undirected_multigraph = nx.MultiGraph() + cls.undirected_multigraph.add_node("n10") + cls.undirected_multigraph.add_edge("n0", "n2", id="e0") + cls.undirected_multigraph.add_edge("n1", "n2", id="e1") + cls.undirected_multigraph.add_edge("n2", "n1", id="e2") + fh = io.BytesIO(cls.undirected_multigraph_data.encode("UTF-8")) + cls.undirected_multigraph_fh = fh + + cls.undirected_multigraph_no_multiedge_data = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G"> + <node id="n0"/> + <node id="n1"/> + <node id="n2"/> + <node id="n10"/> + <edge id="e0" source="n0" target="n2"/> + <edge id="e1" source="n1" target="n2"/> + <edge id="e2" source="n2" target="n3"/> + </graph> +</graphml>""" + cls.undirected_multigraph_no_multiedge = nx.MultiGraph() + cls.undirected_multigraph_no_multiedge.add_node("n10") + cls.undirected_multigraph_no_multiedge.add_edge("n0", "n2", id="e0") + cls.undirected_multigraph_no_multiedge.add_edge("n1", "n2", id="e1") + cls.undirected_multigraph_no_multiedge.add_edge("n2", "n3", id="e2") + fh = io.BytesIO(cls.undirected_multigraph_no_multiedge_data.encode("UTF-8")) + cls.undirected_multigraph_no_multiedge_fh = fh + + cls.multigraph_only_ids_for_multiedges_data = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G"> + <node id="n0"/> + <node id="n1"/> + <node id="n2"/> + <node id="n10"/> + <edge source="n0" target="n2"/> + <edge id="e1" source="n1" target="n2"/> + <edge id="e2" source="n2" target="n1"/> + </graph> +</graphml>""" + cls.multigraph_only_ids_for_multiedges = nx.MultiGraph() + cls.multigraph_only_ids_for_multiedges.add_node("n10") + cls.multigraph_only_ids_for_multiedges.add_edge("n0", "n2") + cls.multigraph_only_ids_for_multiedges.add_edge("n1", "n2", id="e1") + cls.multigraph_only_ids_for_multiedges.add_edge("n2", "n1", id="e2") + fh = io.BytesIO(cls.multigraph_only_ids_for_multiedges_data.encode("UTF-8")) + cls.multigraph_only_ids_for_multiedges_fh = fh + + +class TestReadGraphML(BaseGraphML): + def test_read_simple_directed_graphml(self): + G = self.simple_directed_graph + H = nx.read_graphml(self.simple_directed_fh) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(G.edges()) == sorted(H.edges()) + assert sorted(G.edges(data=True)) == sorted(H.edges(data=True)) + self.simple_directed_fh.seek(0) + + PG = nx.parse_graphml(self.simple_directed_data) + assert sorted(G.nodes()) == sorted(PG.nodes()) + assert sorted(G.edges()) == sorted(PG.edges()) + assert sorted(G.edges(data=True)) == sorted(PG.edges(data=True)) + + def test_read_simple_undirected_graphml(self): + G = self.simple_undirected_graph + H = nx.read_graphml(self.simple_undirected_fh) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + self.simple_undirected_fh.seek(0) + + PG = nx.parse_graphml(self.simple_undirected_data) + assert nodes_equal(G.nodes(), PG.nodes()) + assert edges_equal(G.edges(), PG.edges()) + + def test_read_undirected_multigraph_graphml(self): + G = self.undirected_multigraph + H = nx.read_graphml(self.undirected_multigraph_fh) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + self.undirected_multigraph_fh.seek(0) + + PG = nx.parse_graphml(self.undirected_multigraph_data) + assert nodes_equal(G.nodes(), PG.nodes()) + assert edges_equal(G.edges(), PG.edges()) + + def test_read_undirected_multigraph_no_multiedge_graphml(self): + G = self.undirected_multigraph_no_multiedge + H = nx.read_graphml(self.undirected_multigraph_no_multiedge_fh) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + self.undirected_multigraph_no_multiedge_fh.seek(0) + + PG = nx.parse_graphml(self.undirected_multigraph_no_multiedge_data) + assert nodes_equal(G.nodes(), PG.nodes()) + assert edges_equal(G.edges(), PG.edges()) + + def test_read_undirected_multigraph_only_ids_for_multiedges_graphml(self): + G = self.multigraph_only_ids_for_multiedges + H = nx.read_graphml(self.multigraph_only_ids_for_multiedges_fh) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + self.multigraph_only_ids_for_multiedges_fh.seek(0) + + PG = nx.parse_graphml(self.multigraph_only_ids_for_multiedges_data) + assert nodes_equal(G.nodes(), PG.nodes()) + assert edges_equal(G.edges(), PG.edges()) + + def test_read_attribute_graphml(self): + G = self.attribute_graph + H = nx.read_graphml(self.attribute_fh) + assert nodes_equal(G.nodes(True), sorted(H.nodes(data=True))) + ge = sorted(G.edges(data=True)) + he = sorted(H.edges(data=True)) + for a, b in zip(ge, he): + assert a == b + self.attribute_fh.seek(0) + + PG = nx.parse_graphml(self.attribute_data) + assert sorted(G.nodes(True)) == sorted(PG.nodes(data=True)) + ge = sorted(G.edges(data=True)) + he = sorted(PG.edges(data=True)) + for a, b in zip(ge, he): + assert a == b + + def test_node_default_attribute_graphml(self): + G = self.node_attribute_default_graph + H = nx.read_graphml(self.node_attribute_default_fh) + assert G.graph["node_default"] == H.graph["node_default"] + + def test_directed_edge_in_undirected(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G"> + <node id="n0"/> + <node id="n1"/> + <node id="n2"/> + <edge source="n0" target="n1"/> + <edge source="n1" target="n2" directed='true'/> + </graph> +</graphml>""" + fh = io.BytesIO(s.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_graphml, fh) + pytest.raises(nx.NetworkXError, nx.parse_graphml, s) + + def test_undirected_edge_in_directed(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G" edgedefault='directed'> + <node id="n0"/> + <node id="n1"/> + <node id="n2"/> + <edge source="n0" target="n1"/> + <edge source="n1" target="n2" directed='false'/> + </graph> +</graphml>""" + fh = io.BytesIO(s.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_graphml, fh) + pytest.raises(nx.NetworkXError, nx.parse_graphml, s) + + def test_key_raise(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key id="d0" for="node" attr.name="color" attr.type="string"> + <default>yellow</default> + </key> + <key id="d1" for="edge" attr.name="weight" attr.type="double"/> + <graph id="G" edgedefault="directed"> + <node id="n0"> + <data key="d0">green</data> + </node> + <node id="n1"/> + <node id="n2"> + <data key="d0">blue</data> + </node> + <edge id="e0" source="n0" target="n2"> + <data key="d2">1.0</data> + </edge> + </graph> +</graphml> +""" + fh = io.BytesIO(s.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_graphml, fh) + pytest.raises(nx.NetworkXError, nx.parse_graphml, s) + + def test_hyperedge_raise(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key id="d0" for="node" attr.name="color" attr.type="string"> + <default>yellow</default> + </key> + <key id="d1" for="edge" attr.name="weight" attr.type="double"/> + <graph id="G" edgedefault="directed"> + <node id="n0"> + <data key="d0">green</data> + </node> + <node id="n1"/> + <node id="n2"> + <data key="d0">blue</data> + </node> + <hyperedge id="e0" source="n0" target="n2"> + <endpoint node="n0"/> + <endpoint node="n1"/> + <endpoint node="n2"/> + </hyperedge> + </graph> +</graphml> +""" + fh = io.BytesIO(s.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_graphml, fh) + pytest.raises(nx.NetworkXError, nx.parse_graphml, s) + + def test_multigraph_keys(self): + # Test that reading multigraphs uses edge id attributes as keys + s = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <graph id="G" edgedefault="directed"> + <node id="n0"/> + <node id="n1"/> + <edge id="e0" source="n0" target="n1"/> + <edge id="e1" source="n0" target="n1"/> + </graph> +</graphml> +""" + fh = io.BytesIO(s.encode("UTF-8")) + G = nx.read_graphml(fh) + expected = [("n0", "n1", "e0"), ("n0", "n1", "e1")] + assert sorted(G.edges(keys=True)) == expected + fh.seek(0) + H = nx.parse_graphml(s) + assert sorted(H.edges(keys=True)) == expected + + def test_preserve_multi_edge_data(self): + """ + Test that data and keys of edges are preserved on consequent + write and reads + """ + G = nx.MultiGraph() + G.add_node(1) + G.add_node(2) + G.add_edges_from( + [ + # edges with no data, no keys: + (1, 2), + # edges with only data: + (1, 2, {"key": "data_key1"}), + (1, 2, {"id": "data_id2"}), + (1, 2, {"key": "data_key3", "id": "data_id3"}), + # edges with both data and keys: + (1, 2, 103, {"key": "data_key4"}), + (1, 2, 104, {"id": "data_id5"}), + (1, 2, 105, {"key": "data_key6", "id": "data_id7"}), + ] + ) + fh = io.BytesIO() + nx.write_graphml(G, fh) + fh.seek(0) + H = nx.read_graphml(fh, node_type=int) + assert edges_equal(G.edges(data=True, keys=True), H.edges(data=True, keys=True)) + assert G._adj == H._adj + + Gadj = { + str(node): { + str(nbr): {str(ekey): dd for ekey, dd in key_dict.items()} + for nbr, key_dict in nbr_dict.items() + } + for node, nbr_dict in G._adj.items() + } + fh.seek(0) + HH = nx.read_graphml(fh, node_type=str, edge_key_type=str) + assert Gadj == HH._adj + + fh.seek(0) + string_fh = fh.read() + HH = nx.parse_graphml(string_fh, node_type=str, edge_key_type=str) + assert Gadj == HH._adj + + def test_yfiles_extension(self): + data = """<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xmlns:y="http://www.yworks.com/xml/graphml" + xmlns:yed="http://www.yworks.com/xml/yed/3" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <!--Created by yFiles for Java 2.7--> + <key for="graphml" id="d0" yfiles.type="resources"/> + <key attr.name="url" attr.type="string" for="node" id="d1"/> + <key attr.name="description" attr.type="string" for="node" id="d2"/> + <key for="node" id="d3" yfiles.type="nodegraphics"/> + <key attr.name="Description" attr.type="string" for="graph" id="d4"> + <default/> + </key> + <key attr.name="url" attr.type="string" for="edge" id="d5"/> + <key attr.name="description" attr.type="string" for="edge" id="d6"/> + <key for="edge" id="d7" yfiles.type="edgegraphics"/> + <graph edgedefault="directed" id="G"> + <node id="n0"> + <data key="d3"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="125.0" y="100.0"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" + borderDistance="0.0" fontFamily="Dialog" fontSize="13" + fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" + height="19.1328125" modelName="internal" modelPosition="c" + textColor="#000000" visible="true" width="12.27099609375" + x="8.864501953125" y="5.43359375">1</y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + <node id="n1"> + <data key="d3"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="183.0" y="205.0"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" + borderDistance="0.0" fontFamily="Dialog" fontSize="13" + fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" + height="19.1328125" modelName="internal" modelPosition="c" + textColor="#000000" visible="true" width="12.27099609375" + x="8.864501953125" y="5.43359375">2</y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + <node id="n2"> + <data key="d6" xml:space="preserve"><![CDATA[description +line1 +line2]]></data> + <data key="d3"> + <y:GenericNode configuration="com.yworks.flowchart.terminator"> + <y:Geometry height="40.0" width="80.0" x="950.0" y="286.0"/> + <y:Fill color="#E8EEF7" color2="#B7C9E3" transparent="false"/> + <y:BorderStyle color="#000000" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" + fontFamily="Dialog" fontSize="12" fontStyle="plain" + hasBackgroundColor="false" hasLineColor="false" height="17.96875" + horizontalTextPosition="center" iconTextGap="4" modelName="custom" + textColor="#000000" verticalTextPosition="bottom" visible="true" + width="67.984375" x="6.0078125" xml:space="preserve" + y="11.015625">3<y:LabelModel> + <y:SmartNodeLabelModel distance="4.0"/></y:LabelModel> + <y:ModelParameter><y:SmartNodeLabelModelParameter labelRatioX="0.0" + labelRatioY="0.0" nodeRatioX="0.0" nodeRatioY="0.0" offsetX="0.0" + offsetY="0.0" upX="0.0" upY="-1.0"/></y:ModelParameter></y:NodeLabel> + </y:GenericNode> + </data> + </node> + <edge id="e0" source="n0" target="n1"> + <data key="d7"> + <y:PolyLineEdge> + <y:Path sx="0.0" sy="0.0" tx="0.0" ty="0.0"/> + <y:LineStyle color="#000000" type="line" width="1.0"/> + <y:Arrows source="none" target="standard"/> + <y:BendStyle smoothed="false"/> + </y:PolyLineEdge> + </data> + </edge> + </graph> + <data key="d0"> + <y:Resources/> + </data> +</graphml> +""" + fh = io.BytesIO(data.encode("UTF-8")) + G = nx.read_graphml(fh, force_multigraph=True) + assert list(G.edges()) == [("n0", "n1")] + assert G.has_edge("n0", "n1", key="e0") + assert G.nodes["n0"]["label"] == "1" + assert G.nodes["n1"]["label"] == "2" + assert G.nodes["n2"]["label"] == "3" + assert G.nodes["n0"]["shape_type"] == "rectangle" + assert G.nodes["n1"]["shape_type"] == "rectangle" + assert G.nodes["n2"]["shape_type"] == "com.yworks.flowchart.terminator" + assert G.nodes["n2"]["description"] == "description\nline1\nline2" + fh.seek(0) + G = nx.read_graphml(fh) + assert list(G.edges()) == [("n0", "n1")] + assert G["n0"]["n1"]["id"] == "e0" + assert G.nodes["n0"]["label"] == "1" + assert G.nodes["n1"]["label"] == "2" + assert G.nodes["n2"]["label"] == "3" + assert G.nodes["n0"]["shape_type"] == "rectangle" + assert G.nodes["n1"]["shape_type"] == "rectangle" + assert G.nodes["n2"]["shape_type"] == "com.yworks.flowchart.terminator" + assert G.nodes["n2"]["description"] == "description\nline1\nline2" + + H = nx.parse_graphml(data, force_multigraph=True) + assert list(H.edges()) == [("n0", "n1")] + assert H.has_edge("n0", "n1", key="e0") + assert H.nodes["n0"]["label"] == "1" + assert H.nodes["n1"]["label"] == "2" + assert H.nodes["n2"]["label"] == "3" + + H = nx.parse_graphml(data) + assert list(H.edges()) == [("n0", "n1")] + assert H["n0"]["n1"]["id"] == "e0" + assert H.nodes["n0"]["label"] == "1" + assert H.nodes["n1"]["label"] == "2" + assert H.nodes["n2"]["label"] == "3" + + def test_bool(self): + s = """<?xml version="1.0" encoding="UTF-8"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key id="d0" for="node" attr.name="test" attr.type="boolean"> + <default>false</default> + </key> + <graph id="G" edgedefault="directed"> + <node id="n0"> + <data key="d0">true</data> + </node> + <node id="n1"/> + <node id="n2"> + <data key="d0">false</data> + </node> + <node id="n3"> + <data key="d0">FaLsE</data> + </node> + <node id="n4"> + <data key="d0">True</data> + </node> + <node id="n5"> + <data key="d0">0</data> + </node> + <node id="n6"> + <data key="d0">1</data> + </node> + </graph> +</graphml> +""" + fh = io.BytesIO(s.encode("UTF-8")) + G = nx.read_graphml(fh) + H = nx.parse_graphml(s) + for graph in [G, H]: + assert graph.nodes["n0"]["test"] + assert not graph.nodes["n2"]["test"] + assert not graph.nodes["n3"]["test"] + assert graph.nodes["n4"]["test"] + assert not graph.nodes["n5"]["test"] + assert graph.nodes["n6"]["test"] + + def test_graphml_header_line(self): + good = """<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key id="d0" for="node" attr.name="test" attr.type="boolean"> + <default>false</default> + </key> + <graph id="G"> + <node id="n0"> + <data key="d0">true</data> + </node> + </graph> +</graphml> +""" + bad = """<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<graphml> + <key id="d0" for="node" attr.name="test" attr.type="boolean"> + <default>false</default> + </key> + <graph id="G"> + <node id="n0"> + <data key="d0">true</data> + </node> + </graph> +</graphml> +""" + ugly = """<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<graphml xmlns="https://ghghgh"> + <key id="d0" for="node" attr.name="test" attr.type="boolean"> + <default>false</default> + </key> + <graph id="G"> + <node id="n0"> + <data key="d0">true</data> + </node> + </graph> +</graphml> +""" + for s in (good, bad): + fh = io.BytesIO(s.encode("UTF-8")) + G = nx.read_graphml(fh) + H = nx.parse_graphml(s) + for graph in [G, H]: + assert graph.nodes["n0"]["test"] + + fh = io.BytesIO(ugly.encode("UTF-8")) + pytest.raises(nx.NetworkXError, nx.read_graphml, fh) + pytest.raises(nx.NetworkXError, nx.parse_graphml, ugly) + + def test_read_attributes_with_groups(self): + data = """\ +<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" xmlns:java="http://www.yworks.com/xml/yfiles-common/1.0/java" xmlns:sys="http://www.yworks.com/xml/yfiles-common/markup/primitives/2.0" xmlns:x="http://www.yworks.com/xml/yfiles-common/markup/2.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:y="http://www.yworks.com/xml/graphml" xmlns:yed="http://www.yworks.com/xml/yed/3" xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns http://www.yworks.com/xml/schema/graphml/1.1/ygraphml.xsd"> + <!--Created by yEd 3.17--> + <key attr.name="Description" attr.type="string" for="graph" id="d0"/> + <key for="port" id="d1" yfiles.type="portgraphics"/> + <key for="port" id="d2" yfiles.type="portgeometry"/> + <key for="port" id="d3" yfiles.type="portuserdata"/> + <key attr.name="CustomProperty" attr.type="string" for="node" id="d4"> + <default/> + </key> + <key attr.name="url" attr.type="string" for="node" id="d5"/> + <key attr.name="description" attr.type="string" for="node" id="d6"/> + <key for="node" id="d7" yfiles.type="nodegraphics"/> + <key for="graphml" id="d8" yfiles.type="resources"/> + <key attr.name="url" attr.type="string" for="edge" id="d9"/> + <key attr.name="description" attr.type="string" for="edge" id="d10"/> + <key for="edge" id="d11" yfiles.type="edgegraphics"/> + <graph edgedefault="directed" id="G"> + <data key="d0"/> + <node id="n0"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d6"/> + <data key="d7"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="125.0" y="-255.4611111111111"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" raised="false" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="17.96875" horizontalTextPosition="center" iconTextGap="4" modelName="custom" textColor="#000000" verticalTextPosition="bottom" visible="true" width="11.634765625" x="9.1826171875" y="6.015625">2<y:LabelModel> + <y:SmartNodeLabelModel distance="4.0"/> + </y:LabelModel> + <y:ModelParameter> + <y:SmartNodeLabelModelParameter labelRatioX="0.0" labelRatioY="0.0" nodeRatioX="0.0" nodeRatioY="0.0" offsetX="0.0" offsetY="0.0" upX="0.0" upY="-1.0"/> + </y:ModelParameter> + </y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + <node id="n1" yfiles.foldertype="group"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d5"/> + <data key="d6"/> + <data key="d7"> + <y:ProxyAutoBoundsNode> + <y:Realizers active="0"> + <y:GroupNode> + <y:Geometry height="250.38333333333333" width="140.0" x="-30.0" y="-330.3833333333333"/> + <y:Fill color="#F5F5F5" transparent="false"/> + <y:BorderStyle color="#000000" type="dashed" width="1.0"/> + <y:NodeLabel alignment="right" autoSizePolicy="node_width" backgroundColor="#EBEBEB" borderDistance="0.0" fontFamily="Dialog" fontSize="15" fontStyle="plain" hasLineColor="false" height="21.4609375" horizontalTextPosition="center" iconTextGap="4" modelName="internal" modelPosition="t" textColor="#000000" verticalTextPosition="bottom" visible="true" width="140.0" x="0.0" y="0.0">Group 3</y:NodeLabel> + <y:Shape type="roundrectangle"/> + <y:State closed="false" closedHeight="50.0" closedWidth="50.0" innerGraphDisplayEnabled="false"/> + <y:Insets bottom="15" bottomF="15.0" left="15" leftF="15.0" right="15" rightF="15.0" top="15" topF="15.0"/> + <y:BorderInsets bottom="1" bottomF="1.0" left="0" leftF="0.0" right="0" rightF="0.0" top="1" topF="1.0001736111111086"/> + </y:GroupNode> + <y:GroupNode> + <y:Geometry height="50.0" width="50.0" x="0.0" y="60.0"/> + <y:Fill color="#F5F5F5" transparent="false"/> + <y:BorderStyle color="#000000" type="dashed" width="1.0"/> + <y:NodeLabel alignment="right" autoSizePolicy="node_width" backgroundColor="#EBEBEB" borderDistance="0.0" fontFamily="Dialog" fontSize="15" fontStyle="plain" hasLineColor="false" height="21.4609375" horizontalTextPosition="center" iconTextGap="4" modelName="internal" modelPosition="t" textColor="#000000" verticalTextPosition="bottom" visible="true" width="65.201171875" x="-7.6005859375" y="0.0">Folder 3</y:NodeLabel> + <y:Shape type="roundrectangle"/> + <y:State closed="true" closedHeight="50.0" closedWidth="50.0" innerGraphDisplayEnabled="false"/> + <y:Insets bottom="5" bottomF="5.0" left="5" leftF="5.0" right="5" rightF="5.0" top="5" topF="5.0"/> + <y:BorderInsets bottom="0" bottomF="0.0" left="0" leftF="0.0" right="0" rightF="0.0" top="0" topF="0.0"/> + </y:GroupNode> + </y:Realizers> + </y:ProxyAutoBoundsNode> + </data> + <graph edgedefault="directed" id="n1:"> + <node id="n1::n0" yfiles.foldertype="group"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d5"/> + <data key="d6"/> + <data key="d7"> + <y:ProxyAutoBoundsNode> + <y:Realizers active="0"> + <y:GroupNode> + <y:Geometry height="83.46111111111111" width="110.0" x="-15.0" y="-292.9222222222222"/> + <y:Fill color="#F5F5F5" transparent="false"/> + <y:BorderStyle color="#000000" type="dashed" width="1.0"/> + <y:NodeLabel alignment="right" autoSizePolicy="node_width" backgroundColor="#EBEBEB" borderDistance="0.0" fontFamily="Dialog" fontSize="15" fontStyle="plain" hasLineColor="false" height="21.4609375" horizontalTextPosition="center" iconTextGap="4" modelName="internal" modelPosition="t" textColor="#000000" verticalTextPosition="bottom" visible="true" width="110.0" x="0.0" y="0.0">Group 1</y:NodeLabel> + <y:Shape type="roundrectangle"/> + <y:State closed="false" closedHeight="50.0" closedWidth="50.0" innerGraphDisplayEnabled="false"/> + <y:Insets bottom="15" bottomF="15.0" left="15" leftF="15.0" right="15" rightF="15.0" top="15" topF="15.0"/> + <y:BorderInsets bottom="1" bottomF="1.0" left="0" leftF="0.0" right="0" rightF="0.0" top="1" topF="1.0001736111111086"/> + </y:GroupNode> + <y:GroupNode> + <y:Geometry height="50.0" width="50.0" x="0.0" y="60.0"/> + <y:Fill color="#F5F5F5" transparent="false"/> + <y:BorderStyle color="#000000" type="dashed" width="1.0"/> + <y:NodeLabel alignment="right" autoSizePolicy="node_width" backgroundColor="#EBEBEB" borderDistance="0.0" fontFamily="Dialog" fontSize="15" fontStyle="plain" hasLineColor="false" height="21.4609375" horizontalTextPosition="center" iconTextGap="4" modelName="internal" modelPosition="t" textColor="#000000" verticalTextPosition="bottom" visible="true" width="65.201171875" x="-7.6005859375" y="0.0">Folder 1</y:NodeLabel> + <y:Shape type="roundrectangle"/> + <y:State closed="true" closedHeight="50.0" closedWidth="50.0" innerGraphDisplayEnabled="false"/> + <y:Insets bottom="5" bottomF="5.0" left="5" leftF="5.0" right="5" rightF="5.0" top="5" topF="5.0"/> + <y:BorderInsets bottom="0" bottomF="0.0" left="0" leftF="0.0" right="0" rightF="0.0" top="0" topF="0.0"/> + </y:GroupNode> + </y:Realizers> + </y:ProxyAutoBoundsNode> + </data> + <graph edgedefault="directed" id="n1::n0:"> + <node id="n1::n0::n0"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d6"/> + <data key="d7"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="50.0" y="-255.4611111111111"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" raised="false" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="17.96875" horizontalTextPosition="center" iconTextGap="4" modelName="custom" textColor="#000000" verticalTextPosition="bottom" visible="true" width="11.634765625" x="9.1826171875" y="6.015625">1<y:LabelModel> + <y:SmartNodeLabelModel distance="4.0"/> + </y:LabelModel> + <y:ModelParameter> + <y:SmartNodeLabelModelParameter labelRatioX="0.0" labelRatioY="0.0" nodeRatioX="0.0" nodeRatioY="0.0" offsetX="0.0" offsetY="0.0" upX="0.0" upY="-1.0"/> + </y:ModelParameter> + </y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + <node id="n1::n0::n1"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d6"/> + <data key="d7"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="0.0" y="-255.4611111111111"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" raised="false" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="17.96875" horizontalTextPosition="center" iconTextGap="4" modelName="custom" textColor="#000000" verticalTextPosition="bottom" visible="true" width="11.634765625" x="9.1826171875" y="6.015625">3<y:LabelModel> + <y:SmartNodeLabelModel distance="4.0"/> + </y:LabelModel> + <y:ModelParameter> + <y:SmartNodeLabelModelParameter labelRatioX="0.0" labelRatioY="0.0" nodeRatioX="0.0" nodeRatioY="0.0" offsetX="0.0" offsetY="0.0" upX="0.0" upY="-1.0"/> + </y:ModelParameter> + </y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + </graph> + </node> + <node id="n1::n1" yfiles.foldertype="group"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d5"/> + <data key="d6"/> + <data key="d7"> + <y:ProxyAutoBoundsNode> + <y:Realizers active="0"> + <y:GroupNode> + <y:Geometry height="83.46111111111111" width="110.0" x="-15.0" y="-179.4611111111111"/> + <y:Fill color="#F5F5F5" transparent="false"/> + <y:BorderStyle color="#000000" type="dashed" width="1.0"/> + <y:NodeLabel alignment="right" autoSizePolicy="node_width" backgroundColor="#EBEBEB" borderDistance="0.0" fontFamily="Dialog" fontSize="15" fontStyle="plain" hasLineColor="false" height="21.4609375" horizontalTextPosition="center" iconTextGap="4" modelName="internal" modelPosition="t" textColor="#000000" verticalTextPosition="bottom" visible="true" width="110.0" x="0.0" y="0.0">Group 2</y:NodeLabel> + <y:Shape type="roundrectangle"/> + <y:State closed="false" closedHeight="50.0" closedWidth="50.0" innerGraphDisplayEnabled="false"/> + <y:Insets bottom="15" bottomF="15.0" left="15" leftF="15.0" right="15" rightF="15.0" top="15" topF="15.0"/> + <y:BorderInsets bottom="1" bottomF="1.0" left="0" leftF="0.0" right="0" rightF="0.0" top="1" topF="1.0001736111111086"/> + </y:GroupNode> + <y:GroupNode> + <y:Geometry height="50.0" width="50.0" x="0.0" y="60.0"/> + <y:Fill color="#F5F5F5" transparent="false"/> + <y:BorderStyle color="#000000" type="dashed" width="1.0"/> + <y:NodeLabel alignment="right" autoSizePolicy="node_width" backgroundColor="#EBEBEB" borderDistance="0.0" fontFamily="Dialog" fontSize="15" fontStyle="plain" hasLineColor="false" height="21.4609375" horizontalTextPosition="center" iconTextGap="4" modelName="internal" modelPosition="t" textColor="#000000" verticalTextPosition="bottom" visible="true" width="65.201171875" x="-7.6005859375" y="0.0">Folder 2</y:NodeLabel> + <y:Shape type="roundrectangle"/> + <y:State closed="true" closedHeight="50.0" closedWidth="50.0" innerGraphDisplayEnabled="false"/> + <y:Insets bottom="5" bottomF="5.0" left="5" leftF="5.0" right="5" rightF="5.0" top="5" topF="5.0"/> + <y:BorderInsets bottom="0" bottomF="0.0" left="0" leftF="0.0" right="0" rightF="0.0" top="0" topF="0.0"/> + </y:GroupNode> + </y:Realizers> + </y:ProxyAutoBoundsNode> + </data> + <graph edgedefault="directed" id="n1::n1:"> + <node id="n1::n1::n0"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d6"/> + <data key="d7"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="0.0" y="-142.0"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" raised="false" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="17.96875" horizontalTextPosition="center" iconTextGap="4" modelName="custom" textColor="#000000" verticalTextPosition="bottom" visible="true" width="11.634765625" x="9.1826171875" y="6.015625">5<y:LabelModel> + <y:SmartNodeLabelModel distance="4.0"/> + </y:LabelModel> + <y:ModelParameter> + <y:SmartNodeLabelModelParameter labelRatioX="0.0" labelRatioY="0.0" nodeRatioX="0.0" nodeRatioY="0.0" offsetX="0.0" offsetY="0.0" upX="0.0" upY="-1.0"/> + </y:ModelParameter> + </y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + <node id="n1::n1::n1"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d6"/> + <data key="d7"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="50.0" y="-142.0"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" raised="false" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="17.96875" horizontalTextPosition="center" iconTextGap="4" modelName="custom" textColor="#000000" verticalTextPosition="bottom" visible="true" width="11.634765625" x="9.1826171875" y="6.015625">6<y:LabelModel> + <y:SmartNodeLabelModel distance="4.0"/> + </y:LabelModel> + <y:ModelParameter> + <y:SmartNodeLabelModelParameter labelRatioX="0.0" labelRatioY="0.0" nodeRatioX="0.0" nodeRatioY="0.0" offsetX="0.0" offsetY="0.0" upX="0.0" upY="-1.0"/> + </y:ModelParameter> + </y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + </graph> + </node> + </graph> + </node> + <node id="n2"> + <data key="d4"><![CDATA[CustomPropertyValue]]></data> + <data key="d6"/> + <data key="d7"> + <y:ShapeNode> + <y:Geometry height="30.0" width="30.0" x="125.0" y="-142.0"/> + <y:Fill color="#FFCC00" transparent="false"/> + <y:BorderStyle color="#000000" raised="false" type="line" width="1.0"/> + <y:NodeLabel alignment="center" autoSizePolicy="content" fontFamily="Dialog" fontSize="12" fontStyle="plain" hasBackgroundColor="false" hasLineColor="false" height="17.96875" horizontalTextPosition="center" iconTextGap="4" modelName="custom" textColor="#000000" verticalTextPosition="bottom" visible="true" width="11.634765625" x="9.1826171875" y="6.015625">9<y:LabelModel> + <y:SmartNodeLabelModel distance="4.0"/> + </y:LabelModel> + <y:ModelParameter> + <y:SmartNodeLabelModelParameter labelRatioX="0.0" labelRatioY="0.0" nodeRatioX="0.0" nodeRatioY="0.0" offsetX="0.0" offsetY="0.0" upX="0.0" upY="-1.0"/> + </y:ModelParameter> + </y:NodeLabel> + <y:Shape type="rectangle"/> + </y:ShapeNode> + </data> + </node> + <edge id="n1::n1::e0" source="n1::n1::n0" target="n1::n1::n1"> + <data key="d10"/> + <data key="d11"> + <y:PolyLineEdge> + <y:Path sx="15.0" sy="-0.0" tx="-15.0" ty="-0.0"/> + <y:LineStyle color="#000000" type="line" width="1.0"/> + <y:Arrows source="none" target="standard"/> + <y:BendStyle smoothed="false"/> + </y:PolyLineEdge> + </data> + </edge> + <edge id="n1::n0::e0" source="n1::n0::n1" target="n1::n0::n0"> + <data key="d10"/> + <data key="d11"> + <y:PolyLineEdge> + <y:Path sx="15.0" sy="-0.0" tx="-15.0" ty="-0.0"/> + <y:LineStyle color="#000000" type="line" width="1.0"/> + <y:Arrows source="none" target="standard"/> + <y:BendStyle smoothed="false"/> + </y:PolyLineEdge> + </data> + </edge> + <edge id="e0" source="n1::n0::n0" target="n0"> + <data key="d10"/> + <data key="d11"> + <y:PolyLineEdge> + <y:Path sx="15.0" sy="-0.0" tx="-15.0" ty="-0.0"/> + <y:LineStyle color="#000000" type="line" width="1.0"/> + <y:Arrows source="none" target="standard"/> + <y:BendStyle smoothed="false"/> + </y:PolyLineEdge> + </data> + </edge> + <edge id="e1" source="n1::n1::n1" target="n2"> + <data key="d10"/> + <data key="d11"> + <y:PolyLineEdge> + <y:Path sx="15.0" sy="-0.0" tx="-15.0" ty="-0.0"/> + <y:LineStyle color="#000000" type="line" width="1.0"/> + <y:Arrows source="none" target="standard"/> + <y:BendStyle smoothed="false"/> + </y:PolyLineEdge> + </data> + </edge> + </graph> + <data key="d8"> + <y:Resources/> + </data> +</graphml> +""" + # verify that nodes / attributes are correctly read when part of a group + fh = io.BytesIO(data.encode("UTF-8")) + G = nx.read_graphml(fh) + data = [x for _, x in G.nodes(data=True)] + assert len(data) == 9 + for node_data in data: + assert node_data["CustomProperty"] != "" + + def test_long_attribute_type(self): + # test that graphs with attr.type="long" (as produced by botch and + # dose3) can be parsed + s = """<?xml version='1.0' encoding='utf-8'?> +<graphml xmlns="http://graphml.graphdrawing.org/xmlns" + xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" + xsi:schemaLocation="http://graphml.graphdrawing.org/xmlns + http://graphml.graphdrawing.org/xmlns/1.0/graphml.xsd"> + <key attr.name="cudfversion" attr.type="long" for="node" id="d6" /> + <graph edgedefault="directed"> + <node id="n1"> + <data key="d6">4284</data> + </node> + </graph> +</graphml>""" + fh = io.BytesIO(s.encode("UTF-8")) + G = nx.read_graphml(fh) + expected = [("n1", {"cudfversion": 4284})] + assert sorted(G.nodes(data=True)) == expected + fh.seek(0) + H = nx.parse_graphml(s) + assert sorted(H.nodes(data=True)) == expected + + +class TestWriteGraphML(BaseGraphML): + writer = staticmethod(nx.write_graphml_lxml) + + @classmethod + def setup_class(cls): + BaseGraphML.setup_class() + _ = pytest.importorskip("lxml.etree") + + def test_write_interface(self): + try: + import lxml.etree + + assert nx.write_graphml == nx.write_graphml_lxml + except ImportError: + assert nx.write_graphml == nx.write_graphml_xml + + def test_write_read_simple_directed_graphml(self): + G = self.simple_directed_graph + G.graph["hi"] = "there" + fh = io.BytesIO() + self.writer(G, fh) + fh.seek(0) + H = nx.read_graphml(fh) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(G.edges()) == sorted(H.edges()) + assert sorted(G.edges(data=True)) == sorted(H.edges(data=True)) + self.simple_directed_fh.seek(0) + + def test_GraphMLWriter_add_graphs(self): + gmlw = GraphMLWriter() + G = self.simple_directed_graph + H = G.copy() + gmlw.add_graphs([G, H]) + + def test_write_read_simple_no_prettyprint(self): + G = self.simple_directed_graph + G.graph["hi"] = "there" + G.graph["id"] = "1" + fh = io.BytesIO() + self.writer(G, fh, prettyprint=False) + fh.seek(0) + H = nx.read_graphml(fh) + assert sorted(G.nodes()) == sorted(H.nodes()) + assert sorted(G.edges()) == sorted(H.edges()) + assert sorted(G.edges(data=True)) == sorted(H.edges(data=True)) + self.simple_directed_fh.seek(0) + + def test_write_read_attribute_named_key_ids_graphml(self): + from xml.etree.ElementTree import parse + + G = self.attribute_named_key_ids_graph + fh = io.BytesIO() + self.writer(G, fh, named_key_ids=True) + fh.seek(0) + H = nx.read_graphml(fh) + fh.seek(0) + + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + assert edges_equal(G.edges(data=True), H.edges(data=True)) + self.attribute_named_key_ids_fh.seek(0) + + xml = parse(fh) + # Children are the key elements, and the graph element + children = list(xml.getroot()) + assert len(children) == 4 + + keys = [child.items() for child in children[:3]] + + assert len(keys) == 3 + assert ("id", "edge_prop") in keys[0] + assert ("attr.name", "edge_prop") in keys[0] + assert ("id", "prop2") in keys[1] + assert ("attr.name", "prop2") in keys[1] + assert ("id", "prop1") in keys[2] + assert ("attr.name", "prop1") in keys[2] + + # Confirm the read graph nodes/edge are identical when compared to + # default writing behavior. + default_behavior_fh = io.BytesIO() + nx.write_graphml(G, default_behavior_fh) + default_behavior_fh.seek(0) + H = nx.read_graphml(default_behavior_fh) + + named_key_ids_behavior_fh = io.BytesIO() + nx.write_graphml(G, named_key_ids_behavior_fh, named_key_ids=True) + named_key_ids_behavior_fh.seek(0) + J = nx.read_graphml(named_key_ids_behavior_fh) + + assert all(n1 == n2 for (n1, n2) in zip(H.nodes, J.nodes)) + assert all(e1 == e2 for (e1, e2) in zip(H.edges, J.edges)) + + def test_write_read_attribute_numeric_type_graphml(self): + from xml.etree.ElementTree import parse + + G = self.attribute_numeric_type_graph + fh = io.BytesIO() + self.writer(G, fh, infer_numeric_types=True) + fh.seek(0) + H = nx.read_graphml(fh) + fh.seek(0) + + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + assert edges_equal(G.edges(data=True), H.edges(data=True)) + self.attribute_numeric_type_fh.seek(0) + + xml = parse(fh) + # Children are the key elements, and the graph element + children = list(xml.getroot()) + assert len(children) == 3 + + keys = [child.items() for child in children[:2]] + + assert len(keys) == 2 + assert ("attr.type", "double") in keys[0] + assert ("attr.type", "double") in keys[1] + + def test_more_multigraph_keys(self, tmp_path): + """Writing keys as edge id attributes means keys become strings. + The original keys are stored as data, so read them back in + if `str(key) == edge_id` + This allows the adjacency to remain the same. + """ + G = nx.MultiGraph() + G.add_edges_from([("a", "b", 2), ("a", "b", 3)]) + fname = tmp_path / "test.graphml" + self.writer(G, fname) + H = nx.read_graphml(fname) + assert H.is_multigraph() + assert edges_equal(G.edges(keys=True), H.edges(keys=True)) + assert G._adj == H._adj + + def test_default_attribute(self): + G = nx.Graph(name="Fred") + G.add_node(1, label=1, color="green") + nx.add_path(G, [0, 1, 2, 3]) + G.add_edge(1, 2, weight=3) + G.graph["node_default"] = {"color": "yellow"} + G.graph["edge_default"] = {"weight": 7} + fh = io.BytesIO() + self.writer(G, fh) + fh.seek(0) + H = nx.read_graphml(fh, node_type=int) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + assert G.graph == H.graph + + def test_mixed_type_attributes(self): + G = nx.MultiGraph() + G.add_node("n0", special=False) + G.add_node("n1", special=0) + G.add_edge("n0", "n1", special=False) + G.add_edge("n0", "n1", special=0) + fh = io.BytesIO() + self.writer(G, fh) + fh.seek(0) + H = nx.read_graphml(fh) + assert not H.nodes["n0"]["special"] + assert H.nodes["n1"]["special"] == 0 + assert not H.edges["n0", "n1", 0]["special"] + assert H.edges["n0", "n1", 1]["special"] == 0 + + def test_str_number_mixed_type_attributes(self): + G = nx.MultiGraph() + G.add_node("n0", special="hello") + G.add_node("n1", special=0) + G.add_edge("n0", "n1", special="hello") + G.add_edge("n0", "n1", special=0) + fh = io.BytesIO() + self.writer(G, fh) + fh.seek(0) + H = nx.read_graphml(fh) + assert H.nodes["n0"]["special"] == "hello" + assert H.nodes["n1"]["special"] == 0 + assert H.edges["n0", "n1", 0]["special"] == "hello" + assert H.edges["n0", "n1", 1]["special"] == 0 + + def test_mixed_int_type_number_attributes(self): + np = pytest.importorskip("numpy") + G = nx.MultiGraph() + G.add_node("n0", special=np.int64(0)) + G.add_node("n1", special=1) + G.add_edge("n0", "n1", special=np.int64(2)) + G.add_edge("n0", "n1", special=3) + fh = io.BytesIO() + self.writer(G, fh) + fh.seek(0) + H = nx.read_graphml(fh) + assert H.nodes["n0"]["special"] == 0 + assert H.nodes["n1"]["special"] == 1 + assert H.edges["n0", "n1", 0]["special"] == 2 + assert H.edges["n0", "n1", 1]["special"] == 3 + + def test_multigraph_to_graph(self, tmp_path): + # test converting multigraph to graph if no parallel edges found + G = nx.MultiGraph() + G.add_edges_from([("a", "b", 2), ("b", "c", 3)]) # no multiedges + fname = tmp_path / "test.graphml" + self.writer(G, fname) + H = nx.read_graphml(fname) + assert not H.is_multigraph() + H = nx.read_graphml(fname, force_multigraph=True) + assert H.is_multigraph() + + # add a multiedge + G.add_edge("a", "b", "e-id") + fname = tmp_path / "test.graphml" + self.writer(G, fname) + H = nx.read_graphml(fname) + assert H.is_multigraph() + H = nx.read_graphml(fname, force_multigraph=True) + assert H.is_multigraph() + + def test_write_generate_edge_id_from_attribute(self, tmp_path): + from xml.etree.ElementTree import parse + + G = nx.Graph() + G.add_edges_from([("a", "b"), ("b", "c"), ("a", "c")]) + edge_attributes = {e: str(e) for e in G.edges} + nx.set_edge_attributes(G, edge_attributes, "eid") + fname = tmp_path / "test.graphml" + # set edge_id_from_attribute e.g. "eid" for write_graphml() + self.writer(G, fname, edge_id_from_attribute="eid") + # set edge_id_from_attribute e.g. "eid" for generate_graphml() + generator = nx.generate_graphml(G, edge_id_from_attribute="eid") + + H = nx.read_graphml(fname) + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + # NetworkX adds explicit edge "id" from file as attribute + nx.set_edge_attributes(G, edge_attributes, "id") + assert edges_equal(G.edges(data=True), H.edges(data=True)) + + tree = parse(fname) + children = list(tree.getroot()) + assert len(children) == 2 + edge_ids = [ + edge.attrib["id"] + for edge in tree.getroot().findall( + ".//{http://graphml.graphdrawing.org/xmlns}edge" + ) + ] + # verify edge id value is equal to specified attribute value + assert sorted(edge_ids) == sorted(edge_attributes.values()) + + # check graphml generated from generate_graphml() + data = "".join(generator) + J = nx.parse_graphml(data) + assert sorted(G.nodes()) == sorted(J.nodes()) + assert sorted(G.edges()) == sorted(J.edges()) + # NetworkX adds explicit edge "id" from file as attribute + nx.set_edge_attributes(G, edge_attributes, "id") + assert edges_equal(G.edges(data=True), J.edges(data=True)) + + def test_multigraph_write_generate_edge_id_from_attribute(self, tmp_path): + from xml.etree.ElementTree import parse + + G = nx.MultiGraph() + G.add_edges_from([("a", "b"), ("b", "c"), ("a", "c"), ("a", "b")]) + edge_attributes = {e: str(e) for e in G.edges} + nx.set_edge_attributes(G, edge_attributes, "eid") + fname = tmp_path / "test.graphml" + # set edge_id_from_attribute e.g. "eid" for write_graphml() + self.writer(G, fname, edge_id_from_attribute="eid") + # set edge_id_from_attribute e.g. "eid" for generate_graphml() + generator = nx.generate_graphml(G, edge_id_from_attribute="eid") + + H = nx.read_graphml(fname) + assert H.is_multigraph() + H = nx.read_graphml(fname, force_multigraph=True) + assert H.is_multigraph() + + assert nodes_equal(G.nodes(), H.nodes()) + assert edges_equal(G.edges(), H.edges()) + assert sorted(data.get("eid") for u, v, data in H.edges(data=True)) == sorted( + edge_attributes.values() + ) + # NetworkX uses edge_ids as keys in multigraphs if no key + assert sorted(key for u, v, key in H.edges(keys=True)) == sorted( + edge_attributes.values() + ) + + tree = parse(fname) + children = list(tree.getroot()) + assert len(children) == 2 + edge_ids = [ + edge.attrib["id"] + for edge in tree.getroot().findall( + ".//{http://graphml.graphdrawing.org/xmlns}edge" + ) + ] + # verify edge id value is equal to specified attribute value + assert sorted(edge_ids) == sorted(edge_attributes.values()) + + # check graphml generated from generate_graphml() + graphml_data = "".join(generator) + J = nx.parse_graphml(graphml_data) + assert J.is_multigraph() + + assert nodes_equal(G.nodes(), J.nodes()) + assert edges_equal(G.edges(), J.edges()) + assert sorted(data.get("eid") for u, v, data in J.edges(data=True)) == sorted( + edge_attributes.values() + ) + # NetworkX uses edge_ids as keys in multigraphs if no key + assert sorted(key for u, v, key in J.edges(keys=True)) == sorted( + edge_attributes.values() + ) + + def test_numpy_float64(self, tmp_path): + np = pytest.importorskip("numpy") + wt = np.float64(3.4) + G = nx.Graph([(1, 2, {"weight": wt})]) + fname = tmp_path / "test.graphml" + self.writer(G, fname) + H = nx.read_graphml(fname, node_type=int) + assert G.edges == H.edges + wtG = G[1][2]["weight"] + wtH = H[1][2]["weight"] + assert wtG == pytest.approx(wtH, abs=1e-6) + assert type(wtG) == np.float64 + assert type(wtH) == float + + def test_numpy_float32(self, tmp_path): + np = pytest.importorskip("numpy") + wt = np.float32(3.4) + G = nx.Graph([(1, 2, {"weight": wt})]) + fname = tmp_path / "test.graphml" + self.writer(G, fname) + H = nx.read_graphml(fname, node_type=int) + assert G.edges == H.edges + wtG = G[1][2]["weight"] + wtH = H[1][2]["weight"] + assert wtG == pytest.approx(wtH, abs=1e-6) + assert type(wtG) == np.float32 + assert type(wtH) == float + + def test_numpy_float64_inference(self, tmp_path): + np = pytest.importorskip("numpy") + G = self.attribute_numeric_type_graph + G.edges[("n1", "n1")]["weight"] = np.float64(1.1) + fname = tmp_path / "test.graphml" + self.writer(G, fname, infer_numeric_types=True) + H = nx.read_graphml(fname) + assert G._adj == H._adj + + def test_unicode_attributes(self, tmp_path): + G = nx.Graph() + name1 = chr(2344) + chr(123) + chr(6543) + name2 = chr(5543) + chr(1543) + chr(324) + node_type = str + G.add_edge(name1, "Radiohead", foo=name2) + fname = tmp_path / "test.graphml" + self.writer(G, fname) + H = nx.read_graphml(fname, node_type=node_type) + assert G._adj == H._adj + + def test_unicode_escape(self): + # test for handling json escaped strings in python 2 Issue #1880 + import json + + a = {"a": '{"a": "123"}'} # an object with many chars to escape + sa = json.dumps(a) + G = nx.Graph() + G.graph["test"] = sa + fh = io.BytesIO() + self.writer(G, fh) + fh.seek(0) + H = nx.read_graphml(fh) + assert G.graph["test"] == H.graph["test"] + + +class TestXMLGraphML(TestWriteGraphML): + writer = staticmethod(nx.write_graphml_xml) + + @classmethod + def setup_class(cls): + TestWriteGraphML.setup_class() + + +def test_exception_for_unsupported_datatype_node_attr(): + """Test that a detailed exception is raised when an attribute is of a type + not supported by GraphML, e.g. a list""" + pytest.importorskip("lxml.etree") + # node attribute + G = nx.Graph() + G.add_node(0, my_list_attribute=[0, 1, 2]) + fh = io.BytesIO() + with pytest.raises(TypeError, match="GraphML does not support"): + nx.write_graphml(G, fh) + + +def test_exception_for_unsupported_datatype_edge_attr(): + """Test that a detailed exception is raised when an attribute is of a type + not supported by GraphML, e.g. a list""" + pytest.importorskip("lxml.etree") + # edge attribute + G = nx.Graph() + G.add_edge(0, 1, my_list_attribute=[0, 1, 2]) + fh = io.BytesIO() + with pytest.raises(TypeError, match="GraphML does not support"): + nx.write_graphml(G, fh) + + +def test_exception_for_unsupported_datatype_graph_attr(): + """Test that a detailed exception is raised when an attribute is of a type + not supported by GraphML, e.g. a list""" + pytest.importorskip("lxml.etree") + # graph attribute + G = nx.Graph() + G.graph["my_list_attribute"] = [0, 1, 2] + fh = io.BytesIO() + with pytest.raises(TypeError, match="GraphML does not support"): + nx.write_graphml(G, fh) + + +def test_empty_attribute(): + """Tests that a GraphML string with an empty attribute can be parsed + correctly.""" + s = """<?xml version='1.0' encoding='utf-8'?> + <graphml> + <key id="d1" for="node" attr.name="foo" attr.type="string"/> + <key id="d2" for="node" attr.name="bar" attr.type="string"/> + <graph> + <node id="0"> + <data key="d1">aaa</data> + <data key="d2">bbb</data> + </node> + <node id="1"> + <data key="d1">ccc</data> + <data key="d2"></data> + </node> + </graph> + </graphml>""" + fh = io.BytesIO(s.encode("UTF-8")) + G = nx.read_graphml(fh) + assert G.nodes["0"] == {"foo": "aaa", "bar": "bbb"} + assert G.nodes["1"] == {"foo": "ccc", "bar": ""} diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_leda.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_leda.py new file mode 100644 index 00000000..8ac5ecc3 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_leda.py @@ -0,0 +1,30 @@ +import io + +import networkx as nx + + +class TestLEDA: + def test_parse_leda(self): + data = """#header section \nLEDA.GRAPH \nstring\nint\n-1\n#nodes section\n5 \n|{v1}| \n|{v2}| \n|{v3}| \n|{v4}| \n|{v5}| \n\n#edges section\n7 \n1 2 0 |{4}| \n1 3 0 |{3}| \n2 3 0 |{2}| \n3 4 0 |{3}| \n3 5 0 |{7}| \n4 5 0 |{6}| \n5 1 0 |{foo}|""" + G = nx.parse_leda(data) + G = nx.parse_leda(data.split("\n")) + assert sorted(G.nodes()) == ["v1", "v2", "v3", "v4", "v5"] + assert sorted(G.edges(data=True)) == [ + ("v1", "v2", {"label": "4"}), + ("v1", "v3", {"label": "3"}), + ("v2", "v3", {"label": "2"}), + ("v3", "v4", {"label": "3"}), + ("v3", "v5", {"label": "7"}), + ("v4", "v5", {"label": "6"}), + ("v5", "v1", {"label": "foo"}), + ] + + def test_read_LEDA(self): + fh = io.BytesIO() + data = """#header section \nLEDA.GRAPH \nstring\nint\n-1\n#nodes section\n5 \n|{v1}| \n|{v2}| \n|{v3}| \n|{v4}| \n|{v5}| \n\n#edges section\n7 \n1 2 0 |{4}| \n1 3 0 |{3}| \n2 3 0 |{2}| \n3 4 0 |{3}| \n3 5 0 |{7}| \n4 5 0 |{6}| \n5 1 0 |{foo}|""" + G = nx.parse_leda(data) + fh.write(data.encode("UTF-8")) + fh.seek(0) + Gin = nx.read_leda(fh) + assert sorted(G.nodes()) == sorted(Gin.nodes()) + assert sorted(G.edges()) == sorted(Gin.edges()) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_p2g.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_p2g.py new file mode 100644 index 00000000..e4c50de7 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_p2g.py @@ -0,0 +1,62 @@ +import io + +import networkx as nx +from networkx.readwrite.p2g import read_p2g, write_p2g +from networkx.utils import edges_equal + + +class TestP2G: + @classmethod + def setup_class(cls): + cls.G = nx.Graph(name="test") + e = [("a", "b"), ("b", "c"), ("c", "d"), ("d", "e"), ("e", "f"), ("a", "f")] + cls.G.add_edges_from(e) + cls.G.add_node("g") + cls.DG = nx.DiGraph(cls.G) + + def test_read_p2g(self): + s = b"""\ +name +3 4 +a +1 2 +b + +c +0 2 +""" + bytesIO = io.BytesIO(s) + G = read_p2g(bytesIO) + assert G.name == "name" + assert sorted(G) == ["a", "b", "c"] + edges = [(str(u), str(v)) for u, v in G.edges()] + assert edges_equal(G.edges(), [("a", "c"), ("a", "b"), ("c", "a"), ("c", "c")]) + + def test_write_p2g(self): + s = b"""foo +3 2 +1 +1 +2 +2 +3 + +""" + fh = io.BytesIO() + G = nx.DiGraph() + G.name = "foo" + G.add_edges_from([(1, 2), (2, 3)]) + write_p2g(G, fh) + fh.seek(0) + r = fh.read() + assert r == s + + def test_write_read_p2g(self): + fh = io.BytesIO() + G = nx.DiGraph() + G.name = "foo" + G.add_edges_from([("a", "b"), ("b", "c")]) + write_p2g(G, fh) + fh.seek(0) + H = read_p2g(fh) + assert edges_equal(G.edges(), H.edges()) diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_pajek.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_pajek.py new file mode 100644 index 00000000..317ebe8e --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_pajek.py @@ -0,0 +1,126 @@ +""" +Pajek tests +""" + +import networkx as nx +from networkx.utils import edges_equal, nodes_equal + + +class TestPajek: + @classmethod + def setup_class(cls): + cls.data = """*network Tralala\n*vertices 4\n 1 "A1" 0.0938 0.0896 ellipse x_fact 1 y_fact 1\n 2 "Bb" 0.8188 0.2458 ellipse x_fact 1 y_fact 1\n 3 "C" 0.3688 0.7792 ellipse x_fact 1\n 4 "D2" 0.9583 0.8563 ellipse x_fact 1\n*arcs\n1 1 1 h2 0 w 3 c Blue s 3 a1 -130 k1 0.6 a2 -130 k2 0.6 ap 0.5 l "Bezier loop" lc BlueViolet fos 20 lr 58 lp 0.3 la 360\n2 1 1 h2 0 a1 120 k1 1.3 a2 -120 k2 0.3 ap 25 l "Bezier arc" lphi 270 la 180 lr 19 lp 0.5\n1 2 1 h2 0 a1 40 k1 2.8 a2 30 k2 0.8 ap 25 l "Bezier arc" lphi 90 la 0 lp 0.65\n4 2 -1 h2 0 w 1 k1 -2 k2 250 ap 25 l "Circular arc" c Red lc OrangeRed\n3 4 1 p Dashed h2 0 w 2 c OliveGreen ap 25 l "Straight arc" lc PineGreen\n1 3 1 p Dashed h2 0 w 5 k1 -1 k2 -20 ap 25 l "Oval arc" c Brown lc Black\n3 3 -1 h1 6 w 1 h2 12 k1 -2 k2 -15 ap 0.5 l "Circular loop" c Red lc OrangeRed lphi 270 la 180""" + cls.G = nx.MultiDiGraph() + cls.G.add_nodes_from(["A1", "Bb", "C", "D2"]) + cls.G.add_edges_from( + [ + ("A1", "A1"), + ("A1", "Bb"), + ("A1", "C"), + ("Bb", "A1"), + ("C", "C"), + ("C", "D2"), + ("D2", "Bb"), + ] + ) + + cls.G.graph["name"] = "Tralala" + + def test_parse_pajek_simple(self): + # Example without node positions or shape + data = """*Vertices 2\n1 "1"\n2 "2"\n*Edges\n1 2\n2 1""" + G = nx.parse_pajek(data) + assert sorted(G.nodes()) == ["1", "2"] + assert edges_equal(G.edges(), [("1", "2"), ("1", "2")]) + + def test_parse_pajek(self): + G = nx.parse_pajek(self.data) + assert sorted(G.nodes()) == ["A1", "Bb", "C", "D2"] + assert edges_equal( + G.edges(), + [ + ("A1", "A1"), + ("A1", "Bb"), + ("A1", "C"), + ("Bb", "A1"), + ("C", "C"), + ("C", "D2"), + ("D2", "Bb"), + ], + ) + + def test_parse_pajet_mat(self): + data = """*Vertices 3\n1 "one"\n2 "two"\n3 "three"\n*Matrix\n1 1 0\n0 1 0\n0 1 0\n""" + G = nx.parse_pajek(data) + assert set(G.nodes()) == {"one", "two", "three"} + assert G.nodes["two"] == {"id": "2"} + assert edges_equal( + set(G.edges()), + {("one", "one"), ("two", "one"), ("two", "two"), ("two", "three")}, + ) + + def test_read_pajek(self, tmp_path): + G = nx.parse_pajek(self.data) + # Read data from file + fname = tmp_path / "test.pjk" + with open(fname, "wb") as fh: + fh.write(self.data.encode("UTF-8")) + + Gin = nx.read_pajek(fname) + assert sorted(G.nodes()) == sorted(Gin.nodes()) + assert edges_equal(G.edges(), Gin.edges()) + assert self.G.graph == Gin.graph + for n in G: + assert G.nodes[n] == Gin.nodes[n] + + def test_write_pajek(self): + import io + + G = nx.parse_pajek(self.data) + fh = io.BytesIO() + nx.write_pajek(G, fh) + fh.seek(0) + H = nx.read_pajek(fh) + assert nodes_equal(list(G), list(H)) + assert edges_equal(list(G.edges()), list(H.edges())) + # Graph name is left out for now, therefore it is not tested. + # assert_equal(G.graph, H.graph) + + def test_ignored_attribute(self): + import io + + G = nx.Graph() + fh = io.BytesIO() + G.add_node(1, int_attr=1) + G.add_node(2, empty_attr=" ") + G.add_edge(1, 2, int_attr=2) + G.add_edge(2, 3, empty_attr=" ") + + import warnings + + with warnings.catch_warnings(record=True) as w: + nx.write_pajek(G, fh) + assert len(w) == 4 + + def test_noname(self): + # Make sure we can parse a line such as: *network + # Issue #952 + line = "*network\n" + other_lines = self.data.split("\n")[1:] + data = line + "\n".join(other_lines) + G = nx.parse_pajek(data) + + def test_unicode(self): + import io + + G = nx.Graph() + name1 = chr(2344) + chr(123) + chr(6543) + name2 = chr(5543) + chr(1543) + chr(324) + G.add_edge(name1, "Radiohead", foo=name2) + fh = io.BytesIO() + nx.write_pajek(G, fh) + fh.seek(0) + H = nx.read_pajek(fh) + assert nodes_equal(list(G), list(H)) + assert edges_equal(list(G.edges()), list(H.edges())) + assert G.graph == H.graph diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_sparse6.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_sparse6.py new file mode 100644 index 00000000..344ad0e4 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_sparse6.py @@ -0,0 +1,166 @@ +from io import BytesIO + +import pytest + +import networkx as nx +from networkx.utils import edges_equal, nodes_equal + + +class TestSparseGraph6: + def test_from_sparse6_bytes(self): + data = b":Q___eDcdFcDeFcE`GaJ`IaHbKNbLM" + G = nx.from_sparse6_bytes(data) + assert nodes_equal( + sorted(G.nodes()), + [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17], + ) + assert edges_equal( + G.edges(), + [ + (0, 1), + (0, 2), + (0, 3), + (1, 12), + (1, 14), + (2, 13), + (2, 15), + (3, 16), + (3, 17), + (4, 7), + (4, 9), + (4, 11), + (5, 6), + (5, 8), + (5, 9), + (6, 10), + (6, 11), + (7, 8), + (7, 10), + (8, 12), + (9, 15), + (10, 14), + (11, 13), + (12, 16), + (13, 17), + (14, 17), + (15, 16), + ], + ) + + def test_from_bytes_multigraph_graph(self): + graph_data = b":An" + G = nx.from_sparse6_bytes(graph_data) + assert type(G) == nx.Graph + multigraph_data = b":Ab" + M = nx.from_sparse6_bytes(multigraph_data) + assert type(M) == nx.MultiGraph + + def test_read_sparse6(self): + data = b":Q___eDcdFcDeFcE`GaJ`IaHbKNbLM" + G = nx.from_sparse6_bytes(data) + fh = BytesIO(data) + Gin = nx.read_sparse6(fh) + assert nodes_equal(G.nodes(), Gin.nodes()) + assert edges_equal(G.edges(), Gin.edges()) + + def test_read_many_graph6(self): + # Read many graphs into list + data = b":Q___eDcdFcDeFcE`GaJ`IaHbKNbLM\n" b":Q___dCfDEdcEgcbEGbFIaJ`JaHN`IM" + fh = BytesIO(data) + glist = nx.read_sparse6(fh) + assert len(glist) == 2 + for G in glist: + assert nodes_equal( + G.nodes(), + [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17], + ) + + +class TestWriteSparse6: + """Unit tests for writing graphs in the sparse6 format. + + Most of the test cases were checked against the sparse6 encoder in Sage. + + """ + + def test_null_graph(self): + G = nx.null_graph() + result = BytesIO() + nx.write_sparse6(G, result) + assert result.getvalue() == b">>sparse6<<:?\n" + + def test_trivial_graph(self): + G = nx.trivial_graph() + result = BytesIO() + nx.write_sparse6(G, result) + assert result.getvalue() == b">>sparse6<<:@\n" + + def test_empty_graph(self): + G = nx.empty_graph(5) + result = BytesIO() + nx.write_sparse6(G, result) + assert result.getvalue() == b">>sparse6<<:D\n" + + def test_large_empty_graph(self): + G = nx.empty_graph(68) + result = BytesIO() + nx.write_sparse6(G, result) + assert result.getvalue() == b">>sparse6<<:~?@C\n" + + def test_very_large_empty_graph(self): + G = nx.empty_graph(258049) + result = BytesIO() + nx.write_sparse6(G, result) + assert result.getvalue() == b">>sparse6<<:~~???~?@\n" + + def test_complete_graph(self): + G = nx.complete_graph(4) + result = BytesIO() + nx.write_sparse6(G, result) + assert result.getvalue() == b">>sparse6<<:CcKI\n" + + def test_no_header(self): + G = nx.complete_graph(4) + result = BytesIO() + nx.write_sparse6(G, result, header=False) + assert result.getvalue() == b":CcKI\n" + + def test_padding(self): + codes = (b":Cdv", b":DaYn", b":EaYnN", b":FaYnL", b":GaYnLz") + for n, code in enumerate(codes, start=4): + G = nx.path_graph(n) + result = BytesIO() + nx.write_sparse6(G, result, header=False) + assert result.getvalue() == code + b"\n" + + def test_complete_bipartite(self): + G = nx.complete_bipartite_graph(6, 9) + result = BytesIO() + nx.write_sparse6(G, result) + # Compared with sage + expected = b">>sparse6<<:Nk" + b"?G`cJ" * 9 + b"\n" + assert result.getvalue() == expected + + def test_read_write_inverse(self): + for i in list(range(13)) + [31, 47, 62, 63, 64, 72]: + m = min(2 * i, i * i // 2) + g = nx.random_graphs.gnm_random_graph(i, m, seed=i) + gstr = BytesIO() + nx.write_sparse6(g, gstr, header=False) + # Strip the trailing newline. + gstr = gstr.getvalue().rstrip() + g2 = nx.from_sparse6_bytes(gstr) + assert g2.order() == g.order() + assert edges_equal(g2.edges(), g.edges()) + + def test_no_directed_graphs(self): + with pytest.raises(nx.NetworkXNotImplemented): + nx.write_sparse6(nx.DiGraph(), BytesIO()) + + def test_write_path(self, tmp_path): + # Get a valid temporary file name + fullfilename = str(tmp_path / "test.s6") + # file should be closed now, so write_sparse6 can open it + nx.write_sparse6(nx.null_graph(), fullfilename) + with open(fullfilename, mode="rb") as fh: + assert fh.read() == b">>sparse6<<:?\n" diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_text.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_text.py new file mode 100644 index 00000000..b2b74482 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/tests/test_text.py @@ -0,0 +1,1742 @@ +import random +from itertools import product +from textwrap import dedent + +import pytest + +import networkx as nx + + +def test_generate_network_text_forest_directed(): + # Create a directed forest with labels + graph = nx.balanced_tree(r=2, h=2, create_using=nx.DiGraph) + for node in graph.nodes: + graph.nodes[node]["label"] = "node_" + chr(ord("a") + node) + + node_target = dedent( + """ + ╙── 0 + ├─╼ 1 + │ ├─╼ 3 + │ └─╼ 4 + └─╼ 2 + ├─╼ 5 + └─╼ 6 + """ + ).strip() + + label_target = dedent( + """ + ╙── node_a + ├─╼ node_b + │ ├─╼ node_d + │ └─╼ node_e + └─╼ node_c + ├─╼ node_f + └─╼ node_g + """ + ).strip() + + # Basic node case + ret = nx.generate_network_text(graph, with_labels=False) + assert "\n".join(ret) == node_target + + # Basic label case + ret = nx.generate_network_text(graph, with_labels=True) + assert "\n".join(ret) == label_target + + +def test_write_network_text_empty_graph(): + def _graph_str(g, **kw): + printbuf = [] + nx.write_network_text(g, printbuf.append, end="", **kw) + return "\n".join(printbuf) + + assert _graph_str(nx.DiGraph()) == "╙" + assert _graph_str(nx.Graph()) == "╙" + assert _graph_str(nx.DiGraph(), ascii_only=True) == "+" + assert _graph_str(nx.Graph(), ascii_only=True) == "+" + + +def test_write_network_text_within_forest_glyph(): + g = nx.DiGraph() + g.add_nodes_from([1, 2, 3, 4]) + g.add_edge(2, 4) + lines = [] + write = lines.append + nx.write_network_text(g, path=write, end="") + nx.write_network_text(g, path=write, ascii_only=True, end="") + text = "\n".join(lines) + target = dedent( + """ + ╟── 1 + ╟── 2 + ╎ └─╼ 4 + ╙── 3 + +-- 1 + +-- 2 + : L-> 4 + +-- 3 + """ + ).strip() + assert text == target + + +def test_generate_network_text_directed_multi_tree(): + tree1 = nx.balanced_tree(r=2, h=2, create_using=nx.DiGraph) + tree2 = nx.balanced_tree(r=2, h=2, create_using=nx.DiGraph) + forest = nx.disjoint_union_all([tree1, tree2]) + ret = "\n".join(nx.generate_network_text(forest)) + + target = dedent( + """ + ╟── 0 + ╎ ├─╼ 1 + ╎ │ ├─╼ 3 + ╎ │ └─╼ 4 + ╎ └─╼ 2 + ╎ ├─╼ 5 + ╎ └─╼ 6 + ╙── 7 + ├─╼ 8 + │ ├─╼ 10 + │ └─╼ 11 + └─╼ 9 + ├─╼ 12 + └─╼ 13 + """ + ).strip() + assert ret == target + + tree3 = nx.balanced_tree(r=2, h=2, create_using=nx.DiGraph) + forest = nx.disjoint_union_all([tree1, tree2, tree3]) + ret = "\n".join(nx.generate_network_text(forest, sources=[0, 14, 7])) + + target = dedent( + """ + ╟── 0 + ╎ ├─╼ 1 + ╎ │ ├─╼ 3 + ╎ │ └─╼ 4 + ╎ └─╼ 2 + ╎ ├─╼ 5 + ╎ └─╼ 6 + ╟── 14 + ╎ ├─╼ 15 + ╎ │ ├─╼ 17 + ╎ │ └─╼ 18 + ╎ └─╼ 16 + ╎ ├─╼ 19 + ╎ └─╼ 20 + ╙── 7 + ├─╼ 8 + │ ├─╼ 10 + │ └─╼ 11 + └─╼ 9 + ├─╼ 12 + └─╼ 13 + """ + ).strip() + assert ret == target + + ret = "\n".join( + nx.generate_network_text(forest, sources=[0, 14, 7], ascii_only=True) + ) + + target = dedent( + """ + +-- 0 + : |-> 1 + : | |-> 3 + : | L-> 4 + : L-> 2 + : |-> 5 + : L-> 6 + +-- 14 + : |-> 15 + : | |-> 17 + : | L-> 18 + : L-> 16 + : |-> 19 + : L-> 20 + +-- 7 + |-> 8 + | |-> 10 + | L-> 11 + L-> 9 + |-> 12 + L-> 13 + """ + ).strip() + assert ret == target + + +def test_generate_network_text_undirected_multi_tree(): + tree1 = nx.balanced_tree(r=2, h=2, create_using=nx.Graph) + tree2 = nx.balanced_tree(r=2, h=2, create_using=nx.Graph) + tree2 = nx.relabel_nodes(tree2, {n: n + len(tree1) for n in tree2.nodes}) + forest = nx.union(tree1, tree2) + ret = "\n".join(nx.generate_network_text(forest, sources=[0, 7])) + + target = dedent( + """ + ╟── 0 + ╎ ├── 1 + ╎ │ ├── 3 + ╎ │ └── 4 + ╎ └── 2 + ╎ ├── 5 + ╎ └── 6 + ╙── 7 + ├── 8 + │ ├── 10 + │ └── 11 + └── 9 + ├── 12 + └── 13 + """ + ).strip() + assert ret == target + + ret = "\n".join(nx.generate_network_text(forest, sources=[0, 7], ascii_only=True)) + + target = dedent( + """ + +-- 0 + : |-- 1 + : | |-- 3 + : | L-- 4 + : L-- 2 + : |-- 5 + : L-- 6 + +-- 7 + |-- 8 + | |-- 10 + | L-- 11 + L-- 9 + |-- 12 + L-- 13 + """ + ).strip() + assert ret == target + + +def test_generate_network_text_forest_undirected(): + # Create a directed forest + graph = nx.balanced_tree(r=2, h=2, create_using=nx.Graph) + + node_target0 = dedent( + """ + ╙── 0 + ├── 1 + │ ├── 3 + │ └── 4 + └── 2 + ├── 5 + └── 6 + """ + ).strip() + + # defined starting point + ret = "\n".join(nx.generate_network_text(graph, sources=[0])) + assert ret == node_target0 + + # defined starting point + node_target2 = dedent( + """ + ╙── 2 + ├── 0 + │ └── 1 + │ ├── 3 + │ └── 4 + ├── 5 + └── 6 + """ + ).strip() + ret = "\n".join(nx.generate_network_text(graph, sources=[2])) + assert ret == node_target2 + + +def test_generate_network_text_overspecified_sources(): + """ + When sources are directly specified, we won't be able to determine when we + are in the last component, so there will always be a trailing, leftmost + pipe. + """ + graph = nx.disjoint_union_all( + [ + nx.balanced_tree(r=2, h=1, create_using=nx.DiGraph), + nx.balanced_tree(r=1, h=2, create_using=nx.DiGraph), + nx.balanced_tree(r=2, h=1, create_using=nx.DiGraph), + ] + ) + + # defined starting point + target1 = dedent( + """ + ╟── 0 + ╎ ├─╼ 1 + ╎ └─╼ 2 + ╟── 3 + ╎ └─╼ 4 + ╎ └─╼ 5 + ╟── 6 + ╎ ├─╼ 7 + ╎ └─╼ 8 + """ + ).strip() + + target2 = dedent( + """ + ╟── 0 + ╎ ├─╼ 1 + ╎ └─╼ 2 + ╟── 3 + ╎ └─╼ 4 + ╎ └─╼ 5 + ╙── 6 + ├─╼ 7 + └─╼ 8 + """ + ).strip() + + got1 = "\n".join(nx.generate_network_text(graph, sources=graph.nodes)) + got2 = "\n".join(nx.generate_network_text(graph)) + assert got1 == target1 + assert got2 == target2 + + +def test_write_network_text_iterative_add_directed_edges(): + """ + Walk through the cases going from a disconnected to fully connected graph + """ + graph = nx.DiGraph() + graph.add_nodes_from([1, 2, 3, 4]) + lines = [] + write = lines.append + write("--- initial state ---") + nx.write_network_text(graph, path=write, end="") + for i, j in product(graph.nodes, graph.nodes): + write(f"--- add_edge({i}, {j}) ---") + graph.add_edge(i, j) + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + # defined starting point + target = dedent( + """ + --- initial state --- + ╟── 1 + ╟── 2 + ╟── 3 + ╙── 4 + --- add_edge(1, 1) --- + ╟── 1 ╾ 1 + ╎ └─╼ ... + ╟── 2 + ╟── 3 + ╙── 4 + --- add_edge(1, 2) --- + ╟── 1 ╾ 1 + ╎ ├─╼ 2 + ╎ └─╼ ... + ╟── 3 + ╙── 4 + --- add_edge(1, 3) --- + ╟── 1 ╾ 1 + ╎ ├─╼ 2 + ╎ ├─╼ 3 + ╎ └─╼ ... + ╙── 4 + --- add_edge(1, 4) --- + ╙── 1 ╾ 1 + ├─╼ 2 + ├─╼ 3 + ├─╼ 4 + └─╼ ... + --- add_edge(2, 1) --- + ╙── 2 ╾ 1 + └─╼ 1 ╾ 1 + ├─╼ 3 + ├─╼ 4 + └─╼ ... + --- add_edge(2, 2) --- + ╙── 1 ╾ 1, 2 + ├─╼ 2 ╾ 2 + │ └─╼ ... + ├─╼ 3 + ├─╼ 4 + └─╼ ... + --- add_edge(2, 3) --- + ╙── 1 ╾ 1, 2 + ├─╼ 2 ╾ 2 + │ ├─╼ 3 ╾ 1 + │ └─╼ ... + ├─╼ 4 + └─╼ ... + --- add_edge(2, 4) --- + ╙── 1 ╾ 1, 2 + ├─╼ 2 ╾ 2 + │ ├─╼ 3 ╾ 1 + │ ├─╼ 4 ╾ 1 + │ └─╼ ... + └─╼ ... + --- add_edge(3, 1) --- + ╙── 2 ╾ 1, 2 + ├─╼ 1 ╾ 1, 3 + │ ├─╼ 3 ╾ 2 + │ │ └─╼ ... + │ ├─╼ 4 ╾ 2 + │ └─╼ ... + └─╼ ... + --- add_edge(3, 2) --- + ╙── 3 ╾ 1, 2 + ├─╼ 1 ╾ 1, 2 + │ ├─╼ 2 ╾ 2, 3 + │ │ ├─╼ 4 ╾ 1 + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(3, 3) --- + ╙── 1 ╾ 1, 2, 3 + ├─╼ 2 ╾ 2, 3 + │ ├─╼ 3 ╾ 1, 3 + │ │ └─╼ ... + │ ├─╼ 4 ╾ 1 + │ └─╼ ... + └─╼ ... + --- add_edge(3, 4) --- + ╙── 1 ╾ 1, 2, 3 + ├─╼ 2 ╾ 2, 3 + │ ├─╼ 3 ╾ 1, 3 + │ │ ├─╼ 4 ╾ 1, 2 + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(4, 1) --- + ╙── 2 ╾ 1, 2, 3 + ├─╼ 1 ╾ 1, 3, 4 + │ ├─╼ 3 ╾ 2, 3 + │ │ ├─╼ 4 ╾ 1, 2 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(4, 2) --- + ╙── 3 ╾ 1, 2, 3 + ├─╼ 1 ╾ 1, 2, 4 + │ ├─╼ 2 ╾ 2, 3, 4 + │ │ ├─╼ 4 ╾ 1, 3 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(4, 3) --- + ╙── 4 ╾ 1, 2, 3 + ├─╼ 1 ╾ 1, 2, 3 + │ ├─╼ 2 ╾ 2, 3, 4 + │ │ ├─╼ 3 ╾ 1, 3, 4 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(4, 4) --- + ╙── 1 ╾ 1, 2, 3, 4 + ├─╼ 2 ╾ 2, 3, 4 + │ ├─╼ 3 ╾ 1, 3, 4 + │ │ ├─╼ 4 ╾ 1, 2, 4 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + """ + ).strip() + assert target == text + + +def test_write_network_text_iterative_add_undirected_edges(): + """ + Walk through the cases going from a disconnected to fully connected graph + """ + graph = nx.Graph() + graph.add_nodes_from([1, 2, 3, 4]) + lines = [] + write = lines.append + write("--- initial state ---") + nx.write_network_text(graph, path=write, end="") + for i, j in product(graph.nodes, graph.nodes): + if i == j: + continue + write(f"--- add_edge({i}, {j}) ---") + graph.add_edge(i, j) + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + --- initial state --- + ╟── 1 + ╟── 2 + ╟── 3 + ╙── 4 + --- add_edge(1, 2) --- + ╟── 3 + ╟── 4 + ╙── 1 + └── 2 + --- add_edge(1, 3) --- + ╟── 4 + ╙── 2 + └── 1 + └── 3 + --- add_edge(1, 4) --- + ╙── 2 + └── 1 + ├── 3 + └── 4 + --- add_edge(2, 1) --- + ╙── 2 + └── 1 + ├── 3 + └── 4 + --- add_edge(2, 3) --- + ╙── 4 + └── 1 + ├── 2 + │ └── 3 ─ 1 + └── ... + --- add_edge(2, 4) --- + ╙── 3 + ├── 1 + │ ├── 2 ─ 3 + │ │ └── 4 ─ 1 + │ └── ... + └── ... + --- add_edge(3, 1) --- + ╙── 3 + ├── 1 + │ ├── 2 ─ 3 + │ │ └── 4 ─ 1 + │ └── ... + └── ... + --- add_edge(3, 2) --- + ╙── 3 + ├── 1 + │ ├── 2 ─ 3 + │ │ └── 4 ─ 1 + │ └── ... + └── ... + --- add_edge(3, 4) --- + ╙── 1 + ├── 2 + │ ├── 3 ─ 1 + │ │ └── 4 ─ 1, 2 + │ └── ... + └── ... + --- add_edge(4, 1) --- + ╙── 1 + ├── 2 + │ ├── 3 ─ 1 + │ │ └── 4 ─ 1, 2 + │ └── ... + └── ... + --- add_edge(4, 2) --- + ╙── 1 + ├── 2 + │ ├── 3 ─ 1 + │ │ └── 4 ─ 1, 2 + │ └── ... + └── ... + --- add_edge(4, 3) --- + ╙── 1 + ├── 2 + │ ├── 3 ─ 1 + │ │ └── 4 ─ 1, 2 + │ └── ... + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_iterative_add_random_directed_edges(): + """ + Walk through the cases going from a disconnected to fully connected graph + """ + + rng = random.Random(724466096) + graph = nx.DiGraph() + graph.add_nodes_from([1, 2, 3, 4, 5]) + possible_edges = list(product(graph.nodes, graph.nodes)) + rng.shuffle(possible_edges) + graph.add_edges_from(possible_edges[0:8]) + lines = [] + write = lines.append + write("--- initial state ---") + nx.write_network_text(graph, path=write, end="") + for i, j in possible_edges[8:12]: + write(f"--- add_edge({i}, {j}) ---") + graph.add_edge(i, j) + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + --- initial state --- + ╙── 3 ╾ 5 + └─╼ 2 ╾ 2 + ├─╼ 4 ╾ 4 + │ ├─╼ 5 + │ │ ├─╼ 1 ╾ 1 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(4, 1) --- + ╙── 3 ╾ 5 + └─╼ 2 ╾ 2 + ├─╼ 4 ╾ 4 + │ ├─╼ 5 + │ │ ├─╼ 1 ╾ 1, 4 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(2, 1) --- + ╙── 3 ╾ 5 + └─╼ 2 ╾ 2 + ├─╼ 4 ╾ 4 + │ ├─╼ 5 + │ │ ├─╼ 1 ╾ 1, 4, 2 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(5, 2) --- + ╙── 3 ╾ 5 + └─╼ 2 ╾ 2, 5 + ├─╼ 4 ╾ 4 + │ ├─╼ 5 + │ │ ├─╼ 1 ╾ 1, 4, 2 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- add_edge(1, 5) --- + ╙── 3 ╾ 5 + └─╼ 2 ╾ 2, 5 + ├─╼ 4 ╾ 4 + │ ├─╼ 5 ╾ 1 + │ │ ├─╼ 1 ╾ 1, 4, 2 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + + """ + ).strip() + assert target == text + + +def test_write_network_text_nearly_forest(): + g = nx.DiGraph() + g.add_edge(1, 2) + g.add_edge(1, 5) + g.add_edge(2, 3) + g.add_edge(3, 4) + g.add_edge(5, 6) + g.add_edge(6, 7) + g.add_edge(6, 8) + orig = g.copy() + g.add_edge(1, 8) # forward edge + g.add_edge(4, 2) # back edge + g.add_edge(6, 3) # cross edge + lines = [] + write = lines.append + write("--- directed case ---") + nx.write_network_text(orig, path=write, end="") + write("--- add (1, 8), (4, 2), (6, 3) ---") + nx.write_network_text(g, path=write, end="") + write("--- undirected case ---") + nx.write_network_text(orig.to_undirected(), path=write, sources=[1], end="") + write("--- add (1, 8), (4, 2), (6, 3) ---") + nx.write_network_text(g.to_undirected(), path=write, sources=[1], end="") + text = "\n".join(lines) + target = dedent( + """ + --- directed case --- + ╙── 1 + ├─╼ 2 + │ └─╼ 3 + │ └─╼ 4 + └─╼ 5 + └─╼ 6 + ├─╼ 7 + └─╼ 8 + --- add (1, 8), (4, 2), (6, 3) --- + ╙── 1 + ├─╼ 2 ╾ 4 + │ └─╼ 3 ╾ 6 + │ └─╼ 4 + │ └─╼ ... + ├─╼ 5 + │ └─╼ 6 + │ ├─╼ 7 + │ ├─╼ 8 ╾ 1 + │ └─╼ ... + └─╼ ... + --- undirected case --- + ╙── 1 + ├── 2 + │ └── 3 + │ └── 4 + └── 5 + └── 6 + ├── 7 + └── 8 + --- add (1, 8), (4, 2), (6, 3) --- + ╙── 1 + ├── 2 + │ ├── 3 + │ │ ├── 4 ─ 2 + │ │ └── 6 + │ │ ├── 5 ─ 1 + │ │ ├── 7 + │ │ └── 8 ─ 1 + │ └── ... + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_complete_graph_ascii_only(): + graph = nx.generators.complete_graph(5, create_using=nx.DiGraph) + lines = [] + write = lines.append + write("--- directed case ---") + nx.write_network_text(graph, path=write, ascii_only=True, end="") + write("--- undirected case ---") + nx.write_network_text(graph.to_undirected(), path=write, ascii_only=True, end="") + text = "\n".join(lines) + target = dedent( + """ + --- directed case --- + +-- 0 <- 1, 2, 3, 4 + |-> 1 <- 2, 3, 4 + | |-> 2 <- 0, 3, 4 + | | |-> 3 <- 0, 1, 4 + | | | |-> 4 <- 0, 1, 2 + | | | | L-> ... + | | | L-> ... + | | L-> ... + | L-> ... + L-> ... + --- undirected case --- + +-- 0 + |-- 1 + | |-- 2 - 0 + | | |-- 3 - 0, 1 + | | | L-- 4 - 0, 1, 2 + | | L-- ... + | L-- ... + L-- ... + """ + ).strip() + assert target == text + + +def test_write_network_text_with_labels(): + graph = nx.generators.complete_graph(5, create_using=nx.DiGraph) + for n in graph.nodes: + graph.nodes[n]["label"] = f"Node(n={n})" + lines = [] + write = lines.append + nx.write_network_text(graph, path=write, with_labels=True, ascii_only=False, end="") + text = "\n".join(lines) + # Non trees with labels can get somewhat out of hand with network text + # because we need to immediately show every non-tree edge to the right + target = dedent( + """ + ╙── Node(n=0) ╾ Node(n=1), Node(n=2), Node(n=3), Node(n=4) + ├─╼ Node(n=1) ╾ Node(n=2), Node(n=3), Node(n=4) + │ ├─╼ Node(n=2) ╾ Node(n=0), Node(n=3), Node(n=4) + │ │ ├─╼ Node(n=3) ╾ Node(n=0), Node(n=1), Node(n=4) + │ │ │ ├─╼ Node(n=4) ╾ Node(n=0), Node(n=1), Node(n=2) + │ │ │ │ └─╼ ... + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + """ + ).strip() + assert target == text + + +def test_write_network_text_complete_graphs(): + lines = [] + write = lines.append + for k in [0, 1, 2, 3, 4, 5]: + g = nx.generators.complete_graph(k) + write(f"--- undirected k={k} ---") + nx.write_network_text(g, path=write, end="") + + for k in [0, 1, 2, 3, 4, 5]: + g = nx.generators.complete_graph(k, nx.DiGraph) + write(f"--- directed k={k} ---") + nx.write_network_text(g, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + --- undirected k=0 --- + ╙ + --- undirected k=1 --- + ╙── 0 + --- undirected k=2 --- + ╙── 0 + └── 1 + --- undirected k=3 --- + ╙── 0 + ├── 1 + │ └── 2 ─ 0 + └── ... + --- undirected k=4 --- + ╙── 0 + ├── 1 + │ ├── 2 ─ 0 + │ │ └── 3 ─ 0, 1 + │ └── ... + └── ... + --- undirected k=5 --- + ╙── 0 + ├── 1 + │ ├── 2 ─ 0 + │ │ ├── 3 ─ 0, 1 + │ │ │ └── 4 ─ 0, 1, 2 + │ │ └── ... + │ └── ... + └── ... + --- directed k=0 --- + ╙ + --- directed k=1 --- + ╙── 0 + --- directed k=2 --- + ╙── 0 ╾ 1 + └─╼ 1 + └─╼ ... + --- directed k=3 --- + ╙── 0 ╾ 1, 2 + ├─╼ 1 ╾ 2 + │ ├─╼ 2 ╾ 0 + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- directed k=4 --- + ╙── 0 ╾ 1, 2, 3 + ├─╼ 1 ╾ 2, 3 + │ ├─╼ 2 ╾ 0, 3 + │ │ ├─╼ 3 ╾ 0, 1 + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- directed k=5 --- + ╙── 0 ╾ 1, 2, 3, 4 + ├─╼ 1 ╾ 2, 3, 4 + │ ├─╼ 2 ╾ 0, 3, 4 + │ │ ├─╼ 3 ╾ 0, 1, 4 + │ │ │ ├─╼ 4 ╾ 0, 1, 2 + │ │ │ │ └─╼ ... + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + """ + ).strip() + assert target == text + + +def test_write_network_text_multiple_sources(): + g = nx.DiGraph() + g.add_edge(1, 2) + g.add_edge(1, 3) + g.add_edge(2, 4) + g.add_edge(3, 5) + g.add_edge(3, 6) + g.add_edge(5, 4) + g.add_edge(4, 1) + g.add_edge(1, 5) + lines = [] + write = lines.append + # Use each node as the starting point to demonstrate how the representation + # changes. + nodes = sorted(g.nodes()) + for n in nodes: + write(f"--- source node: {n} ---") + nx.write_network_text(g, path=write, sources=[n], end="") + text = "\n".join(lines) + target = dedent( + """ + --- source node: 1 --- + ╙── 1 ╾ 4 + ├─╼ 2 + │ └─╼ 4 ╾ 5 + │ └─╼ ... + ├─╼ 3 + │ ├─╼ 5 ╾ 1 + │ │ └─╼ ... + │ └─╼ 6 + └─╼ ... + --- source node: 2 --- + ╙── 2 ╾ 1 + └─╼ 4 ╾ 5 + └─╼ 1 + ├─╼ 3 + │ ├─╼ 5 ╾ 1 + │ │ └─╼ ... + │ └─╼ 6 + └─╼ ... + --- source node: 3 --- + ╙── 3 ╾ 1 + ├─╼ 5 ╾ 1 + │ └─╼ 4 ╾ 2 + │ └─╼ 1 + │ ├─╼ 2 + │ │ └─╼ ... + │ └─╼ ... + └─╼ 6 + --- source node: 4 --- + ╙── 4 ╾ 2, 5 + └─╼ 1 + ├─╼ 2 + │ └─╼ ... + ├─╼ 3 + │ ├─╼ 5 ╾ 1 + │ │ └─╼ ... + │ └─╼ 6 + └─╼ ... + --- source node: 5 --- + ╙── 5 ╾ 3, 1 + └─╼ 4 ╾ 2 + └─╼ 1 + ├─╼ 2 + │ └─╼ ... + ├─╼ 3 + │ ├─╼ 6 + │ └─╼ ... + └─╼ ... + --- source node: 6 --- + ╙── 6 ╾ 3 + """ + ).strip() + assert target == text + + +def test_write_network_text_star_graph(): + graph = nx.star_graph(5, create_using=nx.Graph) + lines = [] + write = lines.append + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + ╙── 1 + └── 0 + ├── 2 + ├── 3 + ├── 4 + └── 5 + """ + ).strip() + assert target == text + + +def test_write_network_text_path_graph(): + graph = nx.path_graph(3, create_using=nx.Graph) + lines = [] + write = lines.append + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + ╙── 0 + └── 1 + └── 2 + """ + ).strip() + assert target == text + + +def test_write_network_text_lollipop_graph(): + graph = nx.lollipop_graph(4, 2, create_using=nx.Graph) + lines = [] + write = lines.append + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + ╙── 5 + └── 4 + └── 3 + ├── 0 + │ ├── 1 ─ 3 + │ │ └── 2 ─ 0, 3 + │ └── ... + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_wheel_graph(): + graph = nx.wheel_graph(7, create_using=nx.Graph) + lines = [] + write = lines.append + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + ╙── 1 + ├── 0 + │ ├── 2 ─ 1 + │ │ └── 3 ─ 0 + │ │ └── 4 ─ 0 + │ │ └── 5 ─ 0 + │ │ └── 6 ─ 0, 1 + │ └── ... + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_circular_ladder_graph(): + graph = nx.circular_ladder_graph(4, create_using=nx.Graph) + lines = [] + write = lines.append + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + ╙── 0 + ├── 1 + │ ├── 2 + │ │ ├── 3 ─ 0 + │ │ │ └── 7 + │ │ │ ├── 6 ─ 2 + │ │ │ │ └── 5 ─ 1 + │ │ │ │ └── 4 ─ 0, 7 + │ │ │ └── ... + │ │ └── ... + │ └── ... + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_dorogovtsev_goltsev_mendes_graph(): + graph = nx.dorogovtsev_goltsev_mendes_graph(4, create_using=nx.Graph) + lines = [] + write = lines.append + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + ╙── 15 + ├── 0 + │ ├── 1 ─ 15 + │ │ ├── 2 ─ 0 + │ │ │ ├── 4 ─ 0 + │ │ │ │ ├── 9 ─ 0 + │ │ │ │ │ ├── 22 ─ 0 + │ │ │ │ │ └── 38 ─ 4 + │ │ │ │ ├── 13 ─ 2 + │ │ │ │ │ ├── 34 ─ 2 + │ │ │ │ │ └── 39 ─ 4 + │ │ │ │ ├── 18 ─ 0 + │ │ │ │ ├── 30 ─ 2 + │ │ │ │ └── ... + │ │ │ ├── 5 ─ 1 + │ │ │ │ ├── 12 ─ 1 + │ │ │ │ │ ├── 29 ─ 1 + │ │ │ │ │ └── 40 ─ 5 + │ │ │ │ ├── 14 ─ 2 + │ │ │ │ │ ├── 35 ─ 2 + │ │ │ │ │ └── 41 ─ 5 + │ │ │ │ ├── 25 ─ 1 + │ │ │ │ ├── 31 ─ 2 + │ │ │ │ └── ... + │ │ │ ├── 7 ─ 0 + │ │ │ │ ├── 20 ─ 0 + │ │ │ │ └── 32 ─ 2 + │ │ │ ├── 10 ─ 1 + │ │ │ │ ├── 27 ─ 1 + │ │ │ │ └── 33 ─ 2 + │ │ │ ├── 16 ─ 0 + │ │ │ ├── 23 ─ 1 + │ │ │ └── ... + │ │ ├── 3 ─ 0 + │ │ │ ├── 8 ─ 0 + │ │ │ │ ├── 21 ─ 0 + │ │ │ │ └── 36 ─ 3 + │ │ │ ├── 11 ─ 1 + │ │ │ │ ├── 28 ─ 1 + │ │ │ │ └── 37 ─ 3 + │ │ │ ├── 17 ─ 0 + │ │ │ ├── 24 ─ 1 + │ │ │ └── ... + │ │ ├── 6 ─ 0 + │ │ │ ├── 19 ─ 0 + │ │ │ └── 26 ─ 1 + │ │ └── ... + │ └── ... + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_tree_max_depth(): + orig = nx.balanced_tree(r=1, h=3, create_using=nx.DiGraph) + lines = [] + write = lines.append + write("--- directed case, max_depth=0 ---") + nx.write_network_text(orig, path=write, end="", max_depth=0) + write("--- directed case, max_depth=1 ---") + nx.write_network_text(orig, path=write, end="", max_depth=1) + write("--- directed case, max_depth=2 ---") + nx.write_network_text(orig, path=write, end="", max_depth=2) + write("--- directed case, max_depth=3 ---") + nx.write_network_text(orig, path=write, end="", max_depth=3) + write("--- directed case, max_depth=4 ---") + nx.write_network_text(orig, path=write, end="", max_depth=4) + write("--- undirected case, max_depth=0 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=0) + write("--- undirected case, max_depth=1 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=1) + write("--- undirected case, max_depth=2 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=2) + write("--- undirected case, max_depth=3 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=3) + write("--- undirected case, max_depth=4 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=4) + text = "\n".join(lines) + target = dedent( + """ + --- directed case, max_depth=0 --- + ╙ ... + --- directed case, max_depth=1 --- + ╙── 0 + └─╼ ... + --- directed case, max_depth=2 --- + ╙── 0 + └─╼ 1 + └─╼ ... + --- directed case, max_depth=3 --- + ╙── 0 + └─╼ 1 + └─╼ 2 + └─╼ ... + --- directed case, max_depth=4 --- + ╙── 0 + └─╼ 1 + └─╼ 2 + └─╼ 3 + --- undirected case, max_depth=0 --- + ╙ ... + --- undirected case, max_depth=1 --- + ╙── 0 ─ 1 + └── ... + --- undirected case, max_depth=2 --- + ╙── 0 + └── 1 ─ 2 + └── ... + --- undirected case, max_depth=3 --- + ╙── 0 + └── 1 + └── 2 ─ 3 + └── ... + --- undirected case, max_depth=4 --- + ╙── 0 + └── 1 + └── 2 + └── 3 + """ + ).strip() + assert target == text + + +def test_write_network_text_graph_max_depth(): + orig = nx.erdos_renyi_graph(10, 0.15, directed=True, seed=40392) + lines = [] + write = lines.append + write("--- directed case, max_depth=None ---") + nx.write_network_text(orig, path=write, end="", max_depth=None) + write("--- directed case, max_depth=0 ---") + nx.write_network_text(orig, path=write, end="", max_depth=0) + write("--- directed case, max_depth=1 ---") + nx.write_network_text(orig, path=write, end="", max_depth=1) + write("--- directed case, max_depth=2 ---") + nx.write_network_text(orig, path=write, end="", max_depth=2) + write("--- directed case, max_depth=3 ---") + nx.write_network_text(orig, path=write, end="", max_depth=3) + write("--- undirected case, max_depth=None ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=None) + write("--- undirected case, max_depth=0 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=0) + write("--- undirected case, max_depth=1 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=1) + write("--- undirected case, max_depth=2 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=2) + write("--- undirected case, max_depth=3 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=3) + text = "\n".join(lines) + target = dedent( + """ + --- directed case, max_depth=None --- + ╟── 4 + ╎ ├─╼ 0 ╾ 3 + ╎ ├─╼ 5 ╾ 7 + ╎ │ └─╼ 3 + ╎ │ ├─╼ 1 ╾ 9 + ╎ │ │ └─╼ 9 ╾ 6 + ╎ │ │ ├─╼ 6 + ╎ │ │ │ └─╼ ... + ╎ │ │ ├─╼ 7 ╾ 4 + ╎ │ │ │ ├─╼ 2 + ╎ │ │ │ └─╼ ... + ╎ │ │ └─╼ ... + ╎ │ └─╼ ... + ╎ └─╼ ... + ╙── 8 + --- directed case, max_depth=0 --- + ╙ ... + --- directed case, max_depth=1 --- + ╟── 4 + ╎ └─╼ ... + ╙── 8 + --- directed case, max_depth=2 --- + ╟── 4 + ╎ ├─╼ 0 ╾ 3 + ╎ ├─╼ 5 ╾ 7 + ╎ │ └─╼ ... + ╎ └─╼ 7 ╾ 9 + ╎ └─╼ ... + ╙── 8 + --- directed case, max_depth=3 --- + ╟── 4 + ╎ ├─╼ 0 ╾ 3 + ╎ ├─╼ 5 ╾ 7 + ╎ │ └─╼ 3 + ╎ │ └─╼ ... + ╎ └─╼ 7 ╾ 9 + ╎ ├─╼ 2 + ╎ └─╼ ... + ╙── 8 + --- undirected case, max_depth=None --- + ╟── 8 + ╙── 2 + └── 7 + ├── 4 + │ ├── 0 + │ │ └── 3 + │ │ ├── 1 + │ │ │ └── 9 ─ 7 + │ │ │ └── 6 + │ │ └── 5 ─ 4, 7 + │ └── ... + └── ... + --- undirected case, max_depth=0 --- + ╙ ... + --- undirected case, max_depth=1 --- + ╟── 8 + ╙── 2 ─ 7 + └── ... + --- undirected case, max_depth=2 --- + ╟── 8 + ╙── 2 + └── 7 ─ 4, 5, 9 + └── ... + --- undirected case, max_depth=3 --- + ╟── 8 + ╙── 2 + └── 7 + ├── 4 ─ 0, 5 + │ └── ... + ├── 5 ─ 4, 3 + │ └── ... + └── 9 ─ 1, 6 + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_clique_max_depth(): + orig = nx.complete_graph(5, nx.DiGraph) + lines = [] + write = lines.append + write("--- directed case, max_depth=None ---") + nx.write_network_text(orig, path=write, end="", max_depth=None) + write("--- directed case, max_depth=0 ---") + nx.write_network_text(orig, path=write, end="", max_depth=0) + write("--- directed case, max_depth=1 ---") + nx.write_network_text(orig, path=write, end="", max_depth=1) + write("--- directed case, max_depth=2 ---") + nx.write_network_text(orig, path=write, end="", max_depth=2) + write("--- directed case, max_depth=3 ---") + nx.write_network_text(orig, path=write, end="", max_depth=3) + write("--- undirected case, max_depth=None ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=None) + write("--- undirected case, max_depth=0 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=0) + write("--- undirected case, max_depth=1 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=1) + write("--- undirected case, max_depth=2 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=2) + write("--- undirected case, max_depth=3 ---") + nx.write_network_text(orig.to_undirected(), path=write, end="", max_depth=3) + text = "\n".join(lines) + target = dedent( + """ + --- directed case, max_depth=None --- + ╙── 0 ╾ 1, 2, 3, 4 + ├─╼ 1 ╾ 2, 3, 4 + │ ├─╼ 2 ╾ 0, 3, 4 + │ │ ├─╼ 3 ╾ 0, 1, 4 + │ │ │ ├─╼ 4 ╾ 0, 1, 2 + │ │ │ │ └─╼ ... + │ │ │ └─╼ ... + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- directed case, max_depth=0 --- + ╙ ... + --- directed case, max_depth=1 --- + ╙── 0 ╾ 1, 2, 3, 4 + └─╼ ... + --- directed case, max_depth=2 --- + ╙── 0 ╾ 1, 2, 3, 4 + ├─╼ 1 ╾ 2, 3, 4 + │ └─╼ ... + ├─╼ 2 ╾ 1, 3, 4 + │ └─╼ ... + ├─╼ 3 ╾ 1, 2, 4 + │ └─╼ ... + └─╼ 4 ╾ 1, 2, 3 + └─╼ ... + --- directed case, max_depth=3 --- + ╙── 0 ╾ 1, 2, 3, 4 + ├─╼ 1 ╾ 2, 3, 4 + │ ├─╼ 2 ╾ 0, 3, 4 + │ │ └─╼ ... + │ ├─╼ 3 ╾ 0, 2, 4 + │ │ └─╼ ... + │ ├─╼ 4 ╾ 0, 2, 3 + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + --- undirected case, max_depth=None --- + ╙── 0 + ├── 1 + │ ├── 2 ─ 0 + │ │ ├── 3 ─ 0, 1 + │ │ │ └── 4 ─ 0, 1, 2 + │ │ └── ... + │ └── ... + └── ... + --- undirected case, max_depth=0 --- + ╙ ... + --- undirected case, max_depth=1 --- + ╙── 0 ─ 1, 2, 3, 4 + └── ... + --- undirected case, max_depth=2 --- + ╙── 0 + ├── 1 ─ 2, 3, 4 + │ └── ... + ├── 2 ─ 1, 3, 4 + │ └── ... + ├── 3 ─ 1, 2, 4 + │ └── ... + └── 4 ─ 1, 2, 3 + --- undirected case, max_depth=3 --- + ╙── 0 + ├── 1 + │ ├── 2 ─ 0, 3, 4 + │ │ └── ... + │ ├── 3 ─ 0, 2, 4 + │ │ └── ... + │ └── 4 ─ 0, 2, 3 + └── ... + """ + ).strip() + assert target == text + + +def test_write_network_text_custom_label(): + # Create a directed forest with labels + graph = nx.erdos_renyi_graph(5, 0.4, directed=True, seed=359222358) + for node in graph.nodes: + graph.nodes[node]["label"] = f"Node({node})" + graph.nodes[node]["chr"] = chr(node + ord("a") - 1) + if node % 2 == 0: + graph.nodes[node]["part"] = chr(node + ord("a")) + + lines = [] + write = lines.append + write("--- when with_labels=True, uses the 'label' attr ---") + nx.write_network_text(graph, path=write, with_labels=True, end="", max_depth=None) + write("--- when with_labels=False, uses str(node) value ---") + nx.write_network_text(graph, path=write, with_labels=False, end="", max_depth=None) + write("--- when with_labels is a string, use that attr ---") + nx.write_network_text(graph, path=write, with_labels="chr", end="", max_depth=None) + write("--- fallback to str(node) when the attr does not exist ---") + nx.write_network_text(graph, path=write, with_labels="part", end="", max_depth=None) + + text = "\n".join(lines) + target = dedent( + """ + --- when with_labels=True, uses the 'label' attr --- + ╙── Node(1) + └─╼ Node(3) ╾ Node(2) + ├─╼ Node(0) + │ ├─╼ Node(2) ╾ Node(3), Node(4) + │ │ └─╼ ... + │ └─╼ Node(4) + │ └─╼ ... + └─╼ ... + --- when with_labels=False, uses str(node) value --- + ╙── 1 + └─╼ 3 ╾ 2 + ├─╼ 0 + │ ├─╼ 2 ╾ 3, 4 + │ │ └─╼ ... + │ └─╼ 4 + │ └─╼ ... + └─╼ ... + --- when with_labels is a string, use that attr --- + ╙── a + └─╼ c ╾ b + ├─╼ ` + │ ├─╼ b ╾ c, d + │ │ └─╼ ... + │ └─╼ d + │ └─╼ ... + └─╼ ... + --- fallback to str(node) when the attr does not exist --- + ╙── 1 + └─╼ 3 ╾ c + ├─╼ a + │ ├─╼ c ╾ 3, e + │ │ └─╼ ... + │ └─╼ e + │ └─╼ ... + └─╼ ... + """ + ).strip() + assert target == text + + +def test_write_network_text_vertical_chains(): + graph1 = nx.lollipop_graph(4, 2, create_using=nx.Graph) + graph1.add_edge(0, -1) + graph1.add_edge(-1, -2) + graph1.add_edge(-2, -3) + + graph2 = graph1.to_directed() + graph2.remove_edges_from([(u, v) for u, v in graph2.edges if v > u]) + + lines = [] + write = lines.append + write("--- Undirected UTF ---") + nx.write_network_text(graph1, path=write, end="", vertical_chains=True) + write("--- Undirected ASCI ---") + nx.write_network_text( + graph1, path=write, end="", vertical_chains=True, ascii_only=True + ) + write("--- Directed UTF ---") + nx.write_network_text(graph2, path=write, end="", vertical_chains=True) + write("--- Directed ASCI ---") + nx.write_network_text( + graph2, path=write, end="", vertical_chains=True, ascii_only=True + ) + + text = "\n".join(lines) + target = dedent( + """ + --- Undirected UTF --- + ╙── 5 + │ + 4 + │ + 3 + ├── 0 + │ ├── 1 ─ 3 + │ │ │ + │ │ 2 ─ 0, 3 + │ ├── -1 + │ │ │ + │ │ -2 + │ │ │ + │ │ -3 + │ └── ... + └── ... + --- Undirected ASCI --- + +-- 5 + | + 4 + | + 3 + |-- 0 + | |-- 1 - 3 + | | | + | | 2 - 0, 3 + | |-- -1 + | | | + | | -2 + | | | + | | -3 + | L-- ... + L-- ... + --- Directed UTF --- + ╙── 5 + ╽ + 4 + ╽ + 3 + ├─╼ 0 ╾ 1, 2 + │ ╽ + │ -1 + │ ╽ + │ -2 + │ ╽ + │ -3 + ├─╼ 1 ╾ 2 + │ └─╼ ... + └─╼ 2 + └─╼ ... + --- Directed ASCI --- + +-- 5 + ! + 4 + ! + 3 + |-> 0 <- 1, 2 + | ! + | -1 + | ! + | -2 + | ! + | -3 + |-> 1 <- 2 + | L-> ... + L-> 2 + L-> ... + """ + ).strip() + assert target == text + + +def test_collapse_directed(): + graph = nx.balanced_tree(r=2, h=3, create_using=nx.DiGraph) + lines = [] + write = lines.append + write("--- Original ---") + nx.write_network_text(graph, path=write, end="") + graph.nodes[1]["collapse"] = True + write("--- Collapse Node 1 ---") + nx.write_network_text(graph, path=write, end="") + write("--- Add alternate path (5, 3) to collapsed zone") + graph.add_edge(5, 3) + nx.write_network_text(graph, path=write, end="") + write("--- Collapse Node 0 ---") + graph.nodes[0]["collapse"] = True + nx.write_network_text(graph, path=write, end="") + text = "\n".join(lines) + target = dedent( + """ + --- Original --- + ╙── 0 + ├─╼ 1 + │ ├─╼ 3 + │ │ ├─╼ 7 + │ │ └─╼ 8 + │ └─╼ 4 + │ ├─╼ 9 + │ └─╼ 10 + └─╼ 2 + ├─╼ 5 + │ ├─╼ 11 + │ └─╼ 12 + └─╼ 6 + ├─╼ 13 + └─╼ 14 + --- Collapse Node 1 --- + ╙── 0 + ├─╼ 1 + │ └─╼ ... + └─╼ 2 + ├─╼ 5 + │ ├─╼ 11 + │ └─╼ 12 + └─╼ 6 + ├─╼ 13 + └─╼ 14 + --- Add alternate path (5, 3) to collapsed zone + ╙── 0 + ├─╼ 1 + │ └─╼ ... + └─╼ 2 + ├─╼ 5 + │ ├─╼ 11 + │ ├─╼ 12 + │ └─╼ 3 ╾ 1 + │ ├─╼ 7 + │ └─╼ 8 + └─╼ 6 + ├─╼ 13 + └─╼ 14 + --- Collapse Node 0 --- + ╙── 0 + └─╼ ... + """ + ).strip() + assert target == text + + +def test_collapse_undirected(): + graph = nx.balanced_tree(r=2, h=3, create_using=nx.Graph) + lines = [] + write = lines.append + write("--- Original ---") + nx.write_network_text(graph, path=write, end="", sources=[0]) + graph.nodes[1]["collapse"] = True + write("--- Collapse Node 1 ---") + nx.write_network_text(graph, path=write, end="", sources=[0]) + write("--- Add alternate path (5, 3) to collapsed zone") + graph.add_edge(5, 3) + nx.write_network_text(graph, path=write, end="", sources=[0]) + write("--- Collapse Node 0 ---") + graph.nodes[0]["collapse"] = True + nx.write_network_text(graph, path=write, end="", sources=[0]) + text = "\n".join(lines) + target = dedent( + """ + --- Original --- + ╙── 0 + ├── 1 + │ ├── 3 + │ │ ├── 7 + │ │ └── 8 + │ └── 4 + │ ├── 9 + │ └── 10 + └── 2 + ├── 5 + │ ├── 11 + │ └── 12 + └── 6 + ├── 13 + └── 14 + --- Collapse Node 1 --- + ╙── 0 + ├── 1 ─ 3, 4 + │ └── ... + └── 2 + ├── 5 + │ ├── 11 + │ └── 12 + └── 6 + ├── 13 + └── 14 + --- Add alternate path (5, 3) to collapsed zone + ╙── 0 + ├── 1 ─ 3, 4 + │ └── ... + └── 2 + ├── 5 + │ ├── 11 + │ ├── 12 + │ └── 3 ─ 1 + │ ├── 7 + │ └── 8 + └── 6 + ├── 13 + └── 14 + --- Collapse Node 0 --- + ╙── 0 ─ 1, 2 + └── ... + """ + ).strip() + assert target == text + + +def generate_test_graphs(): + """ + Generate a gauntlet of different test graphs with different properties + """ + import random + + rng = random.Random(976689776) + num_randomized = 3 + + for directed in [0, 1]: + cls = nx.DiGraph if directed else nx.Graph + + for num_nodes in range(17): + # Disconnected graph + graph = cls() + graph.add_nodes_from(range(num_nodes)) + yield graph + + # Randomize graphs + if num_nodes > 0: + for p in [0.1, 0.3, 0.5, 0.7, 0.9]: + for seed in range(num_randomized): + graph = nx.erdos_renyi_graph( + num_nodes, p, directed=directed, seed=rng + ) + yield graph + + yield nx.complete_graph(num_nodes, cls) + + yield nx.path_graph(3, create_using=cls) + yield nx.balanced_tree(r=1, h=3, create_using=cls) + if not directed: + yield nx.circular_ladder_graph(4, create_using=cls) + yield nx.star_graph(5, create_using=cls) + yield nx.lollipop_graph(4, 2, create_using=cls) + yield nx.wheel_graph(7, create_using=cls) + yield nx.dorogovtsev_goltsev_mendes_graph(4, create_using=cls) + + +@pytest.mark.parametrize( + ("vertical_chains", "ascii_only"), + tuple( + [ + (vertical_chains, ascii_only) + for vertical_chains in [0, 1] + for ascii_only in [0, 1] + ] + ), +) +def test_network_text_round_trip(vertical_chains, ascii_only): + """ + Write the graph to network text format, then parse it back in, assert it is + the same as the original graph. Passing this test is strong validation of + both the format generator and parser. + """ + from networkx.readwrite.text import _parse_network_text + + for graph in generate_test_graphs(): + graph = nx.relabel_nodes(graph, {n: str(n) for n in graph.nodes}) + lines = list( + nx.generate_network_text( + graph, vertical_chains=vertical_chains, ascii_only=ascii_only + ) + ) + new = _parse_network_text(lines) + try: + assert new.nodes == graph.nodes + assert new.edges == graph.edges + except Exception: + nx.write_network_text(graph) + raise diff --git a/.venv/lib/python3.12/site-packages/networkx/readwrite/text.py b/.venv/lib/python3.12/site-packages/networkx/readwrite/text.py new file mode 100644 index 00000000..c54901d1 --- /dev/null +++ b/.venv/lib/python3.12/site-packages/networkx/readwrite/text.py @@ -0,0 +1,852 @@ +""" +Text-based visual representations of graphs +""" + +import sys +import warnings +from collections import defaultdict + +import networkx as nx +from networkx.utils import open_file + +__all__ = ["generate_network_text", "write_network_text"] + + +class BaseGlyphs: + @classmethod + def as_dict(cls): + return { + a: getattr(cls, a) + for a in dir(cls) + if not a.startswith("_") and a != "as_dict" + } + + +class AsciiBaseGlyphs(BaseGlyphs): + empty: str = "+" + newtree_last: str = "+-- " + newtree_mid: str = "+-- " + endof_forest: str = " " + within_forest: str = ": " + within_tree: str = "| " + + +class AsciiDirectedGlyphs(AsciiBaseGlyphs): + last: str = "L-> " + mid: str = "|-> " + backedge: str = "<-" + vertical_edge: str = "!" + + +class AsciiUndirectedGlyphs(AsciiBaseGlyphs): + last: str = "L-- " + mid: str = "|-- " + backedge: str = "-" + vertical_edge: str = "|" + + +class UtfBaseGlyphs(BaseGlyphs): + # Notes on available box and arrow characters + # https://en.wikipedia.org/wiki/Box-drawing_character + # https://stackoverflow.com/questions/2701192/triangle-arrow + empty: str = "╙" + newtree_last: str = "╙── " + newtree_mid: str = "╟── " + endof_forest: str = " " + within_forest: str = "╎ " + within_tree: str = "│ " + + +class UtfDirectedGlyphs(UtfBaseGlyphs): + last: str = "└─╼ " + mid: str = "├─╼ " + backedge: str = "╾" + vertical_edge: str = "╽" + + +class UtfUndirectedGlyphs(UtfBaseGlyphs): + last: str = "└── " + mid: str = "├── " + backedge: str = "─" + vertical_edge: str = "│" + + +def generate_network_text( + graph, + with_labels=True, + sources=None, + max_depth=None, + ascii_only=False, + vertical_chains=False, +): + """Generate lines in the "network text" format + + This works via a depth-first traversal of the graph and writing a line for + each unique node encountered. Non-tree edges are written to the right of + each node, and connection to a non-tree edge is indicated with an ellipsis. + This representation works best when the input graph is a forest, but any + graph can be represented. + + This notation is original to networkx, although it is simple enough that it + may be known in existing literature. See #5602 for details. The procedure + is summarized as follows: + + 1. Given a set of source nodes (which can be specified, or automatically + discovered via finding the (strongly) connected components and choosing one + node with minimum degree from each), we traverse the graph in depth first + order. + + 2. Each reachable node will be printed exactly once on it's own line. + + 3. Edges are indicated in one of four ways: + + a. a parent "L-style" connection on the upper left. This corresponds to + a traversal in the directed DFS tree. + + b. a backref "<-style" connection shown directly on the right. For + directed graphs, these are drawn for any incoming edges to a node that + is not a parent edge. For undirected graphs, these are drawn for only + the non-parent edges that have already been represented (The edges that + have not been represented will be handled in the recursive case). + + c. a child "L-style" connection on the lower right. Drawing of the + children are handled recursively. + + d. if ``vertical_chains`` is true, and a parent node only has one child + a "vertical-style" edge is drawn between them. + + 4. The children of each node (wrt the directed DFS tree) are drawn + underneath and to the right of it. In the case that a child node has already + been drawn the connection is replaced with an ellipsis ("...") to indicate + that there is one or more connections represented elsewhere. + + 5. If a maximum depth is specified, an edge to nodes past this maximum + depth will be represented by an ellipsis. + + 6. If a node has a truthy "collapse" value, then we do not traverse past + that node. + + Parameters + ---------- + graph : nx.DiGraph | nx.Graph + Graph to represent + + with_labels : bool | str + If True will use the "label" attribute of a node to display if it + exists otherwise it will use the node value itself. If given as a + string, then that attribute name will be used instead of "label". + Defaults to True. + + sources : List + Specifies which nodes to start traversal from. Note: nodes that are not + reachable from one of these sources may not be shown. If unspecified, + the minimal set of nodes needed to reach all others will be used. + + max_depth : int | None + The maximum depth to traverse before stopping. Defaults to None. + + ascii_only : Boolean + If True only ASCII characters are used to construct the visualization + + vertical_chains : Boolean + If True, chains of nodes will be drawn vertically when possible. + + Yields + ------ + str : a line of generated text + + Examples + -------- + >>> graph = nx.path_graph(10) + >>> graph.add_node("A") + >>> graph.add_node("B") + >>> graph.add_node("C") + >>> graph.add_node("D") + >>> graph.add_edge(9, "A") + >>> graph.add_edge(9, "B") + >>> graph.add_edge(9, "C") + >>> graph.add_edge("C", "D") + >>> graph.add_edge("C", "E") + >>> graph.add_edge("C", "F") + >>> nx.write_network_text(graph) + ╙── 0 + └── 1 + └── 2 + └── 3 + └── 4 + └── 5 + └── 6 + └── 7 + └── 8 + └── 9 + ├── A + ├── B + └── C + ├── D + ├── E + └── F + >>> nx.write_network_text(graph, vertical_chains=True) + ╙── 0 + │ + 1 + │ + 2 + │ + 3 + │ + 4 + │ + 5 + │ + 6 + │ + 7 + │ + 8 + │ + 9 + ├── A + ├── B + └── C + ├── D + ├── E + └── F + """ + from typing import Any, NamedTuple + + class StackFrame(NamedTuple): + parent: Any + node: Any + indents: list + this_islast: bool + this_vertical: bool + + collapse_attr = "collapse" + + is_directed = graph.is_directed() + + if is_directed: + glyphs = AsciiDirectedGlyphs if ascii_only else UtfDirectedGlyphs + succ = graph.succ + pred = graph.pred + else: + glyphs = AsciiUndirectedGlyphs if ascii_only else UtfUndirectedGlyphs + succ = graph.adj + pred = graph.adj + + if isinstance(with_labels, str): + label_attr = with_labels + elif with_labels: + label_attr = "label" + else: + label_attr = None + + if max_depth == 0: + yield glyphs.empty + " ..." + elif len(graph.nodes) == 0: + yield glyphs.empty + else: + # If the nodes to traverse are unspecified, find the minimal set of + # nodes that will reach the entire graph + if sources is None: + sources = _find_sources(graph) + + # Populate the stack with each: + # 1. parent node in the DFS tree (or None for root nodes), + # 2. the current node in the DFS tree + # 2. a list of indentations indicating depth + # 3. a flag indicating if the node is the final one to be written. + # Reverse the stack so sources are popped in the correct order. + last_idx = len(sources) - 1 + stack = [ + StackFrame(None, node, [], (idx == last_idx), False) + for idx, node in enumerate(sources) + ][::-1] + + num_skipped_children = defaultdict(lambda: 0) + seen_nodes = set() + while stack: + parent, node, indents, this_islast, this_vertical = stack.pop() + + if node is not Ellipsis: + skip = node in seen_nodes + if skip: + # Mark that we skipped a parent's child + num_skipped_children[parent] += 1 + + if this_islast: + # If we reached the last child of a parent, and we skipped + # any of that parents children, then we should emit an + # ellipsis at the end after this. + if num_skipped_children[parent] and parent is not None: + # Append the ellipsis to be emitted last + next_islast = True + try_frame = StackFrame( + node, Ellipsis, indents, next_islast, False + ) + stack.append(try_frame) + + # Redo this frame, but not as a last object + next_islast = False + try_frame = StackFrame( + parent, node, indents, next_islast, this_vertical + ) + stack.append(try_frame) + continue + + if skip: + continue + seen_nodes.add(node) + + if not indents: + # Top level items (i.e. trees in the forest) get different + # glyphs to indicate they are not actually connected + if this_islast: + this_vertical = False + this_prefix = indents + [glyphs.newtree_last] + next_prefix = indents + [glyphs.endof_forest] + else: + this_prefix = indents + [glyphs.newtree_mid] + next_prefix = indents + [glyphs.within_forest] + + else: + # Non-top-level items + if this_vertical: + this_prefix = indents + next_prefix = indents + else: + if this_islast: + this_prefix = indents + [glyphs.last] + next_prefix = indents + [glyphs.endof_forest] + else: + this_prefix = indents + [glyphs.mid] + next_prefix = indents + [glyphs.within_tree] + + if node is Ellipsis: + label = " ..." + suffix = "" + children = [] + else: + if label_attr is not None: + label = str(graph.nodes[node].get(label_attr, node)) + else: + label = str(node) + + # Determine if we want to show the children of this node. + if collapse_attr is not None: + collapse = graph.nodes[node].get(collapse_attr, False) + else: + collapse = False + + # Determine: + # (1) children to traverse into after showing this node. + # (2) parents to immediately show to the right of this node. + if is_directed: + # In the directed case we must show every successor node + # note: it may be skipped later, but we don't have that + # information here. + children = list(succ[node]) + # In the directed case we must show every predecessor + # except for parent we directly traversed from. + handled_parents = {parent} + else: + # Showing only the unseen children results in a more + # concise representation for the undirected case. + children = [ + child for child in succ[node] if child not in seen_nodes + ] + + # In the undirected case, parents are also children, so we + # only need to immediately show the ones we can no longer + # traverse + handled_parents = {*children, parent} + + if max_depth is not None and len(indents) == max_depth - 1: + # Use ellipsis to indicate we have reached maximum depth + if children: + children = [Ellipsis] + handled_parents = {parent} + + if collapse: + # Collapsing a node is the same as reaching maximum depth + if children: + children = [Ellipsis] + handled_parents = {parent} + + # The other parents are other predecessors of this node that + # are not handled elsewhere. + other_parents = [p for p in pred[node] if p not in handled_parents] + if other_parents: + if label_attr is not None: + other_parents_labels = ", ".join( + [ + str(graph.nodes[p].get(label_attr, p)) + for p in other_parents + ] + ) + else: + other_parents_labels = ", ".join( + [str(p) for p in other_parents] + ) + suffix = " ".join(["", glyphs.backedge, other_parents_labels]) + else: + suffix = "" + + # Emit the line for this node, this will be called for each node + # exactly once. + if this_vertical: + yield "".join(this_prefix + [glyphs.vertical_edge]) + + yield "".join(this_prefix + [label, suffix]) + + if vertical_chains: + if is_directed: + num_children = len(set(children)) + else: + num_children = len(set(children) - {parent}) + # The next node can be drawn vertically if it is the only + # remaining child of this node. + next_is_vertical = num_children == 1 + else: + next_is_vertical = False + + # Push children on the stack in reverse order so they are popped in + # the original order. + for idx, child in enumerate(children[::-1]): + next_islast = idx == 0 + try_frame = StackFrame( + node, child, next_prefix, next_islast, next_is_vertical + ) + stack.append(try_frame) + + +@open_file(1, "w") +def write_network_text( + graph, + path=None, + with_labels=True, + sources=None, + max_depth=None, + ascii_only=False, + end="\n", + vertical_chains=False, +): + """Creates a nice text representation of a graph + + This works via a depth-first traversal of the graph and writing a line for + each unique node encountered. Non-tree edges are written to the right of + each node, and connection to a non-tree edge is indicated with an ellipsis. + This representation works best when the input graph is a forest, but any + graph can be represented. + + Parameters + ---------- + graph : nx.DiGraph | nx.Graph + Graph to represent + + path : string or file or callable or None + Filename or file handle for data output. + if a function, then it will be called for each generated line. + if None, this will default to "sys.stdout.write" + + with_labels : bool | str + If True will use the "label" attribute of a node to display if it + exists otherwise it will use the node value itself. If given as a + string, then that attribute name will be used instead of "label". + Defaults to True. + + sources : List + Specifies which nodes to start traversal from. Note: nodes that are not + reachable from one of these sources may not be shown. If unspecified, + the minimal set of nodes needed to reach all others will be used. + + max_depth : int | None + The maximum depth to traverse before stopping. Defaults to None. + + ascii_only : Boolean + If True only ASCII characters are used to construct the visualization + + end : string + The line ending character + + vertical_chains : Boolean + If True, chains of nodes will be drawn vertically when possible. + + Examples + -------- + >>> graph = nx.balanced_tree(r=2, h=2, create_using=nx.DiGraph) + >>> nx.write_network_text(graph) + ╙── 0 + ├─╼ 1 + │ ├─╼ 3 + │ └─╼ 4 + └─╼ 2 + ├─╼ 5 + └─╼ 6 + + >>> # A near tree with one non-tree edge + >>> graph.add_edge(5, 1) + >>> nx.write_network_text(graph) + ╙── 0 + ├─╼ 1 ╾ 5 + │ ├─╼ 3 + │ └─╼ 4 + └─╼ 2 + ├─╼ 5 + │ └─╼ ... + └─╼ 6 + + >>> graph = nx.cycle_graph(5) + >>> nx.write_network_text(graph) + ╙── 0 + ├── 1 + │ └── 2 + │ └── 3 + │ └── 4 ─ 0 + └── ... + + >>> graph = nx.cycle_graph(5, nx.DiGraph) + >>> nx.write_network_text(graph, vertical_chains=True) + ╙── 0 ╾ 4 + ╽ + 1 + ╽ + 2 + ╽ + 3 + ╽ + 4 + └─╼ ... + + >>> nx.write_network_text(graph, vertical_chains=True, ascii_only=True) + +-- 0 <- 4 + ! + 1 + ! + 2 + ! + 3 + ! + 4 + L-> ... + + >>> graph = nx.generators.barbell_graph(4, 2) + >>> nx.write_network_text(graph, vertical_chains=False) + ╙── 4 + ├── 5 + │ └── 6 + │ ├── 7 + │ │ ├── 8 ─ 6 + │ │ │ └── 9 ─ 6, 7 + │ │ └── ... + │ └── ... + └── 3 + ├── 0 + │ ├── 1 ─ 3 + │ │ └── 2 ─ 0, 3 + │ └── ... + └── ... + >>> nx.write_network_text(graph, vertical_chains=True) + ╙── 4 + ├── 5 + │ │ + │ 6 + │ ├── 7 + │ │ ├── 8 ─ 6 + │ │ │ │ + │ │ │ 9 ─ 6, 7 + │ │ └── ... + │ └── ... + └── 3 + ├── 0 + │ ├── 1 ─ 3 + │ │ │ + │ │ 2 ─ 0, 3 + │ └── ... + └── ... + + >>> graph = nx.complete_graph(5, create_using=nx.Graph) + >>> nx.write_network_text(graph) + ╙── 0 + ├── 1 + │ ├── 2 ─ 0 + │ │ ├── 3 ─ 0, 1 + │ │ │ └── 4 ─ 0, 1, 2 + │ │ └── ... + │ └── ... + └── ... + + >>> graph = nx.complete_graph(3, create_using=nx.DiGraph) + >>> nx.write_network_text(graph) + ╙── 0 ╾ 1, 2 + ├─╼ 1 ╾ 2 + │ ├─╼ 2 ╾ 0 + │ │ └─╼ ... + │ └─╼ ... + └─╼ ... + """ + if path is None: + # The path is unspecified, write to stdout + _write = sys.stdout.write + elif hasattr(path, "write"): + # The path is already an open file + _write = path.write + elif callable(path): + # The path is a custom callable + _write = path + else: + raise TypeError(type(path)) + + for line in generate_network_text( + graph, + with_labels=with_labels, + sources=sources, + max_depth=max_depth, + ascii_only=ascii_only, + vertical_chains=vertical_chains, + ): + _write(line + end) + + +def _find_sources(graph): + """ + Determine a minimal set of nodes such that the entire graph is reachable + """ + # For each connected part of the graph, choose at least + # one node as a starting point, preferably without a parent + if graph.is_directed(): + # Choose one node from each SCC with minimum in_degree + sccs = list(nx.strongly_connected_components(graph)) + # condensing the SCCs forms a dag, the nodes in this graph with + # 0 in-degree correspond to the SCCs from which the minimum set + # of nodes from which all other nodes can be reached. + scc_graph = nx.condensation(graph, sccs) + supernode_to_nodes = {sn: [] for sn in scc_graph.nodes()} + # Note: the order of mapping differs between pypy and cpython + # so we have to loop over graph nodes for consistency + mapping = scc_graph.graph["mapping"] + for n in graph.nodes: + sn = mapping[n] + supernode_to_nodes[sn].append(n) + sources = [] + for sn in scc_graph.nodes(): + if scc_graph.in_degree[sn] == 0: + scc = supernode_to_nodes[sn] + node = min(scc, key=lambda n: graph.in_degree[n]) + sources.append(node) + else: + # For undirected graph, the entire graph will be reachable as + # long as we consider one node from every connected component + sources = [ + min(cc, key=lambda n: graph.degree[n]) + for cc in nx.connected_components(graph) + ] + sources = sorted(sources, key=lambda n: graph.degree[n]) + return sources + + +def _parse_network_text(lines): + """Reconstructs a graph from a network text representation. + + This is mainly used for testing. Network text is for display, not + serialization, as such this cannot parse all network text representations + because node labels can be ambiguous with the glyphs and indentation used + to represent edge structure. Additionally, there is no way to determine if + disconnected graphs were originally directed or undirected. + + Parameters + ---------- + lines : list or iterator of strings + Input data in network text format + + Returns + ------- + G: NetworkX graph + The graph corresponding to the lines in network text format. + """ + from itertools import chain + from typing import Any, NamedTuple, Union + + class ParseStackFrame(NamedTuple): + node: Any + indent: int + has_vertical_child: int | None + + initial_line_iter = iter(lines) + + is_ascii = None + is_directed = None + + ############## + # Initial Pass + ############## + + # Do an initial pass over the lines to determine what type of graph it is. + # Remember what these lines were, so we can reiterate over them in the + # parsing pass. + initial_lines = [] + try: + first_line = next(initial_line_iter) + except StopIteration: + ... + else: + initial_lines.append(first_line) + # The first character indicates if it is an ASCII or UTF graph + first_char = first_line[0] + if first_char in { + UtfBaseGlyphs.empty, + UtfBaseGlyphs.newtree_mid[0], + UtfBaseGlyphs.newtree_last[0], + }: + is_ascii = False + elif first_char in { + AsciiBaseGlyphs.empty, + AsciiBaseGlyphs.newtree_mid[0], + AsciiBaseGlyphs.newtree_last[0], + }: + is_ascii = True + else: + raise AssertionError(f"Unexpected first character: {first_char}") + + if is_ascii: + directed_glyphs = AsciiDirectedGlyphs.as_dict() + undirected_glyphs = AsciiUndirectedGlyphs.as_dict() + else: + directed_glyphs = UtfDirectedGlyphs.as_dict() + undirected_glyphs = UtfUndirectedGlyphs.as_dict() + + # For both directed / undirected glyphs, determine which glyphs never + # appear as substrings in the other undirected / directed glyphs. Glyphs + # with this property unambiguously indicates if a graph is directed / + # undirected. + directed_items = set(directed_glyphs.values()) + undirected_items = set(undirected_glyphs.values()) + unambiguous_directed_items = [] + for item in directed_items: + other_items = undirected_items + other_supersets = [other for other in other_items if item in other] + if not other_supersets: + unambiguous_directed_items.append(item) + unambiguous_undirected_items = [] + for item in undirected_items: + other_items = directed_items + other_supersets = [other for other in other_items if item in other] + if not other_supersets: + unambiguous_undirected_items.append(item) + + for line in initial_line_iter: + initial_lines.append(line) + if any(item in line for item in unambiguous_undirected_items): + is_directed = False + break + elif any(item in line for item in unambiguous_directed_items): + is_directed = True + break + + if is_directed is None: + # Not enough information to determine, choose undirected by default + is_directed = False + + glyphs = directed_glyphs if is_directed else undirected_glyphs + + # the backedge symbol by itself can be ambiguous, but with spaces around it + # becomes unambiguous. + backedge_symbol = " " + glyphs["backedge"] + " " + + # Reconstruct an iterator over all of the lines. + parsing_line_iter = chain(initial_lines, initial_line_iter) + + ############## + # Parsing Pass + ############## + + edges = [] + nodes = [] + is_empty = None + + noparent = object() # sentinel value + + # keep a stack of previous nodes that could be parents of subsequent nodes + stack = [ParseStackFrame(noparent, -1, None)] + + for line in parsing_line_iter: + if line == glyphs["empty"]: + # If the line is the empty glyph, we are done. + # There shouldn't be anything else after this. + is_empty = True + continue + + if backedge_symbol in line: + # This line has one or more backedges, separate those out + node_part, backedge_part = line.split(backedge_symbol) + backedge_nodes = [u.strip() for u in backedge_part.split(", ")] + # Now the node can be parsed + node_part = node_part.rstrip() + prefix, node = node_part.rsplit(" ", 1) + node = node.strip() + # Add the backedges to the edge list + edges.extend([(u, node) for u in backedge_nodes]) + else: + # No backedge, the tail of this line is the node + prefix, node = line.rsplit(" ", 1) + node = node.strip() + + prev = stack.pop() + + if node in glyphs["vertical_edge"]: + # Previous node is still the previous node, but we know it will + # have exactly one child, which will need to have its nesting level + # adjusted. + modified_prev = ParseStackFrame( + prev.node, + prev.indent, + True, + ) + stack.append(modified_prev) + continue + + # The length of the string before the node characters give us a hint + # about our nesting level. The only case where this doesn't work is + # when there are vertical chains, which is handled explicitly. + indent = len(prefix) + curr = ParseStackFrame(node, indent, None) + + if prev.has_vertical_child: + # In this case we know prev must be the parent of our current line, + # so we don't have to search the stack. (which is good because the + # indentation check wouldn't work in this case). + ... + else: + # If the previous node nesting-level is greater than the current + # nodes nesting-level than the previous node was the end of a path, + # and is not our parent. We can safely pop nodes off the stack + # until we find one with a comparable nesting-level, which is our + # parent. + while curr.indent <= prev.indent: + prev = stack.pop() + + if node == "...": + # The current previous node is no longer a valid parent, + # keep it popped from the stack. + stack.append(prev) + else: + # The previous and current nodes may still be parents, so add them + # back onto the stack. + stack.append(prev) + stack.append(curr) + + # Add the node and the edge to its parent to the node / edge lists. + nodes.append(curr.node) + if prev.node is not noparent: + edges.append((prev.node, curr.node)) + + if is_empty: + # Sanity check + assert len(nodes) == 0 + + # Reconstruct the graph + cls = nx.DiGraph if is_directed else nx.Graph + new = cls() + new.add_nodes_from(nodes) + new.add_edges_from(edges) + return new |