From 10da4248b29b92e18c61323640acea59dd7244dc Mon Sep 17 00:00:00 2001 From: BonfaceKilz Date: Thu, 29 Apr 2021 20:31:26 +0300 Subject: base: data_set: Rewrite data_set using @dataclass @dataclasses should be used to hold only data. Since dataset only encapsulates data, re-writing it using a dataclass makes it more readable and easier to follow. --- wqflask/base/data_set.py | 86 +++++++++++++++++++++++++++--------------------- 1 file changed, 48 insertions(+), 38 deletions(-) (limited to 'wqflask/base') diff --git a/wqflask/base/data_set.py b/wqflask/base/data_set.py index cc5a428c..e3a85ff6 100644 --- a/wqflask/base/data_set.py +++ b/wqflask/base/data_set.py @@ -17,7 +17,10 @@ # at rwilliams@uthsc.edu and xzhou15@uthsc.edu # # This module is used by GeneNetwork project (www.genenetwork.org) - +from dataclasses import dataclass +from dataclasses import field +from dataclasses import InitVar +from typing import Optional, Dict from db.call import fetchall, fetchone, fetch1 from utility.logger import getLogger from utility.tools import USE_GN_SERVER, USE_REDIS, flat_files, flat_file_exists, GN2_BASE_URL @@ -74,11 +77,10 @@ def create_dataset(dataset_name, dataset_type=None, get_samplelist=True, group_n return dataset_class(dataset_name, get_samplelist) +@dataclass class DatasetType: - - def __init__(self, redis_instance): - """Create a dictionary of samples where the value is set to Geno, -Publish or ProbeSet. E.g. + """Create a dictionary of samples where the value is set to Geno, + Publish or ProbeSet. E.g. {'AD-cases-controls-MyersGeno': 'Geno', 'AD-cases-controls-MyersPublish': 'Publish', @@ -89,21 +91,28 @@ Publish or ProbeSet. E.g. 'All Phenotypes': 'Publish', 'B139_K_1206_M': 'ProbeSet', 'B139_K_1206_R': 'ProbeSet' ... - + } """ + redis_instance: InitVar[Redis] + datasets: Optional[Dict] = field(init=False, default_factory=dict) + data: Optional[Dict] = field(init=False) + + def __post_init__(self, redis_instance): self.redis_instance = redis_instance - self.datasets = {} - data = self.redis_instance.get("dataset_structure") + data = redis_instance.get("dataset_structure") if data: self.datasets = json.loads(data) - else: # ZS: I don't think this should ever run unless Redis is emptied + else: + # ZS: I don't think this should ever run unless Redis is + # emptied try: data = json.loads(requests.get( - GN2_BASE_URL + "/api/v_pre1/gen_dropdown", timeout=5).content) - for species in data['datasets']: - for group in data['datasets'][species]: - for dataset_type in data['datasets'][species][group]: - for dataset in data['datasets'][species][group][dataset_type]: + GN2_BASE_URL + "/api/v_pre1/gen_dropdown", + timeout=5).content) + for _species in data['datasets']: + for group in data['datasets'][_species]: + for dataset_type in data['datasets'][_species][group]: + for dataset in data['datasets'][_species][group][dataset_type]: short_dataset_name = dataset[1] if dataset_type == "Phenotypes": new_type = "Publish" @@ -112,15 +121,15 @@ Publish or ProbeSet. E.g. else: new_type = "ProbeSet" self.datasets[short_dataset_name] = new_type - except: + except Exception: # Do nothing pass - - self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) + self.redis_instance.set("dataset_structure", + json.dumps(self.datasets)) + self.data = data def set_dataset_key(self, t, name): - """If name is not in the object's dataset dictionary, set it, and update - dataset_structure in Redis - + """If name is not in the object's dataset dictionary, set it, and + update dataset_structure in Redis args: t: Type of dataset structure which can be: 'mrna_expr', 'pheno', 'other_pheno', 'geno' @@ -128,19 +137,20 @@ Publish or ProbeSet. E.g. """ sql_query_mapping = { - 'mrna_expr': ("""SELECT ProbeSetFreeze.Id FROM """ + - """ProbeSetFreeze WHERE ProbeSetFreeze.Name = "{}" """), - 'pheno': ("""SELECT InfoFiles.GN_AccesionId """ + - """FROM InfoFiles, PublishFreeze, InbredSet """ + - """WHERE InbredSet.Name = '{}' AND """ + - """PublishFreeze.InbredSetId = InbredSet.Id AND """ + - """InfoFiles.InfoPageName = PublishFreeze.Name"""), - 'other_pheno': ("""SELECT PublishFreeze.Name """ + - """FROM PublishFreeze, InbredSet """ + - """WHERE InbredSet.Name = '{}' AND """ + - """PublishFreeze.InbredSetId = InbredSet.Id"""), - 'geno': ("""SELECT GenoFreeze.Id FROM GenoFreeze WHERE """ + - """GenoFreeze.Name = "{}" """) + 'mrna_expr': ("SELECT ProbeSetFreeze.Id FROM " + "ProbeSetFreeze WHERE " + "ProbeSetFreeze.Name = \"%s\" "), + 'pheno': ("SELECT InfoFiles.GN_AccesionId " + "FROM InfoFiles, PublishFreeze, InbredSet " + "WHERE InbredSet.Name = '%s' AND " + "PublishFreeze.InbredSetId = InbredSet.Id AND " + "InfoFiles.InfoPageName = PublishFreeze.Name"), + 'other_pheno': ("SELECT PublishFreeze.Name " + "FROM PublishFreeze, InbredSet " + "WHERE InbredSet.Name = '%s' AND " + "PublishFreeze.InbredSetId = InbredSet.Id"), + 'geno': ("SELECT GenoFreeze.Id FROM GenoFreeze WHERE " + "GenoFreeze.Name = \"%s\" ") } dataset_name_mapping = { @@ -154,22 +164,22 @@ Publish or ProbeSet. E.g. if t in ['pheno', 'other_pheno']: group_name = name.replace("Publish", "") - results = g.db.execute(sql_query_mapping[t].format(group_name)).fetchone() + results = g.db.execute(sql_query_mapping[t] % group_name).fetchone() if results: self.datasets[name] = dataset_name_mapping[t] self.redis_instance.set("dataset_structure", json.dumps(self.datasets)) return True - return None def __call__(self, name): - if name not in self.datasets: for t in ["mrna_expr", "pheno", "other_pheno", "geno"]: - # This has side-effects, with the end result being a truth-y value + # This has side-effects, with the end result being a + # truth-y value if(self.set_dataset_key(t, name)): break - return self.datasets.get(name, None) # Return None if name has not been set + # Return None if name has not been set + return self.datasets.get(name, None) # Do the intensive work at startup one time only -- cgit v1.2.3