1 files changed, 38 insertions, 6 deletions
diff --git a/uploader/phenotypes/models.py b/uploader/phenotypes/models.py
index eb5a189..9324601 100644
--- a/uploader/phenotypes/models.py
+++ b/uploader/phenotypes/models.py
@@ -1,6 +1,7 @@
 """Database and utility functions for phenotypes."""
 from typing import Optional
 from functools import reduce
+from datetime import datetime
 
 import MySQLdb as mdb
 from MySQLdb.cursors import Cursor, DictCursor
@@ -78,7 +79,7 @@ def __phenotype_se__(cursor: Cursor,
                      xref_id: str) -> dict:
     """Fetch standard-error values (if they exist) for a phenotype."""
     _sequery = (
-        "SELECT pxr.Id AS xref_id, pxr.DataId, pse.error, nst.count "
+        "SELECT pxr.Id AS xref_id, pxr.DataId, str.Id AS StrainId, pse.error, nst.count "
         "FROM Phenotype AS pheno "
         "INNER JOIN PublishXRef AS pxr ON pheno.Id=pxr.PhenotypeId "
         "INNER JOIN PublishSE AS pse ON pxr.DataId=pse.DataId "
@@ -90,7 +91,12 @@ def __phenotype_se__(cursor: Cursor,
         "WHERE (str.SpeciesId, pxr.InbredSetId, pf.Id, pxr.Id)=(%s, %s, %s, %s)")
     cursor.execute(_sequery,
                    (species_id, population_id, dataset_id, xref_id))
-    return {row["xref_id"]: dict(row) for row in cursor.fetchall()}
+    return {(row["DataId"], row["StrainId"]): {
+        "xref_id": row["xref_id"],
+        "DataId": row["DataId"],
+        "error": row["error"],
+        "count": row["count"]
+    } for row in cursor.fetchall()}
 
 def __organise_by_phenotype__(pheno, row):
     """Organise disparate data rows into phenotype 'objects'."""
@@ -105,11 +111,10 @@ def __organise_by_phenotype__(pheno, row):
             "Units": row["Units"],
             "Pre_publication_abbreviation": row["Pre_publication_abbreviation"],
             "Post_publication_abbreviation": row["Post_publication_abbreviation"],
+            "xref_id": row["pxr.Id"],
             "data": {
-                #TOD0: organise these by DataId and StrainId
                 **(_pheno["data"] if bool(_pheno) else {}),
-                row["pxr.Id"]: {
-                    "xref_id": row["pxr.Id"],
+                (row["DataId"], row["StrainId"]): {
                     "DataId": row["DataId"],
                     "mean": row["mean"],
                     "Locus": row["Locus"],
@@ -168,7 +173,7 @@ def phenotype_by_id(
                                      species_id,
                                      population_id,
                                      dataset_id,
-                                     xref_id)).items())
+                                     xref_id)).values())
             }
         if bool(_pheno) and len(_pheno.keys()) > 1:
             raise Exception(
@@ -198,3 +203,30 @@ def phenotypes_data(conn: mdb.Connection,
         cursor.execute(_query, (population_id, dataset_id))
         debug_query(cursor)
         return tuple(dict(row) for row in cursor.fetchall())
+
+
+def save_new_dataset(cursor: Cursor,
+                     population_id: int,
+                     dataset_name: str,
+                     dataset_fullname: str,
+                     dataset_shortname: str) -> dict:
+    """Create a new phenotype dataset."""
+    params = {
+        "population_id": population_id,
+        "dataset_name": dataset_name,
+        "dataset_fullname": dataset_fullname,
+        "dataset_shortname": dataset_shortname,
+        "created": datetime.now().date().isoformat(),
+        "public": 2,
+        "confidentiality": 0,
+        "users": None
+    }
+    cursor.execute(
+        "INSERT INTO PublishFreeze(Name, FullName, ShortName, CreateTime, "
+        "public, InbredSetId, confidentiality, AuthorisedUsers) "
+        "VALUES(%(dataset_name)s, %(dataset_fullname)s, %(dataset_shortname)s, "
+        "%(created)s, %(public)s, %(population_id)s, %(confidentiality)s, "
+        "%(users)s)",
+        params)
+    debug_query(cursor)
+    return {**params, "Id": cursor.lastrowid}