Fix bug: Insert ProbeSets if they do not exist.

1 files changed, 13 insertions, 7 deletions

diff --git a/scripts/insert_data.py b/scripts/insert_data.py
index 7e27f93..f8b73f6 100644
--- a/scripts/insert_data.py
+++ b/scripts/insert_data.py
@@ -131,27 +131,33 @@ def insert_means(
     "Insert the means/averages data into the database"
     headings = read_file_headings(filepath)
     strains = strains_info(dbconn, headings[1:], speciesid)
-    check_strains(headings[1:], strains)
+    check_strains(headings[1:], strains),
+    probeset_query = (
+        "INSERT IGNORE INTO ProbeSet(ChipId, Name) "
+        "VALUES (%(ChipId)s, %(ProbeSetId)s) ")
     means_query = (
         "INSERT INTO ProbeSetData "
         "VALUES(%(ProbeSetDataId)s, %(StrainId)s, %(DataValue)s)")
     xref_query = (
         "INSERT INTO ProbeSetXRef(ProbeSetFreezeId, ProbeSetId, DataId) "
         "VALUES(%(ProbeSetFreezeId)s, %(ProbeSetId)s, %(ProbeSetDataId)s)")
-    the_means = (
-        {"ProbeSetFreezeId": datasetid, "ProbeSetDataId": data_id, **mean}
-        for data_id, mean in
-        enumerate(
-            read_datavalues(filepath, headings, strains),
-            start=(last_data_id(dbconn)+1)))
+    the_means = ({
+        "ProbeSetFreezeId": datasetid, "ProbeSetDataId": data_id,
+        "ChipId": platform_id, **mean
+    } for data_id, mean in enumerate(
+        read_datavalues(filepath, headings, strains),
+        start=(last_data_id(dbconn)+1)))
     with dbconn.cursor(cursorclass=DictCursor) as cursor:
         while True:
             means = tuple(take(the_means, 1000))
             if not bool(means):
                 break
+            print(__format_query__(probeset_query, means))
+            print()
             print(__format_query__(means_query, means))
             print()
             print(__format_query__(xref_query, means))
+            cursor.executemany(probeset_query, means)
             cursor.executemany(means_query, means)
             cursor.executemany(xref_query, means)
     return 0