Move difference computation to `run()` function.

1 files changed, 46 insertions, 51 deletions

diff --git a/scripts/phenotypes_bulk_edit.py b/scripts/phenotypes_bulk_edit.py
index 1d8124e..bbce60d 100644
--- a/scripts/phenotypes_bulk_edit.py
+++ b/scripts/phenotypes_bulk_edit.py
@@ -10,6 +10,7 @@ from MySQLdb.cursors import DictCursor
 
 from gn_libs import jobs, mysqldb, sqlite3
 
+from uploader.phenotypes.models import phenotypes_data_by_ids
 import uploader.publications.pubmed as pmed
 from uploader.publications.misc import publications_differences
 from uploader.phenotypes.views import BULK_EDIT_COMMON_FIELDNAMES
@@ -87,52 +88,6 @@ def descriptions_differences(file_data, db_data) -> dict[str, str]:
     return diff
 
 
-def compute_differences(
-        conn,
-        file_contents,
-        pheno_ids,
-        pheno_xref_ids,
-        pubmed_ids
-) -> tuple[tuple[dict, ...], tuple[dict, ...], tuple[dict, ...]]:
-    """Compute differences between data in DB and edited data."""
-    logger.info("Computing differences.")
-    # 1. Basic Phenotype data differences
-    #    a. Descriptions differences
-    desc_diff = descriptions_differences(
-        file_contents, __fetch_phenotypes__(conn, pheno_ids))
-    logger.debug("DESCRIPTIONS DIFFERENCES: %s", desc_diff)
-
-    #    b. Publications differences
-    db_publications = fetch_phenotype_publications(conn, pheno_xref_ids)
-
-    _pubmed_map = {
-        (int(row["PubMed_ID"]) if bool(row["PubMed_ID"]) else None): f"{row['phenotype_id']}::{row['xref_id']}"
-        for row in file_contents
-    }
-    _pub_id_map = {
-        f"{pub['PhenotypeId']}::{pub['xref_id']}": pub["PublicationId"]
-        for pub in db_publications
-    }
-
-    new_publications = update_publications(
-        conn, tuple({
-            **pub, "publication_id": _pub_id_map[_pubmed_map[pub["pubmed_id"]]]
-        } for pub in pmed.fetch_publications(tuple(
-            pubmed_id for pubmed_id in pubmed_ids
-            if pubmed_id not in
-            tuple(row["PubMed_ID"] for row in db_publications)))))
-    logger.debug("New Publications: %s", new_publications)
-    pub_diff = publications_differences(
-        file_contents, db_publications, {
-            row["PubMed_ID" if "PubMed_ID" in row else "pubmed_id"]: row[
-                "PublicationId" if "PublicationId" in row else "publication_id"]
-            for row in db_publications + new_publications})
-    logger.debug("Publications diff: %s", pub_diff)
-    # 2. Data differences
-    # data_diff = data_differences(...)
-    pass
-
-
 def update_descriptions():
     """Update descriptions in the database"""
     logger.info("Updating descriptions")
@@ -216,11 +171,51 @@ def run(conn, job):
     check_ids(conn, pheno_xref_ids)
     check_for_mandatory_fields()
     # stop running here if any errors are found.
-    compute_differences(conn,
-                        file_contents,
-                        pheno_ids,
-                        pheno_xref_ids,
-                        pubmed_ids)
+
+    ### Compute differences
+    logger.info("Computing differences.")
+    # 1. Basic Phenotype data differences
+    #    a. Descriptions differences
+    _desc_diff = descriptions_differences(
+        file_contents, __fetch_phenotypes__(conn, pheno_ids))
+    logger.debug("DESCRIPTIONS DIFFERENCES: %s", _desc_diff)
+
+    #    b. Publications differences
+    _db_publications = fetch_phenotype_publications(conn, pheno_xref_ids)
+    logger.debug("DB PUBLICATIONS: %s", _db_publications)
+
+    _pubmed_map = {
+        (int(row["PubMed_ID"]) if bool(row["PubMed_ID"]) else None): f"{row['phenotype_id']}::{row['xref_id']}"
+        for row in file_contents
+    }
+    _pub_id_map = {
+        f"{pub['PhenotypeId']}::{pub['xref_id']}": pub["PublicationId"]
+        for pub in _db_publications
+    }
+
+    _new_publications = update_publications(
+        conn, tuple({
+            **pub, "publication_id": _pub_id_map[_pubmed_map[pub["pubmed_id"]]]
+        } for pub in pmed.fetch_publications(tuple(
+            pubmed_id for pubmed_id in pubmed_ids
+            if pubmed_id not in
+            tuple(row["PubMed_ID"] for row in _db_publications)))))
+    _pub_diff = publications_differences(
+        file_contents, _db_publications, {
+            row["PubMed_ID" if "PubMed_ID" in row else "pubmed_id"]: row[
+                "PublicationId" if "PublicationId" in row else "publication_id"]
+            for row in _db_publications + _new_publications})
+    logger.debug("Publications diff: %s", _pub_diff)
+    # 2. Data differences
+    # _db_pheno_data = phenotypes_data_by_ids(conn, tuple({
+    #     "population_id": job["metadata"]["population-id"],
+    #     "phenoid": row[0],
+    #     "xref_id": row[1]
+    # } for row in pheno_xref_ids))
+    # logger.debug("Phenotype Data in Database: %s", _db_pheno_data)
+
+    # data_diff = data_differences(...)
+    ### END: Compute differences
     update_descriptions()
     link_publications()
     update_values()