feat: Use of openpyxl for simple ordinal to letter conversion

Author: Alan Christie

poetry.lock

@@ -33,6 +33,7 @@ gunicorn = "^21.2.0"
 im-squonk2-client = "^1.22.2"
 mozilla_django_oidc = "^3.0.0"
 mysql-connector-python = "^8.1.0"
+openpyxl = "^3.1.2"
 pandas = "^1.5.3"
 pandoc = "^2.3"
 psutil = "^5.9.5"

pyproject.toml

@@ -7,6 +7,8 @@
 import zipfile
 from typing import Any, Dict, List, Optional, Tuple
 
+from openpyxl.utils import get_column_letter
+
 os.environ.setdefault("DJANGO_SETTINGS_MODULE", "fragalysis.settings")
 import django
 
@@ -500,19 +502,19 @@ def task(self) -> ComputedSet:
                 len(existing_sets),
                 latest_ordinal,
             )
+        # ordinals are 1-based
         new_ordinal: int = latest_ordinal + 1
 
         # The computed set "name" consists of the "method",
         # today's date and a 2-digit ordinal. The ordinal
         # is used to distinguish between computed sets uploaded
         # with the same method on the same day.
-        cs_name: str = f"{truncated_submitter_method}-{str(today)}-{new_ordinal:0>2}"
+        assert new_ordinal > 0
+        cs_name: str = f"{truncated_submitter_method}-{str(today)}-{get_column_letter(new_ordinal)}"
         logger.info('Creating new ComputedSet "%s"', cs_name)
 
         computed_set: ComputedSet = ComputedSet()
-        computed_set.name = (
-            f"{truncated_submitter_method}-{str(today)}-{new_ordinal:0>2}"
-        )
+        computed_set.name = cs_name
         computed_set.md_ordinal = new_ordinal
         computed_set.upload_date = today
         computed_set.method = truncated_submitter_method

viewer/cset_upload.py

@@ -5,13 +5,12 @@
 @author: Warren
 Script to check sdf file format for Fragalysis upload
 """
-import datetime
 import logging
 
 import validators
 from rdkit import Chem
 
-from viewer.models import ComputedSet, SiteObservation
+from viewer.models import SiteObservation
 
 logger = logging.getLogger(__name__)
 
@@ -24,6 +23,8 @@ def check_property_descriptions():
 
 
 def check_compound_set(description_mol, validate_dict, update=None):
+    del update
+
     # Must have a 'generation_date'
     if not description_mol.HasProp('generation_date'):
         validate_dict = add_warning(
@@ -45,28 +46,6 @@ def check_compound_set(description_mol, validate_dict, update=None):
         )
         return validate_dict
 
-    submitter_dict = {
-        'submitter__name': description_mol.GetProp('submitter_name'),
-        'submitter__email': description_mol.GetProp('submitter_email'),
-        'submitter__institution': description_mol.GetProp('submitter_institution'),
-        'submitter__generation_date': datetime.date(
-            int(y_m_d[0]), int(y_m_d[1]), int(y_m_d[2])
-        ),
-        'submitter__method': description_mol.GetProp('method'),
-    }
-
-    query = ComputedSet.objects.filter(**submitter_dict)
-
-    if len(query) != 0 and not update:
-        validate_dict = add_warning(
-            molecule_name='File error',
-            field='compound set',
-            warning_string="A ComputedSet with the name "
-            + query[0].name
-            + " already exists (change method name in blank mol method field)",
-            validate_dict=validate_dict,
-        )
-
     return validate_dict