fix: correct DataFrame column filtering and enhance debug output

- Fix KeyError in generate_theme_catalog.py: use isCommander column correctly
- DataFrame.get() doesn't work like dict.get() - use column name directly
- Enhanced debug step to print full row data for better diagnostics

.github/workflows/build-similarity-cache.yml

@@ -109,25 +109,23 @@ jobs:
           print(f'Columns: {list(df.columns)}')
           print('')
           
-          # Show first 10 rows with their themeTags
+          # Show first 5 rows completely
-          print('First 10 cards with themeTags:')
+          print('First 5 complete rows:')
-          print('=' * 80)
+          print('=' * 100)
-          for idx, row in df.head(10).iterrows():
+          for idx, row in df.head(5).iterrows():
-              name = row.get('name', 'UNKNOWN')
+              print(f'Row {idx}:')
-              tags = row.get('themeTags', [])
+              for col in df.columns:
-              tag_count = len(tags) if isinstance(tags, list) else 0
+                  value = row[col]
-              print(f'{idx}: {name}')
+                  if isinstance(value, (list, tuple)) or hasattr(value, '__array__'):
-              print(f'   Type: {type(tags).__name__}')
+                      # For array-like, show type and length
-              print(f'   Count: {tag_count}')
+                      try:
-              if tag_count > 0:
+                          length = len(value)
-                  # Show first 5 tags
+                          print(f'  {col}: {type(value).__name__}[{length}] = {value}')
-                  sample = tags[:5] if tag_count > 5 else tags
+                      except:
-                  print(f'   Tags: {sample}')
+                          print(f'  {col}: {type(value).__name__} = {value}')
-                  if tag_count > 5:
+                  else:
-                      print(f'   ... and {tag_count - 5} more')
+                      print(f'  {col}: {value}')
-              else:
+              print('-' * 100)
-                  print(f'   Tags: (empty)')
-              print('')
           "
       
       - name: Generate theme catalog

code/scripts/generate_theme_catalog.py

@@ -247,9 +247,13 @@ def build_theme_catalog(
         all_cards_parquet, theme_variants=theme_variants
     )
     
-    # For commander counts, filter all_cards by is_commander column
+    # For commander counts, filter all_cards by isCommander column
     df_commanders = pd.read_parquet(all_cards_parquet)
-    df_commanders = df_commanders[df_commanders.get('is_commander', False)]
+    if 'isCommander' in df_commanders.columns:
+        df_commanders = df_commanders[df_commanders['isCommander']]
+    else:
+        # Fallback: assume all cards could be commanders if column missing
+        pass
     commander_counts = Counter()
     for tags in df_commanders['themeTags'].tolist():
         if tags is None or (isinstance(tags, float) and pd.isna(tags)):