Andreas/mtg_python_deckbuilder
1
0
Fork 0
mirror of https://github.com/mwisnowski/mtg_python_deckbuilder.git synced 2025-12-16 15:40:12 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix: update generate_theme_catalog to use processed/ directory

Browse source
This commit is contained in:
matt 2025-10-18 22:11:46 -07:00
parent 0e19824372
commit 9e6c3e66e9
1 changed files with 28 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

37
code/scripts/generate_theme_catalog.py
View file

@ -245,21 +245,40 @@ def build_theme_catalog(
used_parquet = False used_parquet = False
if use_parquet and HAS_PARQUET_SUPPORT: if use_parquet and HAS_PARQUET_SUPPORT:
try: try:
# Use dedicated parquet files (matches CSV structure exactly) # Use processed parquet files (M4 migration)
parquet_dir = csv_directory.parent / "card_files" parquet_dir = csv_directory.parent / "card_files" / "processed"
# Load commander counts directly from commander_cards.parquet # Load all card counts from all_cards.parquet (includes commanders)
commander_parquet = parquet_dir / "commander_cards.parquet"
commander_counts = _load_theme_counts_from_parquet(
commander_parquet, theme_variants=theme_variants
)
# Load all card counts from all_cards.parquet to include all themes
all_cards_parquet = parquet_dir / "all_cards.parquet" all_cards_parquet = parquet_dir / "all_cards.parquet"
card_counts = _load_theme_counts_from_parquet( card_counts = _load_theme_counts_from_parquet(
all_cards_parquet, theme_variants=theme_variants all_cards_parquet, theme_variants=theme_variants
) )
# For commander counts, filter all_cards by is_commander column
if all_cards_parquet.exists() and pd is not None:
df_commanders = pd.read_parquet(all_cards_parquet)
df_commanders = df_commanders[df_commanders.get('is_commander', False)]
commander_counts = Counter()
for tags in df_commanders['themeTags'].tolist():
if tags is None or (isinstance(tags, float) and pd.isna(tags)):
continue
from code.deck_builder.theme_catalog_loader import parse_theme_tags, normalize_theme_display, canonical_key
parsed = parse_theme_tags(tags)
if not parsed:
continue
seen = set()
for tag in parsed:
display = normalize_theme_display(tag)
if not display:
continue
key = canonical_key(display)
if key not in seen:
seen.add(key)
commander_counts[key] += 1
theme_variants[key].add(display)
else:
commander_counts = Counter()
used_parquet = True used_parquet = True
print("✓ Loaded theme data from parquet files") print("✓ Loaded theme data from parquet files")
print(f" - Commanders: {len(commander_counts)} themes") print(f" - Commanders: {len(commander_counts)} themes")