Backend scores updates on products

scripts/sync_nutrition_scores.py

@@ -0,0 +1,116 @@
+#!/usr/bin/env python3
+"""
+Copy `nutrition_insights` scores onto the brand tables and the seed catalogs.
+
+WHY THIS EXISTS
+---------------
+`nutrition_score` and `health_score` are stored on `nutrition_insights`, keyed
+by (brand, image_id). Nothing joined that table to `brand_*`, so anyone reading
+product rows straight out of Postgres - which is how the merchant console does
+it - could not see a score at all. The two columns now exist on every brand
+table; this is what fills them the first time.
+
+Afterwards it maintains itself: `enrich_all_products` and the `/upload/nutrition`
+endpoint both call `nutrition_score_sync.sync_after_write` when they finish, so
+the mirror stays within one run of the source. Run this by hand after anything
+that edits `nutrition_insights` outside those two paths - notably
+`purge_non_consumable_nutrition`, which deletes rows and would otherwise leave
+stale scores behind on the brand tables.
+
+SAFE TO RE-RUN
+--------------
+Every statement is guarded with `IS DISTINCT FROM`, so a second run changes
+nothing and reports 0. That makes "0 rows changed" a real assertion that the
+mirror is current, not an artefact of the script doing nothing.
+
+USAGE
+-----
+    python -m scripts.sync_nutrition_scores                  # dry run
+    python -m scripts.sync_nutrition_scores --apply
+    python -m scripts.sync_nutrition_scores --apply --brands "Cadbury,Amul"
+    python -m scripts.sync_nutrition_scores --apply --db-only
+    python -m scripts.sync_nutrition_scores --apply --json-only
+
+Dry run by default. `--apply` writes.
+"""
+from __future__ import annotations
+
+import argparse
+import logging
+import sys
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parents[1]))
+
+from app.infrastructure.settings import DB_HOST, DB_NAME  # noqa: E402
+from app.services.nutrition_score_sync import (  # noqa: E402
+    sync_scores_to_brand_tables,
+    sync_scores_to_seed_files,
+)
+
+logging.basicConfig(level=logging.INFO, format="%(message)s")
+logger = logging.getLogger(__name__)
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(
+        description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
+    parser.add_argument("--apply", action="store_true",
+                        help="commit the changes (default is a dry run)")
+    parser.add_argument("--dry-run", action="store_true",
+                        help="explicit no-op; this is already the default")
+    parser.add_argument("--brands", default=None,
+                        help="comma-separated brand names to limit the run to")
+    parser.add_argument("--db-only", action="store_true",
+                        help="brand tables only, leave the seed JSON alone")
+    parser.add_argument("--json-only", action="store_true",
+                        help="seed JSON only, leave the database alone")
+    args = parser.parse_args()
+
+    if args.db_only and args.json_only:
+        parser.error("--db-only and --json-only are mutually exclusive")
+
+    apply = args.apply and not args.dry_run
+    brands = [b.strip() for b in (args.brands or "").split(",") if b.strip()] or None
+
+    logger.info("Target database: %s / %s", DB_HOST, DB_NAME)
+    logger.info("Mode: %s", "APPLY - this writes" if apply
+                else "DRY RUN - nothing is written")
+    if brands:
+        logger.info("Limited to: %s", ", ".join(brands))
+    logger.info("")
+
+    total = 0
+
+    if not args.json_only:
+        logger.info("Brand tables")
+        logger.info("-" * 52)
+        changed = sync_scores_to_brand_tables(brands, dry_run=not apply)
+        for brand, count in sorted(changed.items(), key=lambda kv: -kv[1]):
+            logger.info("  %-38s %5d", brand, count)
+        if not changed:
+            logger.info("  nothing to change - the mirror is current")
+        total += sum(changed.values())
+        logger.info("")
+
+    if not args.db_only:
+        logger.info("Seed catalogs")
+        logger.info("-" * 52)
+        changed = sync_scores_to_seed_files(brands, dry_run=not apply)
+        for name, count in sorted(changed.items(), key=lambda kv: -kv[1]):
+            logger.info("  %-38s %5d", name, count)
+        if not changed:
+            logger.info("  nothing to change - the files are current")
+        total += sum(changed.values())
+        logger.info("")
+
+    if not apply and total:
+        logger.info("Re-run with --apply to write %d change(s).", total)
+    elif apply:
+        logger.info("Done. %d change(s) written.", total)
+
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())