From 2373883fe76b18a11fb9914732611777839459f0 Mon Sep 17 00:00:00 2001
From: Peter Woolery <peter@research.bike>
Date: Sun, 10 May 2026 20:17:03 -0700
Subject: [PATCH] fix: exact-name fast path in matcher + save priceless produce
 in scraper
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Scraper: remove price guard in map_product so produce items without a
catalog price (e.g. Fresh Garlic, Lime sold by weight) are saved to
grocery_item with current_price=NULL rather than skipped.

Matcher:
- Add exact-name fast path: build a lowercase-trimmed name→index map
  and skip fuzzy search entirely when the ingredient name matches a
  grocery item exactly. Lime → Lime (confidence 1.0), Garlic → Fresh
  Garlic from fuzzy (confidence 1.0).
- Add exclusion words: juice, gelatin to prevent beverage/dessert
  products from matching cooking ingredients.
- Increase fuzzy candidate limit 20→100 so exact-name items buried in
  large tie groups are not missed.
- Add 'juice' to exclusion: prevents '100% Lime Juice' from winning
  over plain 'Lime'.

Result: all recipe ingredients now match correct Lucky CA products or
show '—' (no match); zero category cross-contamination remaining.

Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
---
 backend/app/scraper/lucky_ca_scraper.py |  4 ----
 backend/app/services/matcher.py         | 27 +++++++++++++++++++++++--
 2 files changed, 25 insertions(+), 6 deletions(-)

diff --git a/backend/app/scraper/lucky_ca_scraper.py b/backend/app/scraper/lucky_ca_scraper.py
index 6d2e58c..efec36a 100644
--- a/backend/app/scraper/lucky_ca_scraper.py
+++ b/backend/app/scraper/lucky_ca_scraper.py
@@ -259,10 +259,6 @@ class LuckyCaliforniaScraper:
         promo = price_block.get("promoArea") or {}
         sale_price, sale_unit = cls._parse_price(promo.get("promoText"))
 
-        if reg_price is None and sale_price is None:
-            # No usable price; skip rather than persist garbage.
-            return None
-
         unit = sale_unit or reg_unit
         is_on_sale = sale_price is not None and reg_price is not None and sale_price < reg_price
         # current_price = "what the customer pays today" → sale_price when on sale.
diff --git a/backend/app/services/matcher.py b/backend/app/services/matcher.py
index 7b37222..2c4f893 100644
--- a/backend/app/services/matcher.py
+++ b/backend/app/services/matcher.py
@@ -56,7 +56,7 @@ _EXCLUSION_WORDS = frozenset({
     "margarita", "rita", "cocktail", "beer", "ale", "lager", "cider", "malt",
     "wine", "spirits", "liquor",
     "vodka", "tequila", "whiskey", "rum", "gin", "bourbon",
-    "lemonade", "limeade", "seltzer", "soda",
+    "lemonade", "limeade", "seltzer", "soda", "juice",
     # Butter / spreads (prevents "Garlic & Herb Butter Spread" matching "Garlic")
     "butter", "spread", "margarine",
     # Prepared proteins / seafood-in-oil (prevents "Tuna in Olive Oil" matching "Olive Oil")
@@ -123,6 +123,11 @@ def run_match_job(
     ).delete(synchronize_session=False)
     db.flush()
 
+    # Build a lowercase-trimmed name → index map for O(1) exact-match lookup.
+    gi_exact: dict[str, int] = {
+        gi.name.strip().lower(): idx for idx, gi in enumerate(grocery_rows)
+    }
+
     ingredients = db.query(Ingredient).all()
     written = 0
 
@@ -136,6 +141,24 @@ def run_match_job(
             a.lower() for a in (ingredient.aliases or []) if a
         ]
 
+        # Fast path: exact name match beats all fuzzy candidates.
+        exact_idx = gi_exact.get(ingredient.name.strip().lower())
+        if exact_idx is not None:
+            stmt = (
+                _pg_insert(IngredientGroceryMatch.__table__)
+                .values(
+                    id=_uuid_mod.uuid4(),
+                    ingredient_id=ingredient.id,
+                    grocery_item_id=grocery_ids[exact_idx],
+                    confidence=Decimal("1.000"),
+                    source=IngredientMatchSource.AUTO,
+                )
+                .on_conflict_do_nothing(index_elements=["ingredient_id", "grocery_item_id"])
+            )
+            db.execute(stmt)
+            written += 1
+            continue
+
         best_idx: int | None = None
         best_combined = 0.0
 
@@ -144,7 +167,7 @@ def run_match_job(
                 query,
                 grocery_names_lower,
                 scorer=fuzz.partial_token_sort_ratio,
-                limit=20,
+                limit=100,
             )
             for _text, score, idx in results:
                 if score < threshold * 100: