Add Doris Kitchen household recipe app

home/doris-kitchen/app/services/recipe_search.py

@@ -0,0 +1,144 @@
+from __future__ import annotations
+
+import xml.etree.ElementTree as ET
+from typing import Any
+from urllib import parse, request
+
+from app.config import settings
+from app.services.recipe_importer import normalize_recipe_from_url
+from app.services.repository import repo
+from app.services.scoring import extract_ingredients, extract_tags, score_recipe
+from app.services.store import JsonStore
+
+
+SEARCH_URL = "https://www.bing.com/search?format=rss&q="
+ENGLISH_HINT = " site:allrecipes.com OR site:budgetbytes.com OR site:seriouseats.com OR site:recipetineats.com OR site:spendwithpennies.com OR site:delish.com OR site:food.com"
+ALLOWED_DOMAINS = (
+    "allrecipes.com",
+    "budgetbytes.com",
+    "seriouseats.com",
+    "recipetineats.com",
+    "spendwithpennies.com",
+    "delish.com",
+    "food.com",
+)
+FALLBACK_SEED_URLS = [
+    "https://www.budgetbytes.com/curry-beef-with-peas/",
+    "https://www.recipetineats.com/asian-beef-bowls/",
+    "https://www.spendwithpennies.com/ground-beef-stroganoff/",
+    "https://www.budgetbytes.com/chicken-and-rice-skillet/",
+]
+
+
+class RecipeSearchService:
+    def __init__(self) -> None:
+        self.store = JsonStore()
+
+    def _search_page(self, query: str) -> list[dict[str, str]]:
+        req = request.Request(
+            SEARCH_URL + parse.quote(query + ENGLISH_HINT),
+            headers={
+                "User-Agent": "Mozilla/5.0 (Doris Kitchen)",
+                "Accept-Language": "en-US,en;q=0.9",
+            },
+        )
+        with request.urlopen(req, timeout=30) as resp:
+            rss_text = resp.read().decode("utf-8", "replace")
+        root = ET.fromstring(rss_text)
+        results: list[dict[str, str]] = []
+        for item in root.findall("./channel/item"):
+            href = str(item.findtext("link") or "").strip()
+            title = str(item.findtext("title") or "").strip()
+            snippet = str(item.findtext("description") or "").strip()
+            host = parse.urlparse(href).netloc.lower()
+            if not href.startswith("http"):
+                continue
+            if not any(host.endswith(domain) or host == domain for domain in ALLOWED_DOMAINS):
+                continue
+            results.append({"url": href, "title": title, "snippet": snippet})
+        deduped: list[dict[str, str]] = []
+        seen: set[str] = set()
+        for item in results:
+            if item["url"] in seen:
+                continue
+            seen.add(item["url"])
+            deduped.append(item)
+        return deduped
+
+    def search(self, query: str, limit: int | None = None) -> list[dict[str, Any]]:
+        limit = limit or settings.search_result_limit
+        profile = repo.get_profile()
+        results: list[dict[str, Any]] = []
+        for hit in self._search_page(query):
+            if len(results) >= limit:
+                break
+            try:
+                payload = normalize_recipe_from_url(hit["url"], timeout=12)
+            except Exception:
+                continue
+            score, reasons = score_recipe(query, payload, profile)
+            suggestion = {
+                "id": self.store.new_id("suggestion"),
+                "title": payload.get("name") or hit["title"],
+                "url": hit["url"],
+                "source_query": query,
+                "source_title": hit["title"],
+                "source_snippet": hit["snippet"],
+                "payload": payload,
+                "ingredients": extract_ingredients(payload),
+                "tags": extract_tags(payload),
+                "score": score,
+                "score_reasons": reasons,
+                "status": "suggested",
+                "created_at": self.store.now(),
+                "updated_at": self.store.now(),
+                "kitchenowl_status": "not-imported",
+                "ready_for_auto_import": bool(profile.get("auto_import_enabled") and score >= int(profile.get("auto_import_threshold", settings.auto_import_threshold))),
+            }
+            repo.upsert_suggestion(suggestion)
+            results.append(suggestion)
+        return results
+
+    def _fallback_seed(self) -> list[dict[str, Any]]:
+        profile = repo.get_profile()
+        results: list[dict[str, Any]] = []
+        for url in FALLBACK_SEED_URLS:
+            try:
+                payload = normalize_recipe_from_url(url, timeout=12)
+            except Exception:
+                continue
+            score, reasons = score_recipe("seeded household dinner", payload, profile)
+            suggestion = {
+                "id": self.store.new_id("suggestion"),
+                "title": payload.get("name") or url,
+                "url": url,
+                "source_query": "seeded household dinner",
+                "source_title": payload.get("name") or url,
+                "source_snippet": "Curated fallback suggestion for initial household testing.",
+                "payload": payload,
+                "ingredients": extract_ingredients(payload),
+                "tags": extract_tags(payload),
+                "score": score,
+                "score_reasons": reasons,
+                "status": "suggested",
+                "created_at": self.store.now(),
+                "updated_at": self.store.now(),
+                "kitchenowl_status": "not-imported",
+                "ready_for_auto_import": False,
+            }
+            repo.upsert_suggestion(suggestion)
+            results.append(suggestion)
+        return results
+
+    def seed_queries(self) -> list[dict[str, Any]]:
+        profile = repo.get_profile()
+        queries = profile.get("seed_queries") or []
+        seeded: list[dict[str, Any]] = []
+        for query in queries:
+            seeded.extend(self.search(query, limit=2))
+        if not seeded:
+            seeded = self._fallback_seed()
+        return seeded
+
+
+recipe_search = RecipeSearchService()