manga matching and WebApp

src/ComicInfoBuilder.py

@@ -48,6 +48,7 @@ from MangadexVolumeResolver import MangaDexVolumeResolver
 from MangaBakaWorksResolver import MangaBakaWorksResolver
 from MALResolver import MALResolver
 from AniListResolver import AniListResolver
+from MatchesCache import MatchesCache
 
 try:
     from PIL import Image
@@ -170,7 +171,8 @@ class ComicInfoBuilder:
                  volume_resolver: "MangaDexVolumeResolver | None" = None,
                  works_resolver: "MangaBakaWorksResolver | None" = None,
                  mal_resolver: "MALResolver | None" = None,
-                 al_resolver: "AniListResolver | None" = None):
+                 al_resolver: "AniListResolver | None" = None,
+                 matches_cache: "MatchesCache | None" = None):
         if not manga_title or not str(manga_title).strip():
             raise ValueError("manga_title must not be empty.")
 
@@ -197,6 +199,7 @@ class ComicInfoBuilder:
             request_timeout=request_timeout)
         self._al_resolver = al_resolver or AniListResolver(
             request_timeout=request_timeout)
+        self._matches_cache = matches_cache
 
         self._metadata: "dict | None" = None
         self._pages: list[dict] = []
@@ -353,14 +356,43 @@ class ComicInfoBuilder:
         return series
 
     def _search_best_series(self, title: str):
-        """Searches for `title` and returns the best matching series entry."""
+        """
+        Resolves `title` to a MangaBaka series.
+
+        Lookup order:
+          1. matches.json cache (if attached) — uses the stored series ID
+             to fetch the full series, skipping the search step entirely.
+          2. Fresh MangaBaka search — top hit. The match is persisted to
+             matches.json before being returned so it survives a crash.
+        """
+        if self._matches_cache is not None:
+            cached = self._matches_cache.get(title)
+            if cached and cached.get("mangabakaId"):
+                try:
+                    return self._fetch_series_by_id(cached["mangabakaId"])
+                except Exception as exc:
+                    print(f"[ComicInfoBuilder] cached id "
+                          f"{cached['mangabakaId']} for {title!r} failed "
+                          f"({exc}); falling back to fresh search",
+                          flush=True)
+
         url = f"{self.api_base_url}/series/search"
         resp = self._session.get(
             url, params={"q": title, "page": 1, "limit": 1},
             timeout=self.request_timeout)
         resp.raise_for_status()
         data = resp.json().get("data") or []
-        return data[0] if data else None
+        series = data[0] if data else None
+
+        if series and self._matches_cache is not None:
+            self._matches_cache.add(
+                title,
+                mangabaka_id=series.get("id"),
+                mangabaka_name=series.get("title") or "",
+                image_url=_pick_cover_url(series.get("cover")),
+            )
+
+        return series
 
     def _fetch_series_by_id(self, series_id) -> dict:
         url = f"{self.api_base_url}/series/{series_id}"