Add modular TMDb-first movie pipeline and Discord bot

movie_pipeline.py

@@ -0,0 +1,163 @@
+#!/usr/bin/env python3
+from __future__ import annotations
+
+from datetime import date, timedelta
+
+from app_config import get_settings
+from data_sources.anilist_source import fetch_anilist_movie_by_search
+from data_sources.animeschedule_source import fetch_animeschedule_anime_by_title
+from data_sources.tmdb_source import fetch_tmdb_anime_movies
+
+
+def add_months(d: date, months: int) -> date:
+    year = d.year + ((d.month - 1 + months) // 12)
+    month = ((d.month - 1 + months) % 12) + 1
+    return date(year, month, 1)
+
+
+def format_date(d: date, locale: str) -> str:
+    if locale == "de-DE":
+        return f"{d.day:02d}.{d.month:02d}.{d.year}"
+    return d.isoformat()
+
+
+def select_title(anilist: dict, schedule: dict, tmdb: dict) -> str:
+    return (
+        (schedule.get("english_title") or "").strip()
+        or (anilist.get("title_english") or "").strip()
+        or (anilist.get("title_best") or "").strip()
+        or (tmdb.get("title") or "").strip()
+        or (tmdb.get("original_title") or "").strip()
+    )
+
+
+def structure_movie_record(tmdb_movie: dict, anilist: dict, schedule: dict, locale: str) -> dict:
+    release_date = tmdb_movie.get("release_date")
+    release = format_date(release_date, locale) if isinstance(release_date, date) else "n/a"
+
+    has_anilist = bool(anilist)
+    has_schedule = bool(schedule and (schedule.get("title") or schedule.get("english_title") or schedule.get("url")))
+
+    final_description = (
+        (tmdb_movie.get("overview") or "").strip()
+        or (anilist.get("description") or "").strip()
+        or (schedule.get("description") or "").strip()
+        or "n/a"
+    )
+
+    # User requirement: cover images should come from AniList.
+    cover_image = (anilist.get("cover_image") or "").strip()
+
+    genres_text = (
+        (tmdb_movie.get("genres_text") or "").strip()
+        or (anilist.get("genres_text") or "").strip()
+        or (schedule.get("genres") or "").strip()
+        or "n/a"
+    )
+
+    record = {
+        "title": select_title(anilist, schedule, tmdb_movie),
+        "title_english_anilist": (anilist.get("title_english") or "").strip(),
+        "title_anilist": (anilist.get("title_best") or "").strip(),
+        "title_schedule_english": (schedule.get("english_title") or "").strip(),
+        "title_romaji": (anilist.get("title_romaji") or "").strip() or "n/a",
+        "title_native": (anilist.get("title_native") or "").strip() or "n/a",
+        "studio": (anilist.get("studio_text") or "").strip() or (schedule.get("studios") or "").strip() or "n/a",
+        "genres": genres_text,
+        "tags": (anilist.get("tags_text") or "").strip() or "n/a",
+        "release": release,
+        "anilist_url": (anilist.get("anilist_url") or "").strip() or "n/a",
+        "format": (anilist.get("format") or "").strip() or (schedule.get("format") or "").strip() or "MOVIE",
+        "episodes": anilist.get("episodes") or "n/a",
+        "duration": anilist.get("duration") or "n/a",
+        "source": (anilist.get("source") or "").strip() or "n/a",
+        "cover_image": cover_image,
+        "description": final_description,
+        "schedule_url": (schedule.get("url") or "").strip(),
+        "schedule_title": (schedule.get("title") or "").strip(),
+        "release_source": "TMDb DE",
+        "source_presence": {
+            "tmdb": True,
+            "anilist": has_anilist,
+            "animeschedule": has_schedule,
+        },
+        "tmdb": tmdb_movie,
+        "anilist": anilist,
+        "animeschedule": schedule,
+    }
+    return record
+
+
+def sort_and_structure_movies(tmdb_movies: list[dict], locale: str, schedule_token: str) -> list[dict]:
+    records = []
+    anilist_cache: dict[str, dict | None] = {}
+    schedule_cache: dict[str, dict] = {}
+
+    sorted_tmdb = sorted(
+        tmdb_movies,
+        key=lambda item: item.get("release_date") or date.min,
+        reverse=True,
+    )
+
+    for tmdb_movie in sorted_tmdb:
+        anilist = None
+        for candidate in ((tmdb_movie.get("title") or "").strip(), (tmdb_movie.get("original_title") or "").strip()):
+            if not candidate:
+                continue
+            anilist = fetch_anilist_movie_by_search(candidate, anilist_cache)
+            if anilist:
+                break
+        anilist = anilist or {}
+
+        schedule_candidates = [
+            (anilist.get("title_english") or "").strip(),
+            (anilist.get("title_best") or "").strip(),
+            (tmdb_movie.get("title") or "").strip(),
+            (tmdb_movie.get("original_title") or "").strip(),
+        ]
+
+        best_schedule = {}
+        best_schedule_score = 0
+        seen_schedule_candidates = set()
+        for schedule_candidate in schedule_candidates:
+            if not schedule_candidate:
+                continue
+            norm_key = schedule_candidate.lower()
+            if norm_key in seen_schedule_candidates:
+                continue
+            seen_schedule_candidates.add(norm_key)
+
+            candidate_result = fetch_animeschedule_anime_by_title(schedule_candidate, schedule_token, schedule_cache)
+            score = int(candidate_result.get("match_score") or 0)
+            if score > best_schedule_score:
+                best_schedule = candidate_result
+                best_schedule_score = score
+
+            # Prefer results that have explicit English title when scores are tied.
+            if score == best_schedule_score and score > 0:
+                if not (best_schedule.get("english_title") or "").strip() and (candidate_result.get("english_title") or "").strip():
+                    best_schedule = candidate_result
+
+        schedule = best_schedule if best_schedule_score > 0 else {}
+
+        records.append(structure_movie_record(tmdb_movie, anilist, schedule, locale))
+
+    return records
+
+
+def get_upcoming_movie_records(
+    locale: str,
+    today: date | None = None,
+    anime_schedule_token: str | None = None,
+    tmdb_read_access_token: str | None = None,
+) -> list[dict]:
+    settings = get_settings()
+    schedule_token = (anime_schedule_token or settings.animeschedule_api_token).strip()
+    tmdb_token = (tmdb_read_access_token or settings.tmdb_read_access_token).strip()
+
+    current_day = today or date.today()
+    month_start = date(current_day.year, current_day.month, 1)
+    end_date = add_months(month_start, 2) - timedelta(days=1)
+
+    tmdb_movies = fetch_tmdb_anime_movies(current_day, end_date, tmdb_token, language=locale)
+    return sort_and_structure_movies(tmdb_movies, locale, schedule_token)