finalising the Querries and the rest
This commit is contained in:
ProgrammGamer
@@ -4,30 +4,13 @@ from __future__ import annotations
|
||||
import json
|
||||
import urllib.error
|
||||
import urllib.request
|
||||
from datetime import date
|
||||
from datetime import date, timedelta
|
||||
|
||||
ANILIST_URL = "https://graphql.anilist.co"
|
||||
|
||||
|
||||
def normalize_title(text: str) -> str:
|
||||
cleaned = "".join(ch.lower() if ch.isalnum() else " " for ch in (text or ""))
|
||||
return " ".join(cleaned.split())
|
||||
|
||||
|
||||
def safe_date(sd: dict | None) -> date | None:
|
||||
if not sd:
|
||||
return None
|
||||
year = sd.get("year") or 0
|
||||
month = sd.get("month") or 0
|
||||
day = sd.get("day") or 0
|
||||
if year <= 0 or month <= 0:
|
||||
return None
|
||||
if day <= 0:
|
||||
day = 1
|
||||
try:
|
||||
return date(year, month, day)
|
||||
except ValueError:
|
||||
return None
|
||||
def yyyymmdd(value: date) -> int:
|
||||
return value.year * 10000 + value.month * 100 + value.day
|
||||
|
||||
|
||||
def post_graphql(query: str, variables: dict) -> dict:
|
||||
@@ -46,78 +29,69 @@ def post_graphql(query: str, variables: dict) -> dict:
|
||||
raise RuntimeError(f"HTTP {exc.code}: {body}") from exc
|
||||
|
||||
|
||||
def pick_best_title(title: dict) -> str:
|
||||
english = (title.get("english") or "").strip()
|
||||
if english:
|
||||
return english
|
||||
romaji = (title.get("romaji") or "").strip()
|
||||
if romaji:
|
||||
return romaji
|
||||
native = (title.get("native") or "").strip()
|
||||
return native or ""
|
||||
|
||||
|
||||
def map_anilist_media(media: dict | None) -> dict:
|
||||
def map_anilist_movie(media: dict | None) -> dict:
|
||||
media = media or {}
|
||||
title = media.get("title") or {}
|
||||
start_date = media.get("startDate") or {}
|
||||
studios = ((media.get("studios") or {}).get("nodes") or [])
|
||||
studio_names = [((node or {}).get("name") or "").strip() for node in studios]
|
||||
studio_names = [name for name in studio_names if name]
|
||||
|
||||
genres = [str(item).strip() for item in (media.get("genres") or [])]
|
||||
genres = [g for g in genres if g]
|
||||
studio_names = []
|
||||
for node in studios:
|
||||
name = ((node or {}).get("name") or "").strip()
|
||||
if name:
|
||||
studio_names.append(name)
|
||||
|
||||
genres = []
|
||||
for entry in media.get("genres") or []:
|
||||
text = str(entry or "").strip()
|
||||
if text:
|
||||
genres.append(text)
|
||||
|
||||
tags_raw = media.get("tags") or []
|
||||
tags = []
|
||||
for tag in tags_raw:
|
||||
for tag in media.get("tags") or []:
|
||||
if not isinstance(tag, dict):
|
||||
continue
|
||||
name = (tag.get("name") or "").strip()
|
||||
rank = int(tag.get("rank") or 0)
|
||||
if name:
|
||||
tags.append({"name": name, "rank": rank})
|
||||
if name and rank >= 70:
|
||||
tags.append(name)
|
||||
|
||||
mapped = {
|
||||
"id": media.get("id"),
|
||||
"title_best": pick_best_title(title),
|
||||
return {
|
||||
"anilist_id": int(media.get("id") or 0),
|
||||
"title_english": (title.get("english") or "").strip(),
|
||||
"title_romaji": (title.get("romaji") or "").strip(),
|
||||
"title_native": (title.get("native") or "").strip(),
|
||||
"start_date": safe_date(media.get("startDate")),
|
||||
"format": (media.get("format") or "").strip(),
|
||||
"episodes": media.get("episodes"),
|
||||
"duration": media.get("duration"),
|
||||
"source": str(media.get("source") or "").replace("_", " ").title(),
|
||||
"description": (media.get("description") or "").strip(),
|
||||
"genres": genres,
|
||||
"genres_text": ", ".join(genres) if genres else "",
|
||||
"tags": tags,
|
||||
"tags_text": ", ".join(tag["name"] for tag in tags if tag["rank"] >= 70) or "",
|
||||
"studio_names": studio_names,
|
||||
"studio_text": ", ".join(studio_names) if studio_names else "",
|
||||
"anilist_url": (media.get("siteUrl") or "").strip(),
|
||||
"cover_image": (((media.get("coverImage") or {}).get("large")) or "").strip(),
|
||||
"raw": media,
|
||||
"anilist_url": (media.get("siteUrl") or "").strip(),
|
||||
"start_year": int(start_date.get("year") or 0),
|
||||
"format": (media.get("format") or "").strip() or "MOVIE",
|
||||
"description": (media.get("description") or "").strip(),
|
||||
"genres_text": ", ".join(genres),
|
||||
"tags_text": ", ".join(tags),
|
||||
"studio_text": ", ".join(studio_names),
|
||||
}
|
||||
return mapped
|
||||
|
||||
|
||||
def fetch_anilist_movie_by_search(search_text: str, cache: dict[str, dict | None]) -> dict | None:
|
||||
key = normalize_title(search_text)
|
||||
if key in cache:
|
||||
return cache[key]
|
||||
def fetch_anilist_movie_candidates(today: date, years_window: int = 1) -> list[dict]:
|
||||
start_date = today - timedelta(days=365 * years_window)
|
||||
end_date = today + timedelta(days=365 * years_window)
|
||||
|
||||
query = """
|
||||
query ($search: String, $perPage: Int) {
|
||||
Page(page: 1, perPage: $perPage) {
|
||||
media(type: ANIME, format: MOVIE, search: $search, sort: [SEARCH_MATCH, POPULARITY_DESC]) {
|
||||
query ($page: Int, $perPage: Int, $start: FuzzyDateInt, $end: FuzzyDateInt) {
|
||||
Page(page: $page, perPage: $perPage) {
|
||||
pageInfo { hasNextPage }
|
||||
media(
|
||||
type: ANIME
|
||||
format: MOVIE
|
||||
countryOfOrigin: JP
|
||||
sort: [POPULARITY_DESC, START_DATE_DESC]
|
||||
startDate_greater: $start
|
||||
startDate_lesser: $end
|
||||
) {
|
||||
id
|
||||
title { english romaji native }
|
||||
startDate { year month day }
|
||||
startDate { year }
|
||||
format
|
||||
episodes
|
||||
duration
|
||||
source
|
||||
description(asHtml: false)
|
||||
genres
|
||||
tags { name rank }
|
||||
@@ -129,48 +103,39 @@ def fetch_anilist_movie_by_search(search_text: str, cache: dict[str, dict | None
|
||||
}
|
||||
"""
|
||||
|
||||
try:
|
||||
data = post_graphql(query, {"search": search_text, "perPage": 5})
|
||||
except Exception:
|
||||
cache[key] = None
|
||||
return None
|
||||
page = 1
|
||||
results = []
|
||||
seen_ids = set()
|
||||
|
||||
if "errors" in data:
|
||||
cache[key] = None
|
||||
return None
|
||||
while True:
|
||||
payload = post_graphql(
|
||||
query,
|
||||
{
|
||||
"page": page,
|
||||
"perPage": 50,
|
||||
"start": yyyymmdd(start_date) - 1,
|
||||
"end": yyyymmdd(end_date) + 1,
|
||||
},
|
||||
)
|
||||
|
||||
candidates = data.get("data", {}).get("Page", {}).get("media", [])
|
||||
if not candidates:
|
||||
cache[key] = None
|
||||
return None
|
||||
if "errors" in payload:
|
||||
raise RuntimeError(payload["errors"])
|
||||
|
||||
wanted = normalize_title(search_text)
|
||||
best = None
|
||||
best_score = -1
|
||||
for media in candidates:
|
||||
title = media.get("title") or {}
|
||||
options = [title.get("english"), title.get("romaji"), title.get("native")]
|
||||
score = 0
|
||||
for option in options:
|
||||
normalized = normalize_title(str(option or ""))
|
||||
if not normalized:
|
||||
page_data = payload.get("data", {}).get("Page", {})
|
||||
for media in page_data.get("media", []):
|
||||
mapped = map_anilist_movie(media)
|
||||
anilist_id = mapped.get("anilist_id") or 0
|
||||
if anilist_id <= 0 or anilist_id in seen_ids:
|
||||
continue
|
||||
if normalized == wanted:
|
||||
score = max(score, 3)
|
||||
elif wanted and (wanted in normalized or normalized in wanted):
|
||||
score = max(score, 2)
|
||||
elif normalized.split(" ")[:2] == wanted.split(" ")[:2]:
|
||||
score = max(score, 1)
|
||||
if score > best_score:
|
||||
best_score = score
|
||||
best = media
|
||||
if score == 3:
|
||||
break
|
||||
seen_ids.add(anilist_id)
|
||||
|
||||
if best_score <= 0:
|
||||
cache[key] = None
|
||||
return None
|
||||
if not mapped.get("title_english") and not mapped.get("title_romaji") and not mapped.get("title_native"):
|
||||
continue
|
||||
|
||||
mapped = map_anilist_media(best) if best else None
|
||||
cache[key] = mapped
|
||||
return mapped
|
||||
results.append(mapped)
|
||||
|
||||
if not page_data.get("pageInfo", {}).get("hasNextPage"):
|
||||
break
|
||||
page += 1
|
||||
|
||||
return results
|
||||
|
||||
Reference in New Issue
Block a user