feat(launcher): migrate aniskip resolution to launcher script opts

2026-03-03 06:22:41 -08:00 · 2026-03-03 00:37:50 -08:00
parent 10ef535f9a
commit 092c56f98f
10 changed files with 712 additions and 7 deletions
--- a/launcher/aniskip-metadata.ts
+++ b/launcher/aniskip-metadata.ts
@@ -2,11 +2,22 @@ import path from 'node:path';
 import { spawnSync } from 'node:child_process';
 import { commandExists } from './util.js';

+export type AniSkipLookupStatus =
+  | 'ready'
+  | 'missing_mal_id'
+  | 'missing_episode'
+  | 'missing_payload'
+  | 'lookup_failed';
+
 export interface AniSkipMetadata {
  title: string;
  season: number | null;
  episode: number | null;
  source: 'guessit' | 'fallback';
+  malId: number | null;
+  introStart: number | null;
+  introEnd: number | null;
+  lookupStatus?: AniSkipLookupStatus;
 }

 interface InferAniSkipDeps {
@@ -14,6 +25,50 @@ interface InferAniSkipDeps {
  runGuessit: (mediaPath: string) => string | null;
 }

+interface MalSearchResult {
+  id?: unknown;
+  name?: unknown;
+}
+
+interface MalSearchCategory {
+  items?: unknown;
+}
+
+interface MalSearchResponse {
+  categories?: unknown;
+}
+
+interface AniSkipIntervalPayload {
+  start_time?: unknown;
+  end_time?: unknown;
+}
+
+interface AniSkipSkipItemPayload {
+  skip_type?: unknown;
+  interval?: unknown;
+}
+
+interface AniSkipPayloadResponse {
+  found?: unknown;
+  results?: unknown;
+}
+
+const MAL_PREFIX_API = 'https://myanimelist.net/search/prefix.json?type=anime&keyword=';
+const ANISKIP_PAYLOAD_API = 'https://api.aniskip.com/v1/skip-times/';
+const MAL_USER_AGENT = 'SubMiner-launcher/ani-skip';
+const MAL_MATCH_STOPWORDS = new Set([
+  'the',
+  'this',
+  'that',
+  'world',
+  'animated',
+  'series',
+  'season',
+  'no',
+  'on',
+  'and',
+]);
+
 function toPositiveInt(value: unknown): number | null {
  if (typeof value === 'number' && Number.isFinite(value) && value > 0) {
    return Math.floor(value);
@@ -27,6 +82,217 @@ function toPositiveInt(value: unknown): number | null {
  return null;
 }

+function toPositiveNumber(value: unknown): number | null {
+  if (typeof value === 'number' && Number.isFinite(value) && value > 0) {
+    return value;
+  }
+  if (typeof value === 'string') {
+    const parsed = Number.parseFloat(value);
+    if (Number.isFinite(parsed) && parsed > 0) {
+      return parsed;
+    }
+  }
+  return null;
+}
+
+function normalizeForMatch(value: string): string {
+  return value
+    .toLowerCase()
+    .replace(/[^\w]+/g, ' ')
+    .replace(/\s+/g, ' ')
+    .trim();
+}
+
+function tokenizeMatchWords(value: string): string[] {
+  const words = normalizeForMatch(value)
+    .split(' ')
+    .filter((word) => word.length >= 3);
+  return words.filter((word) => !MAL_MATCH_STOPWORDS.has(word));
+}
+
+function titleOverlapScore(expectedTitle: string, candidateTitle: string): number {
+  const expected = normalizeForMatch(expectedTitle);
+  const candidate = normalizeForMatch(candidateTitle);
+
+  if (!expected || !candidate) return 0;
+
+  if (candidate.includes(expected)) return 120;
+
+  const expectedTokens = tokenizeMatchWords(expectedTitle);
+  if (expectedTokens.length === 0) return 0;
+
+  const candidateSet = new Set(tokenizeMatchWords(candidateTitle));
+  let score = 0;
+  let matched = 0;
+
+  for (const token of expectedTokens) {
+    if (candidateSet.has(token)) {
+      score += 30;
+      matched += 1;
+    } else {
+      score -= 20;
+    }
+  }
+
+  if (matched === 0) {
+    score -= 80;
+  }
+
+  const coverage = matched / expectedTokens.length;
+  if (expectedTokens.length >= 2) {
+    if (coverage >= 0.8) score += 30;
+    else if (coverage >= 0.6) score += 10;
+    else score -= 50;
+  } else if (coverage >= 1) {
+    score += 10;
+  }
+
+  return score;
+}
+
+function hasAnySequelMarker(candidateTitle: string): boolean {
+  const normalized = ` ${normalizeForMatch(candidateTitle)} `;
+  if (!normalized.trim()) return false;
+
+  const markers = [
+    'season 2',
+    'season 3',
+    'season 4',
+    '2nd season',
+    '3rd season',
+    '4th season',
+    'second season',
+    'third season',
+    'fourth season',
+    ' ii ',
+    ' iii ',
+    ' iv ',
+  ];
+  return markers.some((marker) => normalized.includes(marker));
+}
+
+function seasonSignalScore(requestedSeason: number | null, candidateTitle: string): number {
+  const season = toPositiveInt(requestedSeason);
+  if (!season || season < 1) return 0;
+
+  const normalized = ` ${normalizeForMatch(candidateTitle)} `;
+  if (!normalized.trim()) return 0;
+
+  if (season === 1) {
+    return hasAnySequelMarker(candidateTitle) ? -60 : 20;
+  }
+
+  const numericMarker = ` season ${season} `;
+  const ordinalMarker = ` ${season}th season `;
+  if (normalized.includes(numericMarker) || normalized.includes(ordinalMarker)) {
+    return 40;
+  }
+
+  const romanAliases = {
+    2: [' ii ', ' second season ', ' 2nd season '],
+    3: [' iii ', ' third season ', ' 3rd season '],
+    4: [' iv ', ' fourth season ', ' 4th season '],
+    5: [' v ', ' fifth season ', ' 5th season '],
+  } as const;
+
+  const aliases = romanAliases[season] ?? [];
+  return aliases.some((alias) => normalized.includes(alias)) ? 40 : hasAnySequelMarker(candidateTitle) ? -20 : 5;
+}
+
+function toMalSearchItems(payload: unknown): MalSearchResult[] {
+  const parsed = payload as MalSearchResponse;
+  const categories = Array.isArray(parsed?.categories) ? parsed.categories : null;
+  if (!categories) return [];
+
+  const items: MalSearchResult[] = [];
+  for (const category of categories) {
+    const typedCategory = category as MalSearchCategory;
+    const rawItems = Array.isArray(typedCategory?.items) ? typedCategory.items : [];
+    for (const rawItem of rawItems) {
+      const item = rawItem as Record<string, unknown>;
+      items.push({
+        id: item?.id,
+        name: item?.name,
+      });
+    }
+  }
+  return items;
+}
+
+function normalizeEpisodePayload(value: unknown): number | null {
+  return toPositiveNumber(value);
+}
+
+function parseAniSkipPayload(payload: unknown): { start: number; end: number } | null {
+  const parsed = payload as AniSkipPayloadResponse;
+  const results = Array.isArray(parsed?.results) ? parsed.results : null;
+  if (!results) return null;
+
+  for (const rawResult of results) {
+    const result = rawResult as AniSkipSkipItemPayload;
+    if (result.skip_type !== 'op' || typeof result.interval !== 'object' || result.interval === null) {
+      continue;
+    }
+    const interval = result.interval as AniSkipIntervalPayload;
+    const start = normalizeEpisodePayload(interval?.start_time);
+    const end = normalizeEpisodePayload(interval?.end_time);
+    if (start !== null && end !== null && end > start) {
+      return { start, end };
+    }
+  }
+
+  return null;
+}
+
+async function fetchJson<T>(url: string): Promise<T | null> {
+  const response = await fetch(url, {
+    headers: {
+      'User-Agent': MAL_USER_AGENT,
+    },
+  });
+  if (!response.ok) return null;
+  try {
+    return (await response.json()) as T;
+  } catch {
+    return null;
+  }
+}
+
+async function resolveMalIdFromTitle(title: string, season: number | null): Promise<number | null> {
+  const lookup = season && season > 1 ? `${title} Season ${season}` : title;
+  const payload = await fetchJson<unknown>(`${MAL_PREFIX_API}${encodeURIComponent(lookup)}`);
+  const items = toMalSearchItems(payload);
+  if (!items.length) return null;
+
+  let bestScore = Number.NEGATIVE_INFINITY;
+  let bestMalId: number | null = null;
+
+  for (const item of items) {
+    const id = toPositiveInt(item.id);
+    if (!id) continue;
+    const name = typeof item.name === 'string' ? item.name : '';
+    if (!name) continue;
+
+    const score = titleOverlapScore(title, name) + seasonSignalScore(season, name);
+    if (score > bestScore) {
+      bestScore = score;
+      bestMalId = id;
+    }
+  }
+
+  return bestMalId;
+}
+
+async function fetchAniSkipPayload(
+  malId: number,
+  episode: number,
+): Promise<{ start: number; end: number } | null> {
+  const payload = await fetchJson<unknown>(`${ANISKIP_PAYLOAD_API}${malId}/${episode}?types=op&types=ed`);
+  const parsed = payload as AniSkipPayloadResponse;
+  if (!parsed || parsed.found !== true) return null;
+  return parseAniSkipPayload(parsed);
+}
+
 function detectEpisodeFromName(baseName: string): number | null {
  const patterns = [
    /[Ss]\d+[Ee](\d{1,3})/,
@@ -133,6 +399,10 @@ export function parseAniSkipGuessitJson(stdout: string, mediaPath: string): AniS
      season,
      episode: episodeFromDirect ?? episodeFromList,
      source: 'guessit',
+      malId: null,
+      introStart: null,
+      introEnd: null,
+      lookupStatus: 'lookup_failed',
    };
  } catch {
    return null;
@@ -171,9 +441,70 @@ export function inferAniSkipMetadataForFile(
    season: detectSeasonFromNameOrDir(mediaPath),
    episode: detectEpisodeFromName(baseName),
    source: 'fallback',
+    malId: null,
+    introStart: null,
+    introEnd: null,
+    lookupStatus: 'lookup_failed',
  };
 }

+export async function resolveAniSkipMetadataForFile(mediaPath: string): Promise<AniSkipMetadata> {
+  const inferred = inferAniSkipMetadataForFile(mediaPath);
+  if (!inferred.title) {
+    return { ...inferred, lookupStatus: 'lookup_failed' };
+  }
+
+  try {
+    const malId = await resolveMalIdFromTitle(inferred.title, inferred.season);
+    if (!malId) {
+      return {
+        ...inferred,
+        malId: null,
+        introStart: null,
+        introEnd: null,
+        lookupStatus: 'missing_mal_id',
+      };
+    }
+
+    if (!inferred.episode) {
+      return {
+        ...inferred,
+        malId,
+        introStart: null,
+        introEnd: null,
+        lookupStatus: 'missing_episode',
+      };
+    }
+
+    const payload = await fetchAniSkipPayload(malId, inferred.episode);
+    if (!payload) {
+      return {
+        ...inferred,
+        malId,
+        introStart: null,
+        introEnd: null,
+        lookupStatus: 'missing_payload',
+      };
+    }
+
+    return {
+      ...inferred,
+      malId,
+      introStart: payload.start,
+      introEnd: payload.end,
+      lookupStatus: 'ready',
+    };
+  } catch {
+    return {
+      ...inferred,
+      malId: inferred.malId,
+      introStart: inferred.introStart,
+      introEnd: inferred.introEnd,
+      lookupStatus: 'lookup_failed',
+    };
+  }
+}
+
 function sanitizeScriptOptValue(value: string): string {
  return value
    .replace(/,/g, ' ')
@@ -182,6 +513,28 @@ function sanitizeScriptOptValue(value: string): string {
    .trim();
 }

+function buildLauncherAniSkipPayload(aniSkipMetadata: AniSkipMetadata): string | null {
+  if (!aniSkipMetadata.malId || !aniSkipMetadata.introStart || !aniSkipMetadata.introEnd) {
+    return null;
+  }
+  if (aniSkipMetadata.introEnd <= aniSkipMetadata.introStart) {
+    return null;
+  }
+  const payload = {
+    found: true,
+    results: [
+      {
+        skip_type: 'op',
+        interval: {
+          start_time: aniSkipMetadata.introStart,
+          end_time: aniSkipMetadata.introEnd,
+        },
+      },
+    ],
+  };
+  return encodeURIComponent(JSON.stringify(payload));
+}
+
 export function buildSubminerScriptOpts(
  appPath: string,
  socketPath: string,
@@ -200,5 +553,21 @@ export function buildSubminerScriptOpts(
  if (aniSkipMetadata && aniSkipMetadata.episode && aniSkipMetadata.episode > 0) {
    parts.push(`subminer-aniskip_episode=${aniSkipMetadata.episode}`);
  }
+  if (aniSkipMetadata && aniSkipMetadata.malId && aniSkipMetadata.malId > 0) {
+    parts.push(`subminer-aniskip_mal_id=${aniSkipMetadata.malId}`);
+  }
+  if (aniSkipMetadata && aniSkipMetadata.introStart !== null && aniSkipMetadata.introStart > 0) {
+    parts.push(`subminer-aniskip_intro_start=${aniSkipMetadata.introStart}`);
+  }
+  if (aniSkipMetadata && aniSkipMetadata.introEnd !== null && aniSkipMetadata.introEnd > 0) {
+    parts.push(`subminer-aniskip_intro_end=${aniSkipMetadata.introEnd}`);
+  }
+  if (aniSkipMetadata?.lookupStatus) {
+    parts.push(`subminer-aniskip_lookup_status=${sanitizeScriptOptValue(aniSkipMetadata.lookupStatus)}`);
+  }
+  const aniskipPayload = aniSkipMetadata ? buildLauncherAniSkipPayload(aniSkipMetadata) : null;
+  if (aniskipPayload) {
+    parts.push(`subminer-aniskip_payload=${sanitizeScriptOptValue(aniskipPayload)}`);
+  }
  return parts.join(',');
 }