import path from "node:path";
import type { Cue } from "./vtt";
import type { TrackFields } from "./captionTracks";
import { readFile } from "fs-extra";
import { getPaths } from "./paths";

export type { Platform } from "./platform";
import type { Platform } from "./platform";
import type { VideoState } from "./availability";

export type TranscriptSummary = {
  slug: string;
  id: string;
  channelSlug: string;
  title: string;
  uploadDate: string;
  duration: number;
  channel: string;
  description: string;
  // Keyword tags from metadata.info.json (YouTube/Odysee), searchable via the
  // "tags" scope. Empty when the platform/archive carries none.
  tags: string[];
  isLivestream: boolean;
  ageRestricted: boolean;
  platform: Platform;
  webpageUrl: string;
  // HLS master-playlist URL for platforms with no iframe embed (Kick VODs).
  // Undefined for every other platform. Played via react-player/file + hls.js.
  hlsUrl?: string;
  // A plain media file URL a native <video> plays and seeks (archive.org: the
  // record's browser-playable file, lib/archiveOrg.ts). Undefined everywhere
  // else. New with the archiveorg platform, so no cached or normalized record
  // predates it and no cache version moved.
  mediaUrl?: string;
  // Curated tag ids (common/lib/curatedTags.ts) — the operator's cross-channel
  // vocabulary, NOT the yt-dlp keywords in `tags` above. OMITTED when empty, so
  // an untagged corpus's pages stay byte-identical to the ones already on disk
  // and the build's sha1 skip still short-circuits.
  curatedTags?: string[];
  // The day a VOD mirror's video was RECORDED (`YYYYMMDD`, the shape of
  // `uploadDate`), read off its title by its channel's `recordedDate` rule
  // (lib/recordedDate.ts) when the index builds the record. OMITTED when the
  // channel has no rule or the title gives no date, so every other channel's
  // pages stay byte-identical. Coverage reads it first (`coverageDate`).
  recordedDate?: string;
};

export type DisplaySummary = {
  slug: string;
  id: string;
  channelSlug: string;
  title: string;
  uploadDate: string;
  date: string;
  duration: string;
  channel: string;
  isLivestream: boolean;
  ageRestricted: boolean;
  // Legacy booleans, derived from `state`. Still emitted because the MCP's
  // RemoteSource reads summaries pages from live origins it does not control,
  // so version skew between a hub and its members is real: an older reader must
  // keep seeing deleted/unlisted, and an older *site* must keep answering a
  // newer reader.
  isDeleted: boolean;
  isUnlisted: boolean;
  // Presence on the source platform. Omitted when "available" (absent =>
  // available), so a page with no missing videos stays byte-identical to the
  // one already on disk and the build's downstream hashing still short-circuits.
  state?: VideoState;
  platform: Platform;
  webpageUrl: string;
  // Curated tag ids (common/lib/curatedTags.ts) — the operator's cross-channel
  // vocabulary. A DisplaySummary carries no yt-dlp keywords, but the sibling
  // TranscriptSummary.tags does and these are NOT those. OMITTED when empty,
  // for the same byte-identical-pages reason as `state` above.
  curatedTags?: string[];
};

// `track` / `altTracks` (lib/captionTracks.ts): the primary's track id and the
// other English tracks whose words differ from it — present only on a record
// that has such a track.
export type TranscriptDetail = TranscriptSummary & {
  cues: Cue[] | undefined;
} & TrackFields;

export type TranscriptPage = TranscriptDetail[];

export async function countTranscripts(): Promise<number> {
  const manifestPath = path.join(getPaths().exportSummariesDir, "manifest.json");
  const raw = await readFile(manifestPath, "utf8");
  const parsed = JSON.parse(raw) as { totalCount?: number };
  if (typeof parsed.totalCount !== "number") {
    throw new Error("manifest.json missing totalCount; run `pnpm build:index`");
  }
  return parsed.totalCount;
}

export { formatDate, formatDuration } from "./format";