import fs from "node:fs";
import path from "node:path";
import os from "node:os";
// curatedTags.ts is pure (no imports of its own), so this cannot cycle.
import { TAGS_FILENAME } from "./curatedTags";

export type Paths = {
  monorepoRoot: string;
  transcriptsDir: string;
  channelsDir: string;
  // Root of the saved-video store: persisted source-video containers (the
  // keep-latest persistence rule) are MOVED out of the per-video data dir into
  // savedVideosDir/<slug>/<videoId>/, leaving only a small saved-video.json
  // pointer in the data dir. Defaults under transcriptsDir but is overridable
  // via SAVED_VIDEOS_DIR so the (large) source videos can live on a separate
  // disk. A channel may further override the root via ChannelConfig.savedVideosDir.
  // See common/lib/savedVideo.ts.
  savedVideosDir: string;
  // Per-site config lives under sitesDir/<siteId>/site.json (+ chart-templates.json).
  // See common/lib/site.ts. A "site" is a selection + presentation layer over the
  // single global channel pool; channel downloads are never duplicated per site.
  sitesDir: string;
  // The Archilyzer hub/homepage (the `homepage` SSG package): a single, instance-
  // level cross-site landing site, NOT a content site. Its config + chart
  // dashboard live under sitesDir/_homepage. The leading underscore is
  // intentionally not a valid siteId (SITE_ID_RE), so listSiteIds() ignores it.
  // See common/lib/homepage.ts.
  homepageDir: string;
  homepageConfigFile: string;
  homepageChartTemplatesFile: string;
  jobsDir: string;
  // Scratch area for remote-worker transcription requests: uploaded audio and
  // the produced transcript.json live under workerScratchDir/<remoteJobId>/ and
  // are cleaned up once the requesting instance pulls the result (or cancels).
  workerScratchDir: string;
  // Persistent state for the cron-driven sync scheduler (per-channel backoff +
  // a rolling run log). Survives restarts, unlike the in-memory job registry.
  // See common/jobs/syncSchedulerState.ts.
  schedulerStateFile: string;
  // Persistent fairness state for the automatic priority-queue runners
  // (round-robin / weighted-fair cursors + a rolling pick log). Best-effort:
  // drift across a restart is cosmetic. See common/jobs/autoQueueState.ts.
  autoQueueStateFile: string;
  // Persisted "default" worker arrangement: the set of worker ids that should
  // start enabled on the next server launch (workers not listed start
  // disabled). Written by the Workers page "Set as default" button; applied by
  // the worker pool on first use. See common/jobs/workerDefaults.ts.
  workerDefaultsFile: string;
  // Saved monitor-widget presets: named /widget query strings the builder and
  // the in-widget menu both load from. A preset stores the LINK, not a parsed
  // config, so a preset and a shared link are literally the same artifact. See
  // common/lib/widgetPresets.ts.
  widgetPresetsFile: string;
  // The two CHANGELOG.md files the "cut release" flow reads and rewrites. They
  // are TRACKED source files, and cutting a release optionally makes a real git
  // commit — so e2e must be able to point them somewhere disposable. Overridable
  // via EDITOR_CHANGELOG_FILE / EXPORT_CHANGELOG_FILE; the editor's dev:test and
  // start:test scripts set them to editor/test-changelog.md and
  // editor/test-export-changelog.md (both gitignored) — the second since
  // POST /api/ops/cut-release made the export changelog reachable from a spec.
  // Without this the changelog spec rewrites the repo's own changelog and can
  // leave a stray "Release editor <version>" commit behind.
  editorChangelogFile: string;
  exportChangelogFile: string;
  lmdbPath: string;
  exportDir: string;
  // The dir Next.js serves at "/" (also holds checked-in static assets). For
  // multi-site builds this is composed one site at a time from the staging dirs
  // below: per-site summaries/subs-manifest/stats/chart-templates plus the
  // site's subset of the shared transcripts/subs trees.
  exportPublicDir: string;
  exportSummariesDir: string;
  exportTranscriptsDir: string;
  exportSubsDir: string;
  // Served per-channel social-post page tree (/posts/<slug>/{manifest,page-NNNN}.json).
  // The parallel corpus to transcripts/subs — see common/lib/posts.ts.
  exportPostsDir: string;
  // Served per-channel AI-digest page tree
  // (/digests/<slug>/{manifest,page-NNNN}.json). The DERIVED corpus: chapters
  // and topic tags generated from a transcript, composed with human overrides
  // at index time. Sparse by design — only videos that have been digested
  // appear in a manifest's slugToPage, which is what lets the viewer hide its
  // Digest control without a per-video fetch. See common/lib/digests.ts.
  exportDigestsDir: string;
  exportStatsDir: string;
  // Staging area (NOT served) where the shared index + per-site aggregates are
  // built before composition. Shared per-channel transcript/subs pages are
  // identical across sites and built once under exportSharedDir; per-site
  // summaries/subs-manifest/stats land under exportSitesIndexDir/<siteId>/.
  exportIndexDir: string;
  exportSharedDir: string;
  exportSharedTranscriptsDir: string;
  exportSharedSubsDir: string;
  exportSharedPostsDir: string;
  exportSharedDigestsDir: string;
  exportSitesIndexDir: string;
  // Per-site extracted static output (`out/`) from an isolated (Docker) build,
  // keyed exportBuildsDir/<siteId>. Sibling of .export-index. The wrangler deploy
  // reads a site's bundle from here in docker mode (in basic mode the deploy
  // reads export/out directly). Env-overridable so test data stays isolated.
  exportBuildsDir: string;
  settingsFile: string;
  chartsConfigFile: string;
  // Global (cross-site) search-alias dictionary. Per-site aliases live at
  // sitesDir/<siteId>/search-aliases.json (see common/lib/site.ts); the two are
  // merged into each site's shipped /search-aliases.json at compose time. In the
  // data dir (not monorepoRoot — the legacy chartsConfigFile location there is
  // migration-only). See common/lib/aliasesStore.ts.
  globalAliasesFile: string;
  // Curated per-video tags: the corpus-wide vocabulary AND every assignment
  // (<transcriptsDir>/tags.json). Per-site presentation overlays live at
  // sitesDir/<siteId>/tags.json (see common/lib/site.ts). Authoritative, and
  // real curated data — never hand-edited. See common/lib/curatedTagsStore.ts.
  globalTagsFile: string;
  ytdlpBin: string;
  whisperBin: string;
  whisperModel: string;
  ffmpegBin: string;
  // ffprobe binary, used to measure a downloaded audio file's actual duration
  // for the download-time short-audio guard (common/ytdlp/ffprobeDuration.ts).
  // Ships alongside ffmpeg.
  ffprobeBin: string;
  // rsync binary used to mirror the saved-video store to a backup destination
  // (Phase 4 of the video-persistence feature). See
  // common/controller/backupSavedVideos.ts.
  rsyncBin: string;
  // aria2c, which fetches an archive.org file over BitTorrent
  // (lib/archiveOrgTorrent-server.ts). Optional: without it the file is
  // downloaded directly.
  aria2cBin: string;
  // findmnt (util-linux): the read-only identity probe behind storage
  // locations — which volume a root is on, and where a UUID is mounted now.
  // See common/lib/storageVolumes.ts. Never required: every call fails open to
  // "identity unknown", which is also what a container answers.
  findmntBin: string;
  // udisksctl: the ONE storage subprocess that changes the machine — mounting
  // an attached-but-unmounted volume by UUID, offered on /storage. Optional in
  // the same way findmnt is; without it the button is simply not offered.
  udisksctlBin: string;
  // gallery-dl, the primary X/Twitter post fetcher (see
  // common/social/xGalleryDlFetcher.ts). A light headless subprocess — the same
  // shape the codebase already manages for yt-dlp and whisper. NOT bundled;
  // install it separately and point GALLERY_DL_BIN at it if it isn't on PATH.
  galleryDlBin: string;
  // parakeet (overlapping-segment stitching) app. parakeetBin is the standalone
  // wrapper script invoked as the app binary; parakeetCliBin is the underlying
  // parakeet-cli it drives; parakeetModel is the default .gguf model.
  parakeetBin: string;
  parakeetCliBin: string;
  parakeetModel: string;
  // Speaker-diarization wrapper script (scripts/diarize.mjs), invoked the way
  // parakeetBin is. Overridable via DIARIZE_BIN — which is also how the e2e
  // suite swaps in a fake diarizer.
  diarizeBin: string;
  // Base URL of the local ollama server, the local-GPU digest lane
  // (common/lib/digestApps.ts posts to `${ollamaUrl}/api/chat`). The FIRST
  // URL-valued entry in Paths, so it is normalized here the way
  // remoteTranscribe.ts normalizes a worker base: trailing slashes stripped, so
  // callers can always concatenate a leading-slash path.
  ollamaUrl: string;
  // The `claude` CLI, driving the opt-in metered digest lane (off by default —
  // settings.digest.remoteEnabled). Not bundled; install it separately and point
  // CLAUDE_BIN at it if it isn't on PATH.
  claudeBin: string;
  // The operator's PRIVATE config dir, outside the repo (~/.config/archilyzer
  // by default). Holds the two inputs of `archilyzer source publish`
  // (common/publish/source.ts), which are never committed:
  //   sourceScrubFile    — git-filter-repo `lhs==>rhs` rules for the mirror
  //   sourceDenylistFile — literals the published source must never contain
  configDir: string;
  sourceScrubFile: string;
  sourceDenylistFile: string;
  // Where `source publish` makes its scratch clone (removed afterwards).
  sourceScratchDir: string;
  // The history pages' render cache (publish/sourceHistory.ts): the XDG
  // cache dir's archilyzer/source-history — ~/.cache unless XDG_CACHE_HOME
  // says otherwise (an empty one is unset, as the XDG spec has it). Never
  // inside the checkout or the public dir: the step renders without it there.
  sourceHistoryCacheDir: string;
  // stagit, which renders the source's history pages (/source/git/). An
  // operator-installed tool, never vendored: `stagit` on PATH, then
  // ~/.local/bin/stagit (publish/sourceHistory.ts resolveStagit). Without it
  // the publish goes on without the history pages.
  stagitBin: string;
};

let cached: Paths | null = null;

// Every path in getPaths() is built on the repo root, which findMonorepoRoot()
// finds by walking up from `process.cwd()`. Turbopack evaluates `process.cwd()`
// statically, and a path op on a value derived from it becomes an asset
// reference — to every file under it when the path is a directory
// (plans/FACTS.md, "A path joined from `process.cwd()` …"). So every join on
// such a value goes through this one opted-out call. Nothing changes at run
// time. The comment sits before a named first argument, not a spread: that is
// the form Turbopack's own advice shows.
function under(first: string, ...rest: string[]): string {
  return path.join(/* turbopackIgnore: true */ first, ...rest);
}

export function getPaths(): Paths {
  if (cached) return cached;
  const monorepoRoot = findMonorepoRoot();
  const transcriptsDir =
    process.env.TRANSCRIPTS_DIR ?? under(monorepoRoot, "transcripts");
  const exportDir = under(monorepoRoot, "export");
  const exportPublicDir =
    process.env.EXPORT_PUBLIC_DIR ?? under(exportDir, "public");
  // Staging sibling of the served public dir (so it lands inside the test data
  // root when EXPORT_PUBLIC_DIR is overridden). Not served; composed into
  // exportPublicDir per site by the build:site step.
  const exportIndexDir =
    process.env.EXPORT_INDEX_DIR ??
    under(path.dirname(/* turbopackIgnore: true */ exportPublicDir), ".export-index");
  const exportSharedDir = under(exportIndexDir, "shared");
  const sitesDir = process.env.SITES_DIR ?? under(transcriptsDir, "sites");
  const homepageDir = under(sitesDir, "_homepage");
  const configDir =
    process.env.ARCHILYZER_CONFIG_DIR ??
    path.join(os.homedir(), ".config", "archilyzer");
  cached = {
    monorepoRoot,
    transcriptsDir,
    channelsDir: under(transcriptsDir, "channels"),
    savedVideosDir:
      process.env.SAVED_VIDEOS_DIR ?? under(transcriptsDir, "saved-videos"),
    sitesDir,
    homepageDir,
    homepageConfigFile: under(homepageDir, "homepage.json"),
    homepageChartTemplatesFile: under(homepageDir, "chart-templates.json"),
    jobsDir: under(transcriptsDir, ".jobs"),
    workerScratchDir: under(transcriptsDir, ".worker-scratch"),
    schedulerStateFile: under(transcriptsDir, ".scheduler", "state.json"),
    autoQueueStateFile: under(transcriptsDir, ".auto-queue", "state.json"),
    workerDefaultsFile: under(transcriptsDir, ".workers", "defaults.json"),
    widgetPresetsFile: under(transcriptsDir, ".widget", "presets.json"),
    lmdbPath: under(transcriptsDir, "index.mdb"),
    exportDir,
    exportPublicDir,
    exportSummariesDir: under(exportPublicDir, "summaries"),
    exportTranscriptsDir: under(exportPublicDir, "transcripts"),
    exportSubsDir: under(exportPublicDir, "subs"),
    exportPostsDir: under(exportPublicDir, "posts"),
    exportDigestsDir: under(exportPublicDir, "digests"),
    exportStatsDir: under(exportPublicDir, "stats"),
    exportIndexDir,
    exportSharedDir,
    exportSharedTranscriptsDir: under(exportSharedDir, "transcripts"),
    exportSharedSubsDir: under(exportSharedDir, "subs"),
    exportSharedPostsDir: under(exportSharedDir, "posts"),
    exportSharedDigestsDir: under(exportSharedDir, "digests"),
    exportSitesIndexDir: under(exportIndexDir, "sites"),
    exportBuildsDir:
      process.env.EXPORT_BUILDS_DIR ??
      under(path.dirname(/* turbopackIgnore: true */ exportPublicDir), ".export-builds"),
    settingsFile:
      process.env.SETTINGS_FILE ?? under(monorepoRoot, "settings.json"),
    editorChangelogFile:
      process.env.EDITOR_CHANGELOG_FILE ??
      under(monorepoRoot, "editor", "CHANGELOG.md"),
    exportChangelogFile:
      process.env.EXPORT_CHANGELOG_FILE ??
      under(exportDir, "CHANGELOG.md"),
    chartsConfigFile:
      process.env.CHARTS_CONFIG_FILE ??
      under(monorepoRoot, "chart-templates.json"),
    globalAliasesFile:
      process.env.SEARCH_ALIASES_FILE ??
      under(transcriptsDir, "search-aliases.json"),
    globalTagsFile:
      process.env.CURATED_TAGS_FILE ??
      under(transcriptsDir, TAGS_FILENAME),
    ytdlpBin: process.env.YTDLP_BIN ?? "yt-dlp",
    whisperBin: process.env.WHISPER_BIN ?? "whisper-cli",
    whisperModel:
      process.env.WHISPER_MODEL ??
      path.join(
        os.homedir(),
        "whispercpp",
        "whisper.cpp",
        "models",
        "ggml-base.en.bin",
      ),
    ffmpegBin: process.env.FFMPEG_BIN ?? "ffmpeg",
    ffprobeBin: process.env.FFPROBE_BIN ?? "ffprobe",
    rsyncBin: process.env.RSYNC_BIN ?? "rsync",
    aria2cBin: process.env.ARIA2C_BIN ?? "aria2c",
    findmntBin: process.env.FINDMNT_BIN ?? "findmnt",
    udisksctlBin: process.env.UDISKSCTL_BIN ?? "udisksctl",
    galleryDlBin: process.env.GALLERY_DL_BIN ?? "gallery-dl",
    parakeetBin:
      process.env.PARAKEET_STITCH_BIN ??
      under(monorepoRoot, "scripts", "parakeet-stitch.mjs"),
    parakeetCliBin: process.env.PARAKEET_CLI ?? "parakeet-cli",
    parakeetModel: process.env.PARAKEET_MODEL ?? "",
    // Speaker-diarization wrapper, same shape as parakeetBin: a script we own,
    // overridable by env so the engine behind it can be swapped (sherpa-onnx
    // today, pyannote later) without touching any caller.
    diarizeBin:
      process.env.DIARIZE_BIN ??
      under(monorepoRoot, "scripts", "diarize.mjs"),
    ollamaUrl: (process.env.OLLAMA_URL ?? "http://127.0.0.1:11434").replace(
      /\/+$/,
      "",
    ),
    claudeBin: process.env.CLAUDE_BIN ?? "claude",
    configDir,
    sourceScrubFile:
      process.env.SOURCE_SCRUB_FILE ?? path.join(configDir, "source-scrub.txt"),
    sourceDenylistFile:
      process.env.SOURCE_DENYLIST_FILE ??
      path.join(configDir, "source-denylist.txt"),
    sourceScratchDir: process.env.ARCHILYZER_SOURCE_SCRATCH ?? os.tmpdir(),
    sourceHistoryCacheDir: path.join(
      process.env.XDG_CACHE_HOME || path.join(os.homedir(), ".cache"),
      "archilyzer",
      "source-history",
    ),
    stagitBin: process.env.STAGIT_BIN ?? "stagit",
  };
  return cached;
}

// Walks up from the working directory to the workspace root; falls back to
// the working directory itself (an app's own directory). Every path op on
// these cwd-derived values opts out of Turbopack's tracing, so what the
// fallback evaluates to at build time never becomes an asset reference.
function findMonorepoRoot(): string {
  const start = path.resolve(/* turbopackIgnore: true */ process.cwd());
  let dir = start;
  for (let i = 0; i < 8; i++) {
    if (fs.existsSync(path.join(/* turbopackIgnore: true */ dir, "pnpm-workspace.yaml"))) return dir;
    const parent = path.dirname(/* turbopackIgnore: true */ dir);
    if (parent === dir) break;
    dir = parent;
  }
  return start;
}