meetus/ui/doocus-app/server/doocusApi.ts

/**
 * Vite dev middleware exposing the local doocus tree index to the app.
 *
 * Reads `docs-output/index.json` (produced by process_tree.py) — a replica of
 * the whole local drive tree — plus the extracted `<file>.doocus/content.md`
 * sidecars, and streams the ORIGINAL files by relative path (resolved against
 * the index's recorded root). All reads are local; nothing leaves the machine.
 *
 * Routes:
 *   GET /api/tree                 the index.json tree
 *   GET /api/detail?path=<rel>    node + extracted content + meta + inline text
 *   GET /api/original?path=<rel>  stream the original file (native viewer / packaging)
 */
import type { Plugin } from 'vite'
import type { ServerResponse } from 'node:http'
import fs from 'node:fs'
import fsp from 'node:fs/promises'
import path from 'node:path'

interface Options {
  outputDir: string
}

interface Node {
  path: string
  name: string
  ext: string
  family: string
  mode: 'extracted' | 'meeting' | 'link'
  bytes: number
  modified: string
  url: string | null
  out?: string
}

interface Index {
  root: string
  generatedAt: string
  counts: Record<string, number>
  files: Node[]
}

// Text-native files: readable as-is, so the "original" doubles as its own text.
const TEXT_EXTS = new Set(['md', 'markdown', 'txt', 'json', 'yaml', 'yml', 'csv', 'html', 'htm'])

const MIME: Record<string, string> = {
  '.pdf': 'application/pdf', '.png': 'image/png', '.jpg': 'image/jpeg', '.jpeg': 'image/jpeg',
  '.html': 'text/html', '.htm': 'text/html', '.md': 'text/markdown', '.txt': 'text/plain',
  '.csv': 'text/csv', '.json': 'application/json', '.yaml': 'text/yaml', '.yml': 'text/yaml',
  '.docx': 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
  '.pptx': 'application/vnd.openxmlformats-officedocument.presentationml.presentation',
  '.xlsx': 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
  '.mp4': 'video/mp4',
}

export function doocusApi(opts: Options): Plugin {
  const outputDir = path.resolve(opts.outputDir)
  const indexPath = path.join(outputDir, 'index.json')

  function readIndex(): Index | null {
    try {
      return JSON.parse(fs.readFileSync(indexPath, 'utf-8'))
    } catch {
      return null
    }
  }

  /** Resolve a node's original file, rejecting traversal outside the index root. */
  function originalAbs(index: Index, rel: string): string | null {
    const root = path.resolve(index.root)
    const abs = path.resolve(root, rel)
    if (abs !== root && !abs.startsWith(root + path.sep)) return null
    return abs
  }

  return {
    name: 'doocus-api',
    configureServer(server) {
      server.config.logger.info(`[doocus-api] serving tree from: ${indexPath}`)
      server.middlewares.use('/api', (req, res, next) => {
        const url = new URL(req.url ?? '/', 'http://localhost')
        const parts = url.pathname.split('/').filter(Boolean)
        handle(res, parts, url.searchParams).catch((err) => {
          server.config.logger.error(`[doocus-api] ${String(err)}`)
          sendJson(res, 500, { error: String(err) })
        }).then((handled) => { if (!handled) next() })
      })
    },
  }

  async function handle(res: ServerResponse, parts: string[], q: URLSearchParams): Promise<boolean> {
    if (parts[0] === 'tree') {
      const index = readIndex()
      if (!index) { sendJson(res, 404, { error: `index.json not found in ${outputDir}` }); return true }
      sendJson(res, 200, index)
      return true
    }

    if (parts[0] === 'detail') {
      await getDetail(res, q.get('path') ?? '')
      return true
    }

    if (parts[0] === 'original') {
      await serveOriginal(res, q.get('path') ?? '')
      return true
    }

    return false
  }

  async function getDetail(res: ServerResponse, rel: string): Promise<void> {
    const index = readIndex()
    if (!index) { sendJson(res, 404, { error: 'no index' }); return }
    const node = index.files.find((f) => f.path === rel)
    if (!node) { sendJson(res, 404, { error: 'node not found' }); return }

    let content: string | null = null      // extracted search text (heavy formats)
    let meta: unknown = null
    let text: string | null = null         // inline text for text-native originals

    if (node.mode === 'extracted' && node.out) {
      try { content = await fsp.readFile(path.join(outputDir, node.out, 'content.md'), 'utf-8') } catch { /* */ }
      try { meta = JSON.parse(await fsp.readFile(path.join(outputDir, node.out, 'meta.json'), 'utf-8')) } catch { /* */ }
    } else if (node.mode === 'link' && TEXT_EXTS.has(node.ext)) {
      const abs = originalAbs(index, rel)
      if (abs) { try { text = await fsp.readFile(abs, 'utf-8') } catch { /* */ } }
    }

    sendJson(res, 200, {
      node,
      content,
      meta,
      text,
      mime: MIME[path.extname(node.name).toLowerCase()] ?? 'application/octet-stream',
      originalUrl: `/api/original?path=${encodeURIComponent(rel)}`,
    })
  }

  async function serveOriginal(res: ServerResponse, rel: string): Promise<void> {
    const index = readIndex()
    if (!index) { sendJson(res, 404, { error: 'no index' }); return }
    const abs = originalAbs(index, rel)
    if (!abs || !isFile(abs)) { sendJson(res, 404, { error: 'original not reachable' }); return }
    res.setHeader('Content-Type', MIME[path.extname(abs).toLowerCase()] ?? 'application/octet-stream')
    res.setHeader('Cache-Control', 'no-cache')
    fs.createReadStream(abs).pipe(res)
  }
}

function sendJson(res: ServerResponse, status: number, obj: unknown): void {
  const body = JSON.stringify(obj)
  res.statusCode = status
  res.setHeader('Content-Type', 'application/json')
  res.setHeader('Content-Length', Buffer.byteLength(body))
  res.end(body)
}

function isFile(p: string): boolean {
  try { return fs.statSync(p).isFile() } catch { return false }
}