Integration-kit/scripts/checkLinks.js

#!/usr/bin/env node
// Every relative link in this repo's markdown must resolve to a file that exists,
// and every in-page anchor must match a heading in the file it points at.
//
// WHAT THIS DOES NOT DO: fetch anything. External URLs are not checked, on
// purpose. This kit points at a self-hosted Gitea, so an HTTP check would fail on
// a runner without credentials, flake when the host is busy, and teach us to
// ignore red. What breaks in practice is a relative path after a file moves, and
// that is answerable offline with certainty.
//
// It DOES check that every http(s) link into the RunicGateway host names a
// branch, because `.../src/branch/main/...` survives and `.../src/commit/<sha>/...`
// pins a reader to a snapshot of a document we want them reading the current
// version of.
//
// Usage: node scripts/checkLinks.js            (repo root)
//        node scripts/checkLinks.js --quiet

const fs = require('fs')
const path = require('path')

const { stripFences } = require('./lib/markdown')

const ROOT = path.resolve(__dirname, '..')
const QUIET = process.argv.includes('--quiet')

// Directories that hold no prose we own.
const SKIP_DIRS = new Set(['.git', 'node_modules', 'dist'])

/** Every markdown file in the repo, repo-relative, sorted. */
function markdownFiles(dir = ROOT, out = []) {
  for (const entry of fs.readdirSync(dir, { withFileTypes: true })) {
    if (entry.isDirectory()) {
      if (SKIP_DIRS.has(entry.name)) continue
      markdownFiles(path.join(dir, entry.name), out)
    } else if (entry.name.toLowerCase().endsWith('.md')) {
      out.push(path.relative(ROOT, path.join(dir, entry.name)).split(path.sep).join('/'))
    }
  }
  return out.sort()
}

// Fenced code blocks are stripped before links are read (`lib/markdown.js`): a
// fence can legitimately contain a path that does not exist — a directory listing
// of a project the reader has not created yet — and flagging those would make the
// check useless in exactly the document type this repo is made of.

/** Inline `[text](target)` links and `[ref]: target` definitions, with line numbers. */
function linksIn(text) {
  const found = []
  const lines = stripFences(text).split(/\r?\n/)
  lines.forEach((line, i) => {
    // Skip inline code spans: `[a](b)` inside backticks is an example, not a link.
    const bare = line.replace(/`[^`]*`/g, '')
    for (const m of bare.matchAll(/\[[^\]]*\]\(([^)\s]+)(?:\s+"[^"]*")?\)/g)) {
      found.push({ target: m[1], line: i + 1 })
    }
    const def = /^\s{0,3}\[[^\]]+\]:\s*(\S+)/.exec(bare)
    if (def) found.push({ target: def[1], line: i + 1 })
  })
  return found
}

/** GitHub/Gitea-style heading slugs, for anchor checking. */
function anchorsIn(text) {
  const slugs = new Set()
  for (const line of stripFences(text).split(/\r?\n/)) {
    const m = /^\s{0,3}#{1,6}\s+(.*?)\s*#*\s*$/.exec(line)
    if (!m) continue
    const slug = m[1]
      .replace(/`/g, '')
      .replace(/\[([^\]]*)\]\([^)]*\)/g, '$1')
      .toLowerCase()
      .replace(/[^\w\s-]/gu, '')
      .trim()
      .replace(/\s+/g, '-')
    if (slug) slugs.add(slug)
  }
  return slugs
}

const files = markdownFiles()
const anchorCache = new Map()
function anchorsOf(relPath) {
  if (!anchorCache.has(relPath)) {
    anchorCache.set(relPath, anchorsIn(fs.readFileSync(path.join(ROOT, relPath), 'utf8')))
  }
  return anchorCache.get(relPath)
}

const problems = []
let checked = 0

for (const file of files) {
  const text = fs.readFileSync(path.join(ROOT, file), 'utf8')
  const selfAnchors = anchorsIn(text)

  for (const { target, line } of linksIn(text)) {
    const where = `${file}:${line}`

    if (/^(mailto:|tel:)/i.test(target)) continue

    if (/^https?:\/\//i.test(target)) {
      checked++
      // Not fetched — but a permalink to a moving document is still wrong.
      if (/gitea\.whitlocktech\.com\/.*\/src\/commit\//.test(target)) {
        problems.push(`${where}: links to a commit snapshot, not a branch — ${target}`)
      }
      continue
    }

    if (target.startsWith('#')) {
      checked++
      const slug = decodeURIComponent(target.slice(1)).toLowerCase()
      if (!selfAnchors.has(slug)) problems.push(`${where}: no heading matches ${target}`)
      continue
    }

    checked++
    const [rawPath, rawAnchor] = target.split('#')
    // A PR/issue template's text is INLINED into a pull request or issue body, and
    // Gitea resolves relative links in those against the repo root — not against
    // `.gitea/`, where the file itself lives. So `[CONTRIBUTING.md](CONTRIBUTING.md)`
    // is correct in a template and would be wrong anywhere else. Resolve those from
    // the root, or this check reports every template link as broken and gets muted.
    const base = file.startsWith('.gitea/') ? ROOT : path.dirname(path.join(ROOT, file))
    const resolved = path.resolve(base, decodeURIComponent(rawPath))
    const rel = path.relative(ROOT, resolved).split(path.sep).join('/')

    if (rel.startsWith('..')) {
      problems.push(`${where}: points outside the repo — ${target}`)
      continue
    }
    if (!fs.existsSync(resolved)) {
      problems.push(`${where}: no such file — ${target}`)
      continue
    }
    if (rawAnchor && resolved.toLowerCase().endsWith('.md')) {
      const slug = decodeURIComponent(rawAnchor).toLowerCase()
      if (!anchorsOf(rel).has(slug)) {
        problems.push(`${where}: ${rawPath} has no heading matching #${rawAnchor}`)
      }
    }
  }
}

if (problems.length) {
  console.error(`checkLinks: ${problems.length} problem(s) in ${files.length} file(s):\n`)
  for (const p of problems) console.error(`  ${p}`)
  process.exit(1)
}

if (!QUIET) {
  console.log(`checkLinks: ${checked} link(s) across ${files.length} markdown file(s) — OK`)
}