mirror of
https://github.com/stablyai/orca.git
synced 2026-11-05 19:47:27 +00:00
Blue Daydreaming
0b604956ec
Only srcset is a comma-separated candidate list; src and href hold one URL. Splitting them invented missing targets for paths containing commas or spaces and could pass a missing target whose fragments were tracked. Fixes #34815
103 lines
3.4 KiB
JavaScript
103 lines
3.4 KiB
JavaScript
import { execFileSync } from 'node:child_process'
|
|
import { readdirSync, readFileSync } from 'node:fs'
|
|
import path from 'node:path'
|
|
import process from 'node:process'
|
|
import { pathToFileURL } from 'node:url'
|
|
|
|
// Docs-only CI skips some media owners; check index paths because GitHub cannot render uncommitted docs-site/feature-wall assets.
|
|
const TRANSLATED_README_DIR = path.join('docs', 'readme')
|
|
const EXTERNAL_TARGET = /^(?:[a-z][a-z0-9+.-]*:|#|\/\/)/i
|
|
const HTML_ATTRIBUTE = /\b(src|srcset|href)\s*=\s*(?:"([^"]*)"|'([^']*)')/g
|
|
const MARKDOWN_LINK = /!?\[[^\]]*\]\(([^)\s]+)(?:\s+"[^"]*")?\)/g
|
|
|
|
function readmeFiles(root) {
|
|
const translated = readdirSync(path.join(root, TRANSLATED_README_DIR))
|
|
.filter((name) => name.endsWith('.md'))
|
|
.sort()
|
|
.map((name) => path.posix.join('docs', 'readme', name))
|
|
return ['README.md', ...translated]
|
|
}
|
|
|
|
// Bound ls-files to referenced pathspecs; full-tree output overflows the child buffer.
|
|
function trackedFiles(root, candidates) {
|
|
if (candidates.length === 0) {
|
|
return new Set()
|
|
}
|
|
const stdout = execFileSync(
|
|
'git',
|
|
['--literal-pathspecs', 'ls-files', '-z', '--', ...candidates],
|
|
{ cwd: root, encoding: 'utf8' }
|
|
)
|
|
return new Set(stdout.split('\0').filter(Boolean))
|
|
}
|
|
|
|
function* localTargets(markdown) {
|
|
for (const [, attribute, doubleQuoted, singleQuoted] of markdown.matchAll(HTML_ATTRIBUTE)) {
|
|
const value = doubleQuoted ?? singleQuoted
|
|
// Only srcset lists candidates; each begins with its URL. src/href hold one whole URL.
|
|
const targets =
|
|
attribute === 'srcset'
|
|
? value.split(',').map((candidate) => candidate.trim().split(/\s+/)[0])
|
|
: [value.trim()]
|
|
for (const target of targets) {
|
|
if (target) {
|
|
yield target
|
|
}
|
|
}
|
|
}
|
|
for (const match of markdown.matchAll(MARKDOWN_LINK)) {
|
|
yield match[1].replace(/^<|>$/g, '')
|
|
}
|
|
}
|
|
|
|
function resolveTarget(readme, target) {
|
|
const bare = target.split(/[?#]/)[0]
|
|
if (!bare) {
|
|
return null
|
|
}
|
|
const resolved = path.posix.normalize(
|
|
path.posix.join(path.posix.dirname(readme), decodeURIComponent(bare))
|
|
)
|
|
return resolved.startsWith('../') ? null : resolved
|
|
}
|
|
|
|
function collectLinks(root) {
|
|
const links = []
|
|
for (const readme of readmeFiles(root)) {
|
|
const markdown = readFileSync(path.join(root, readme), 'utf8')
|
|
for (const target of new Set(localTargets(markdown))) {
|
|
if (EXTERNAL_TARGET.test(target)) {
|
|
continue
|
|
}
|
|
links.push({ readme, target, resolved: resolveTarget(readme, target) })
|
|
}
|
|
}
|
|
return links
|
|
}
|
|
|
|
export function findBrokenReadmeLinks(root) {
|
|
const links = collectLinks(root)
|
|
const candidates = [...new Set(links.map((link) => link.resolved).filter(Boolean))]
|
|
const tracked = trackedFiles(root, candidates)
|
|
return links.filter(({ resolved }) => resolved === null || !tracked.has(resolved))
|
|
}
|
|
|
|
export function main(root = process.cwd()) {
|
|
const broken = findBrokenReadmeLinks(root)
|
|
if (broken.length > 0) {
|
|
console.error(`README local link check failed with ${broken.length} broken link(s):`)
|
|
for (const { readme, target, resolved } of broken) {
|
|
console.error(
|
|
`- ${readme}: ${target} -> ${resolved ?? 'outside the repository'} is not tracked`
|
|
)
|
|
}
|
|
return 1
|
|
}
|
|
console.log('README local link check passed.')
|
|
return 0
|
|
}
|
|
|
|
if (process.argv[1] && import.meta.url === pathToFileURL(process.argv[1]).href) {
|
|
process.exit(main())
|
|
}
|