Last active
November 28, 2020 13:44
-
-
Save andreich1980/a58deabb237e13c162b581530d262b93 to your computer and use it in GitHub Desktop.
Script that helps to detect broken links in md files (specifically - laravel/docs)
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
<?php | |
/** | |
* Parses *.md files in the current directory for broken links. | |
* - [Link](#broken-anchor) without <a name="broken-anchor"></a> within the same file | |
* - [Link](another-file#broken-anchor) without <a name="broken-anchor"></a> in the another-file.md | |
*/ | |
$ok = $fail = 0; | |
foreach (glob('*.md') as $file) { | |
if ($brokenLinks = findBrokenLinks($file)) { | |
printError($file, $brokenLinks); | |
$fail++; | |
} else { | |
printOk($file, $brokenLinks); | |
$ok++; | |
} | |
} | |
print PHP_EOL . "OK: {$ok}. FAIL: {$fail}."; | |
function findBrokenLinks($file) | |
{ | |
$contents = file_get_contents($file); | |
// Find all the links | |
preg_match_all('/\[[^]]+\]\(([-a-z0-9#]+)\)/', $contents, $links); | |
// Exclude external links | |
$links = array_filter($links[1], function ($link) { | |
return 'http' !== substr($link, 0, 4) && 'ftp' !== substr($link, 0, 3); | |
}); | |
// Filter broken links | |
return array_filter($links, function ($link) use ($contents) { | |
if ('#' !== substr($link, 0, 1)) { | |
[$file, $link] = parseFileData($link); | |
// if the link was [Link](another-file) without an achor then just check file existance | |
if (is_null($link)) { | |
return !file_exists($file); | |
} | |
$contents = file_get_contents($file); | |
} else { | |
$link = substr($link, 1); | |
} | |
return false === strpos($contents, "<a name=\"{$link}\"></a>"); | |
}); | |
} | |
function parseFileData($link) | |
{ | |
$linkData = explode('#', $link); | |
$link = $linkData[1] ?? null; | |
$file = "{$linkData[0]}.md"; | |
return [ | |
file_exists($file) ? file_get_contents($file) : null, | |
$linkData[1] ?? null | |
]; | |
} | |
function printOk($file, $links) | |
{ | |
printInfo($file, $links, "OK", "\033[32m"); | |
} | |
function printError($file, $links) | |
{ | |
printInfo($file, $links, "FAIL", "\033[31m"); | |
} | |
function printInfo($file, $links, $caption = "INFO", $color = "\033[0m") | |
{ | |
$default = "\033[0m"; | |
print "[{$color}{$caption}{$default}] {$file}" . PHP_EOL; | |
foreach ($links as $link) { | |
print " - {$link}" . PHP_EOL; | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment