X-Git-Url: http://git.shiar.net/minimedit.git/blobdiff_plain/41811556ca2ca3ca29191bee012c35e5f40a3995..dcf14e6ed4093fd464367806e2211bb4bec3c8ba:/article.inc.php diff --git a/article.inc.php b/article.inc.php index 336511e..503c5a1 100644 --- a/article.inc.php +++ b/article.inc.php @@ -17,16 +17,32 @@ function showdate($parts) class ArchiveArticle { public $raw, $preface, $title, $body; + public $meta = []; function __construct($path) { - $this->page = $path; - $this->link = preg_replace('{(?:/index)?\.html$}', '', $path); - if (file_exists($this->page)) { - $this->raw = file_get_contents($this->page); - @list ($this->preface, $this->title, $this->body) = - preg_split('{
(.*?)
}s', $this->story, $bodyp)) { + if ($override = @$this->meta['og:description']) { + # prefer specific page description if found in metadata + return $override; + } + + if (preg_match('{ + (?: \s+ ||
\s* (.*?)
+ }sx', $this->raw, $bodyp, PREG_OFFSET_CAPTURE)) { + # fallback paragraph contents following the page header + if ($bodyp[1][1] < 256) { + return $bodyp[1][0]; + } + } + + # starting paragraph for documents without title (assumed simple/partial) + if (strpos($this->raw, '\s* (.*?)
+ }sx', $this->raw, $bodyp)) { return $bodyp[1]; } } @@ -95,6 +153,11 @@ class ArchiveArticle } function image() { + if ($override = @$this->meta['og:image']) { + # prefer specific page image if found in metadata + return $override; + } + if ( preg_match('/\bsrc="([^"]*)"/', $this->img, $src) ) { return $src[1]; } @@ -108,3 +171,41 @@ class ArchiveArticle ); } } + +class PageSearch +{ + function __construct($path = '.') + { + $this->iterator = new RecursiveCallbackFilterIterator( + new RecursiveDirectoryIterator($path), + function ($current) { + if ($current->getFilename()[0] === '.') { + # skip hidden files and directories + return FALSE; + } + if ($current->isLink()) { + # ignore symlinks, original contents only + return FALSE; + } + if ($current->isDir()) { + # traverse subdirectories unless untracked in any amount + return !file_exists("$current/.gitignore"); + } + # match **/*.html + return preg_match('/(?getFilename()); + } + ); + } + + function files() + { + # order alphabetically by link + $dir = iterator_to_array(new RecursiveIteratorIterator($this->iterator)); + array_walk($dir, function (&$row, $name) { + # prepare values for sorting (directory index first) + $row = preg_replace('{/index\.html$}', '', $name); + }); + asort($dir); + return $dir; + } +}