class ArchiveArticle
{
public $raw, $preface, $title, $body;
+ public $meta = [];
function __construct($path)
{
- $this->page = $path;
- $this->link = preg_replace('{(?:/index)?\.html$}', '', $path);
+ $this->page = preg_replace('{^\.(?:/|$)}', '', $path);
+ $this->link = preg_replace('{(?:/index)?\.html$}', '', $this->page);
if (file_exists($this->page)) {
$this->raw = file_get_contents($this->page);
+
+ if (preg_match_all('{
+ \G <meta \s+ property="( [^"]+ )" \s+ content="( [^"]* )" > \s*
+ }x', $this->raw, $meta)) {
+ $matchlen = array_sum(array_map('strlen', $meta[0]));
+ $this->raw = substr($this->raw, $matchlen); # delete matched contents
+ $this->meta = array_combine($meta[1], $meta[2]); # [property => content]
+ }
+
@list ($this->preface, $this->title, $this->body) =
- preg_split('{<h2>(.*?)</h2>\s*}', $this->raw, 2, PREG_SPLIT_DELIM_CAPTURE);
+ preg_split('{<h2>(.*?)</h2>\s*}s', $this->raw, 2, PREG_SPLIT_DELIM_CAPTURE);
}
}
function safetitle()
{
- return trim(strip_tags($this->title));
+ return trim($this->meta['og:title'] ?? strip_tags($this->title));
}
function name()
{
}
return $this->body;
}
+
function teaser()
{
- if (preg_match('{
- <meta \s+ name="description" [^>]* content="([^">]*)"
- }x', $this->preface, $meta)) {
- # prefer specific page description if found (assume before title)
- #TODO: strip from body contents
- return $meta[1];
+ if ($override = @$this->meta['og:description']) {
+ # prefer specific page description if found in metadata
+ return $override;
}
+
if (preg_match('{
- </h2> (?: \s+ | <p\sclass="nav\b.*?</p> | <div[^>]*> )* <p> (.*?) </p>
+ </h2> (?: \s+ | <p\sclass="nav\b.*?</p> | <div[^>]*> )* <p> \s* (.*?) </p>
}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, '<h2') === FALSE and preg_match('{
+ \A (?: <div [^>]*> \s* )* <p> \s* (.*?) </p>
+ }sx', $this->raw, $bodyp)) {
+ return $bodyp[1];
+ }
}
function img()
}
function image()
{
+ if ($override = @$this->meta['og:image']) {
+ # prefer specific page image if found in metadata
+ return $override;
+ }
+
if ( preg_match('/\bsrc="([^"]*)"/', $this->img, $src) ) {
return $src[1];
}
);
}
}
+
+class PageSearch
+{
+ function __construct($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("$current/.gitignore");
+ }
+ # match **/*.html
+ return preg_match('/(?<!\.inc)\.html$/', $current->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;
+ }
+}