article.inc.php

   1 <?php
   2 global $monthname;
   3 $monthname = ['?',
   4         'januari', 'februari', 'maart', 'april', 'mei', 'juni',
   5         'juli', 'augustus', 'september', 'oktober', 'november', 'december',
   6 ];
   7
   8 function showdate($parts)
   9 {
  10         global $monthname;
  11         return implode(' ', array_filter([
  12                 intval(@$parts[2]), $parts[1] > 0 ? $monthname[intval($parts[1])] : '', $parts[0],
  13                 count($parts) > 5 ? "$parts[3]:$parts[4]" : '',
  14         ]));
  15 }
  16
  17 class ArchiveArticle
  18 {
  19         public $raw, $preface, $title, $body;
  20         public $meta = [];
  21
  22         function __construct($path)
  23         {
  24                 $this->page = preg_replace('{^\.(?:/|$)}', '', $path);
  25                 $this->link = preg_replace('{(?:/index)?\.html$}', '', $this->page);
  26                 $this->raw($this->page);
  27         }
  28
  29         function raw($page)
  30         {
  31                 if (!file_exists($page)) {
  32                         return;
  33                 }
  34                 $this->raw = file_get_contents($page);
  35
  36                 if (preg_match_all('{
  37                         \G <meta \s+ property="( [^"]+ )" \s+ content="( [^"]* )" > \s*
  38                 }x', $this->raw, $meta)) {
  39                         $matchlen = array_sum(array_map('strlen', $meta[0]));
  40                         $this->raw = substr($this->raw, $matchlen); # delete matched contents
  41                         $this->meta = array_combine($meta[1], $meta[2]); # [property => content]
  42                 }
  43
  44                 @list ($this->preface, $this->title, $this->body) =
  45                         preg_split('{<h2>(.*?)</h2>\s*}s', $this->raw, 2, PREG_SPLIT_DELIM_CAPTURE);
  46         }
  47
  48         function __get($col)
  49         {
  50                 return $this->$col = $this->$col();  # run method and cache
  51         }
  52
  53         function safetitle()
  54         {
  55                 return trim($this->meta['og:title'] ?? strip_tags($this->title));
  56         }
  57         function name()
  58         {
  59                 return $this->safetitle ?: $this->link;
  60         }
  61
  62         function last()
  63         {
  64                 return filemtime($this->page);
  65         }
  66         function lastiso()
  67         {
  68                 return date(DATE_ATOM, $this->last);
  69         }
  70
  71         function dateparts()
  72         {
  73                 preg_match('< / (\d{4}) [/-] (\d{2}) (?:- (\d{2}) )? - >x', $this->page, $ymd);
  74                 array_shift($ymd);
  75                 return $ymd;
  76         }
  77         function dateiso()
  78         {
  79                 return implode('-', $this->dateparts()) . 'T12:00:00+02:00';
  80         }
  81         function date()
  82         {
  83                 return showdate($this->dateparts);
  84         }
  85
  86         function story()
  87         {
  88                 if ( preg_match('{
  89                         \n (?: < (?: p | figure [^>]* ) >\s* )+ (<img\ [^>]*>) | \n <hr\ />
  90                 }x', $this->body, $img, PREG_OFFSET_CAPTURE) ) {
  91                         # strip part after matching divider (image)
  92                         if (isset($img[1])) {
  93                                 $this->img = $img[1][0];
  94                         }
  95                         return substr($this->body, 0, $img[0][1]);
  96                 }
  97                 return $this->body;
  98         }
  99
 100         function teaser()
 101         {
 102                 if ($override = @$this->meta['og:description']) {
 103                         # prefer specific page description if found in metadata
 104                         return $override;
 105                 }
 106
 107                 if (preg_match('{
 108                         </h2> (?: \s+ | <p\sclass="nav\b.*?</p> | <div[^>]*> )* <p> \s* (.*?) </p>
 109                 }sx', $this->raw, $bodyp, PREG_OFFSET_CAPTURE)) {
 110                         # fallback paragraph contents following the page header
 111                         if ($bodyp[1][1] < 256) {
 112                                 return $bodyp[1][0];
 113                         }
 114                 }
 115
 116                 # starting paragraph for documents without title (assumed simple/partial)
 117                 if (strpos($this->raw, '<h2') === FALSE and preg_match('{
 118                         \A (?: <div [^>]*> \s* )* <p> \s* (.*?) </p>
 119                 }sx', $this->raw, $bodyp)) {
 120                         return $bodyp[1];
 121                 }
 122         }
 123
 124         function img()
 125         {
 126                 $this->img = NULL;
 127                 $this->story;
 128                 return $this->img;
 129         }
 130         function image()
 131         {
 132                 if ($override = @$this->meta['og:image']) {
 133                         # prefer specific page image if found in metadata
 134                         return $override;
 135                 }
 136
 137                 if ( preg_match('/\bsrc="([^"]*)"/', $this->img, $src) ) {
 138                         return $src[1];
 139                 }
 140         }
 141         function thumb($size = '300x')
 142         {
 143                 if (!$this->image or $this->image[0] !== '/') return;
 144                 return preg_replace(
 145                         ['{^(?:/thumb/[^/]*)?}', '/\.groot(?=\.\w+$)/'], ["thumb/$size", ''],
 146                         $this->image
 147                 );
 148         }
 149 }
 150
 151 class PageSearch
 152 {
 153         function __construct($path = '.')
 154         {
 155                 $this->iterator = new RecursiveCallbackFilterIterator(
 156                         new RecursiveDirectoryIterator($path),
 157                         function ($current) {
 158                                 if ($current->getFilename()[0] === '.') {
 159                                         # skip hidden files and directories
 160                                         return FALSE;
 161                                 }
 162                                 if ($current->isLink()) {
 163                                         # ignore symlinks, original contents only
 164                                         return FALSE;
 165                                 }
 166                                 if ($current->isDir()) {
 167                                         # traverse subdirectories unless untracked in any amount
 168                                         return !file_exists("$current/.gitignore");
 169                                 }
 170                                 # match **/*.html
 171                                 return preg_match('/(?<!\.inc)\.html$/', $current->getFilename());
 172                         }
 173                 );
 174         }
 175
 176         function files()
 177         {
 178                 # order alphabetically by link
 179                 $dir = iterator_to_array(new RecursiveIteratorIterator($this->iterator));
 180                 array_walk($dir, function (&$row, $name) {
 181                         # prepare values for sorting (directory index first)
 182                         $row = preg_replace('{/index\.html$}', '', $name);
 183                 });
 184                 asort($dir);
 185                 return $dir;
 186         }
 187 }