| 1 |
<?php |
| 2 |
// This is the path to markdown.php |
| 3 |
if (!defined('WORDPRESS_README_MARKDOWN')) { |
| 4 |
if (defined('AUTOMATTIC_README_MARKDOWN')) |
| 5 |
define('WORDPRESS_README_MARKDOWN', AUTOMATTIC_README_MARKDOWN); |
| 6 |
else |
| 7 |
define('WORDPRESS_README_MARKDOWN', JLTMA_PATH . '/lib/markdown.php'); |
| 8 |
} |
| 9 |
|
| 10 |
class WordPress_Readme_Parser |
| 11 |
{ |
| 12 |
|
| 13 |
function __construct() |
| 14 |
{ |
| 15 |
// This space intentially blank |
| 16 |
} |
| 17 |
|
| 18 |
function parse_readme($file) |
| 19 |
{ |
| 20 |
$file_contents = @implode('', @file($file)); |
| 21 |
return $this->parse_readme_contents($file_contents); |
| 22 |
} |
| 23 |
|
| 24 |
function parse_readme_contents($file_contents) |
| 25 |
{ |
| 26 |
$file_contents = str_replace(array("\r\n", "\r"), "\n", $file_contents); |
| 27 |
$file_contents = trim($file_contents); |
| 28 |
if (0 === strpos($file_contents, "\xEF\xBB\xBF")) |
| 29 |
$file_contents = substr($file_contents, 3); |
| 30 |
|
| 31 |
// Markdown transformations |
| 32 |
$file_contents = preg_replace("|^###([^#]+)#*?\s*?\n|im", '=$1=' . "\n", $file_contents); |
| 33 |
$file_contents = preg_replace("|^##([^#]+)#*?\s*?\n|im", '==$1==' . "\n", $file_contents); |
| 34 |
$file_contents = preg_replace("|^#([^#]+)#*?\s*?\n|im", '===$1===' . "\n", $file_contents); |
| 35 |
|
| 36 |
// === Plugin Name === |
| 37 |
// Must be the very first thing. |
| 38 |
if (!preg_match('|^===(.*)===|', $file_contents, $_name)) |
| 39 |
return array(); // require a name |
| 40 |
$name = trim($_name[1], '='); |
| 41 |
$name = $this->sanitize_text($name); |
| 42 |
|
| 43 |
$file_contents = $this->chop_string($file_contents, $_name[0]); |
| 44 |
|
| 45 |
|
| 46 |
// Requires at least: 1.5 |
| 47 |
if (preg_match('|Requires at least:(.*)|i', $file_contents, $_requires_at_least)) |
| 48 |
$requires_at_least = $this->sanitize_text($_requires_at_least[1]); |
| 49 |
else |
| 50 |
$requires_at_least = NULL; |
| 51 |
|
| 52 |
|
| 53 |
// Tested up to: 2.1 |
| 54 |
if (preg_match('|Tested up to:(.*)|i', $file_contents, $_tested_up_to)) |
| 55 |
$tested_up_to = $this->sanitize_text($_tested_up_to[1]); |
| 56 |
else |
| 57 |
$tested_up_to = NULL; |
| 58 |
|
| 59 |
|
| 60 |
// Stable tag: 10.4-ride-the-fire-eagle-danger-day |
| 61 |
if (preg_match('|Stable tag:(.*)|i', $file_contents, $_stable_tag)) |
| 62 |
$stable_tag = $this->sanitize_text($_stable_tag[1]); |
| 63 |
else |
| 64 |
$stable_tag = NULL; // we assume trunk, but don't set it here to tell the difference between specified trunk and default trunk |
| 65 |
|
| 66 |
|
| 67 |
// Tags: some tag, another tag, we like tags |
| 68 |
if (preg_match('|Tags:(.*)|i', $file_contents, $_tags)) { |
| 69 |
$tags = preg_split('|,[\s]*?|', trim($_tags[1])); |
| 70 |
foreach (array_keys($tags) as $t) |
| 71 |
$tags[$t] = $this->sanitize_text($tags[$t]); |
| 72 |
} else { |
| 73 |
$tags = array(); |
| 74 |
} |
| 75 |
|
| 76 |
|
| 77 |
// Contributors: markjaquith, mdawaffe, zefrank |
| 78 |
$contributors = array(); |
| 79 |
if (preg_match('|Contributors:(.*)|i', $file_contents, $_contributors)) { |
| 80 |
$temp_contributors = preg_split('|,[\s]*|', trim($_contributors[1])); |
| 81 |
foreach (array_keys($temp_contributors) as $c) { |
| 82 |
$tmp_sanitized = $this->user_sanitize($temp_contributors[$c]); |
| 83 |
if (strlen(trim($tmp_sanitized)) > 0) |
| 84 |
$contributors[$c] = $tmp_sanitized; |
| 85 |
unset($tmp_sanitized); |
| 86 |
} |
| 87 |
} |
| 88 |
|
| 89 |
|
| 90 |
// Donate Link: URL |
| 91 |
if (preg_match('|Donate link:(.*)|i', $file_contents, $_donate_link)) |
| 92 |
$donate_link = esc_url($_donate_link[1]); |
| 93 |
else |
| 94 |
$donate_link = NULL; |
| 95 |
|
| 96 |
|
| 97 |
// togs, conts, etc are optional and order shouldn't matter. So we chop them only after we've grabbed their values. |
| 98 |
foreach (array('tags', 'contributors', 'requires_at_least', 'tested_up_to', 'stable_tag', 'donate_link') as $chop) { |
| 99 |
if ($$chop) { |
| 100 |
$_chop = '_' . $chop; |
| 101 |
$file_contents = $this->chop_string($file_contents, ${$_chop}[0]); |
| 102 |
} |
| 103 |
} |
| 104 |
|
| 105 |
$file_contents = trim($file_contents); |
| 106 |
|
| 107 |
|
| 108 |
// short-description fu |
| 109 |
if (!preg_match('/(^(.*?))^[\s]*=+?[\s]*.+?[\s]*=+?/ms', $file_contents, $_short_description)) |
| 110 |
$_short_description = array(1 => &$file_contents, 2 => &$file_contents); |
| 111 |
$short_desc_filtered = $this->sanitize_text($_short_description[2]); |
| 112 |
$short_desc_length = strlen($short_desc_filtered); |
| 113 |
$short_description = substr($short_desc_filtered, 0, 150); |
| 114 |
if ($short_desc_length > strlen($short_description)) |
| 115 |
$truncated = true; |
| 116 |
else |
| 117 |
$truncated = false; |
| 118 |
if ($_short_description[1]) |
| 119 |
$file_contents = $this->chop_string($file_contents, $_short_description[1]); // yes, the [1] is intentional |
| 120 |
|
| 121 |
// == Section == |
| 122 |
// Break into sections |
| 123 |
// $_sections[0] will be the title of the first section, $_sections[1] will be the content of the first section |
| 124 |
// the array alternates from there: title2, content2, title3, content3... and so forth |
| 125 |
$_sections = preg_split('/^[\s]*==[\s]*(.+?)[\s]*==/m', $file_contents, -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY); |
| 126 |
|
| 127 |
$sections = array(); |
| 128 |
for ($i = 1; $i <= count($_sections); $i += 2) { |
| 129 |
$_sections[$i] = preg_replace('/^[\s]*=[\s]+(.+?)[\s]+=/m', '<h4>$1</h4>', isset($_sections[$i]) ? $_sections[$i] : ""); |
| 130 |
$_sections[$i] = $this->filter_text($_sections[$i], true); |
| 131 |
$title = $this->sanitize_text($_sections[$i - 1]); |
| 132 |
$sections[str_replace(' ', '_', strtolower($title))] = array('title' => $title, 'content' => $_sections[$i]); |
| 133 |
} |
| 134 |
|
| 135 |
|
| 136 |
// Special sections |
| 137 |
// This is where we nab our special sections, so we can enforce their order and treat them differently, if needed |
| 138 |
// upgrade_notice is not a section, but parse it like it is for now |
| 139 |
$final_sections = array(); |
| 140 |
foreach (array('description', 'installation', 'frequently_asked_questions', 'screenshots', 'changelog', 'change_log', 'upgrade_notice') as $special_section) { |
| 141 |
if (isset($sections[$special_section])) { |
| 142 |
$final_sections[$special_section] = $sections[$special_section]['content']; |
| 143 |
unset($sections[$special_section]); |
| 144 |
} |
| 145 |
} |
| 146 |
if (isset($final_sections['change_log']) && empty($final_sections['changelog'])) |
| 147 |
$final_sections['changelog'] = $final_sections['change_log']; |
| 148 |
|
| 149 |
|
| 150 |
$final_screenshots = array(); |
| 151 |
if (isset($final_sections['screenshots'])) { |
| 152 |
preg_match_all('|<li>(.*?)</li>|s', $final_sections['screenshots'], $screenshots, PREG_SET_ORDER); |
| 153 |
if ($screenshots) { |
| 154 |
foreach ((array) $screenshots as $ss) |
| 155 |
$final_screenshots[] = $ss[1]; |
| 156 |
} |
| 157 |
} |
| 158 |
|
| 159 |
// Parse the upgrade_notice section specially: |
| 160 |
// 1.0 => blah, 1.1 => fnord |
| 161 |
// if ( isset($final_sections['upgrade_notice']) ) { |
| 162 |
// $upgrade_notice = array(); |
| 163 |
// $split = preg_split( '#<h4>(.*?)</h4>#', $final_sections['upgrade_notice'], -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY ); |
| 164 |
// for ( $i = 0; $i < count( $split ); $i += 2 ) |
| 165 |
// $upgrade_notice[$this->sanitize_text( $split[$i] )] = substr( $this->sanitize_text( $split[$i + 1] ), 0, 300 ); |
| 166 |
// unset( $final_sections['upgrade_notice'] ); |
| 167 |
// } |
| 168 |
|
| 169 |
// No description? |
| 170 |
// No problem... we'll just fall back to the old style of description |
| 171 |
// We'll even let you use markup this time! |
| 172 |
$excerpt = false; |
| 173 |
if (!isset($final_sections['description'])) { |
| 174 |
$final_sections = array_merge(array('description' => $this->filter_text($_short_description[2], true)), $final_sections); |
| 175 |
$excerpt = true; |
| 176 |
} |
| 177 |
|
| 178 |
|
| 179 |
// dump the non-special sections into $remaining_content |
| 180 |
// their order will be determined by their original order in the readme.txt |
| 181 |
$remaining_content = ''; |
| 182 |
foreach ($sections as $s_name => $s_data) { |
| 183 |
$remaining_content .= "\n<h3>{$s_data['title']}</h3>\n{$s_data['content']}"; |
| 184 |
} |
| 185 |
$remaining_content = trim($remaining_content); |
| 186 |
|
| 187 |
|
| 188 |
// All done! |
| 189 |
// $r['tags'] and $r['contributors'] are simple arrays |
| 190 |
// $r['sections'] is an array with named elements |
| 191 |
$r = array( |
| 192 |
'name' => $name, |
| 193 |
'tags' => $tags, |
| 194 |
'requires_at_least' => $requires_at_least, |
| 195 |
'tested_up_to' => $tested_up_to, |
| 196 |
'stable_tag' => $stable_tag, |
| 197 |
'contributors' => $contributors, |
| 198 |
'donate_link' => $donate_link, |
| 199 |
'short_description' => $short_description, |
| 200 |
'screenshots' => $final_screenshots, |
| 201 |
'is_excerpt' => $excerpt, |
| 202 |
'is_truncated' => $truncated, |
| 203 |
'sections' => $final_sections, |
| 204 |
'remaining_content' => $remaining_content, |
| 205 |
// 'upgrade_notice' => $upgrade_notice |
| 206 |
); |
| 207 |
|
| 208 |
return $r; |
| 209 |
} |
| 210 |
|
| 211 |
function chop_string($string, $chop) |
| 212 |
{ // chop a "prefix" from a string: Agressive! uses strstr not 0 === strpos |
| 213 |
if ($_string = strstr($string, $chop)) { |
| 214 |
$_string = substr($_string, strlen($chop)); |
| 215 |
return trim($_string); |
| 216 |
} else { |
| 217 |
return trim($string); |
| 218 |
} |
| 219 |
} |
| 220 |
|
| 221 |
function user_sanitize($text, $strict = false) |
| 222 |
{ // whitelisted chars |
| 223 |
if (function_exists('user_sanitize')) // bbPress native |
| 224 |
return user_sanitize($text, $strict); |
| 225 |
|
| 226 |
if ($strict) { |
| 227 |
$text = preg_replace('/[^a-z0-9-]/i', '', $text); |
| 228 |
$text = preg_replace('|-+|', '-', $text); |
| 229 |
} else { |
| 230 |
$text = preg_replace('/[^a-z0-9_-]/i', '', $text); |
| 231 |
} |
| 232 |
return $text; |
| 233 |
} |
| 234 |
|
| 235 |
function sanitize_text($text) |
| 236 |
{ // not fancy |
| 237 |
$text = strip_tags($text); |
| 238 |
$text = esc_html($text); |
| 239 |
$text = trim($text); |
| 240 |
return $text; |
| 241 |
} |
| 242 |
|
| 243 |
function filter_text($text, $markdown = false) |
| 244 |
{ // fancy, Markdown |
| 245 |
$text = trim($text); |
| 246 |
|
| 247 |
$text = call_user_func(array(__CLASS__, 'code_trick'), $text, $markdown); // A better parser than Markdown's for: backticks -> CODE |
| 248 |
|
| 249 |
if ($markdown) { // Parse markdown. |
| 250 |
if (!function_exists('Markdown')) |
| 251 |
require(WORDPRESS_README_MARKDOWN); |
| 252 |
$text = Markdown($text); |
| 253 |
} |
| 254 |
|
| 255 |
$allowed = array( |
| 256 |
'a' => array( |
| 257 |
'href' => array(), |
| 258 |
'title' => array(), |
| 259 |
'rel' => array() |
| 260 |
), |
| 261 |
'blockquote' => array('cite' => array()), |
| 262 |
'br' => array(), |
| 263 |
'p' => array(), |
| 264 |
'code' => array(), |
| 265 |
'pre' => array(), |
| 266 |
'em' => array(), |
| 267 |
'strong' => array(), |
| 268 |
'ul' => array(), |
| 269 |
'ol' => array(), |
| 270 |
'li' => array(), |
| 271 |
'h3' => array(), |
| 272 |
'h4' => array() |
| 273 |
); |
| 274 |
|
| 275 |
$text = balanceTags($text); |
| 276 |
|
| 277 |
$text = wp_kses($text, $allowed); |
| 278 |
$text = trim($text); |
| 279 |
return $text; |
| 280 |
} |
| 281 |
|
| 282 |
function code_trick($text, $markdown) |
| 283 |
{ // Don't use bbPress native function - it's incompatible with Markdown |
| 284 |
// If doing markdown, first take any user formatted code blocks and turn them into backticks so that |
| 285 |
// markdown will preserve things like underscores in code blocks |
| 286 |
if ($markdown) |
| 287 |
$text = preg_replace_callback("!(<pre><code>|<code>)(.*?)(</code></pre>|</code>)!s", array(__CLASS__, 'decodeit'), $text); |
| 288 |
|
| 289 |
$text = str_replace(array("\r\n", "\r"), "\n", $text); |
| 290 |
if (!$markdown) { |
| 291 |
// This gets the "inline" code blocks, but can't be used with Markdown. |
| 292 |
$text = preg_replace_callback("|(`)(.*?)`|", array(__CLASS__, 'encodeit'), $text); |
| 293 |
// This gets the "block level" code blocks and converts them to PRE CODE |
| 294 |
$text = preg_replace_callback("!(^|\n)`(.*?)`!s", array(__CLASS__, 'encodeit'), $text); |
| 295 |
} else { |
| 296 |
// Markdown can do inline code, we convert bbPress style block level code to Markdown style |
| 297 |
$text = preg_replace_callback("!(^|\n)([ \t]*?)`(.*?)`!s", array(__CLASS__, 'indent'), $text); |
| 298 |
} |
| 299 |
return $text; |
| 300 |
} |
| 301 |
|
| 302 |
function indent($matches) |
| 303 |
{ |
| 304 |
$text = $matches[3]; |
| 305 |
$text = preg_replace('|^|m', $matches[2] . ' ', $text); |
| 306 |
return $matches[1] . $text; |
| 307 |
} |
| 308 |
|
| 309 |
function encodeit($matches) |
| 310 |
{ |
| 311 |
if (function_exists('encodeit')) // bbPress native |
| 312 |
return encodeit($matches); |
| 313 |
|
| 314 |
$text = trim($matches[2]); |
| 315 |
$text = htmlspecialchars($text, ENT_QUOTES); |
| 316 |
$text = str_replace(array("\r\n", "\r"), "\n", $text); |
| 317 |
$text = preg_replace("|\n\n\n+|", "\n\n", $text); |
| 318 |
$text = str_replace('&lt;', '<', $text); |
| 319 |
$text = str_replace('&gt;', '>', $text); |
| 320 |
$text = "<code>$text</code>"; |
| 321 |
if ("`" != $matches[1]) |
| 322 |
$text = "<pre>$text</pre>"; |
| 323 |
return $text; |
| 324 |
} |
| 325 |
|
| 326 |
function decodeit($matches) |
| 327 |
{ |
| 328 |
if (function_exists('decodeit')) // bbPress native |
| 329 |
return decodeit($matches); |
| 330 |
|
| 331 |
$text = $matches[2]; |
| 332 |
$trans_table = array_flip(get_html_translation_table(HTML_ENTITIES)); |
| 333 |
$text = strtr($text, $trans_table); |
| 334 |
$text = str_replace('<br />', '', $text); |
| 335 |
$text = str_replace('&', '&', $text); |
| 336 |
$text = str_replace(''', "'", $text); |
| 337 |
if ('<pre><code>' == $matches[1]) |
| 338 |
$text = "\n$text\n"; |
| 339 |
return "`$text`"; |
| 340 |
} |
| 341 |
} // end class |
| 342 |
|
| 343 |
// $li = new WordPress_Readme_Parser(); |
| 344 |
// Class Automattic_Readme extends WordPress_Readme_Parser {} |
| 345 |
|
