seriot.ch

".$this->runSpanGamut($matches[1])."

".$this->runSpanGamut($matches[2])."

" . $item . "

` blocks. 1092 | # 1093 | $text = preg_replace_callback('{ 1094 | (?:\n\n|\A) 1095 | ( # $1 = the code block -- one or more lines, starting with a space/tab 1096 | (?: 1097 | (?:[ ]{'.$this->tab_width.'} | \t) # Lines must start with a tab or a tab-width of spaces 1098 | .*\n+ 1099 | )+ 1100 | ) 1101 | ((?=^[ ]{0,'.$this->tab_width.'}\S)|\Z) # Lookahead for non-space at line-start, or end of doc 1102 | }xm', 1103 | array(&$this, '_doCodeBlocks_callback'), $text); 1104 | 1105 | return $text; 1106 | } 1107 | function _doCodeBlocks_callback($matches) { 1108 | $codeblock = $matches[1]; 1109 | 1110 | $codeblock = $this->encodeCode($this->outdent($codeblock)); 1111 | // $codeblock = $this->detab($codeblock); 1112 | # trim leading newlines and trailing whitespace 1113 | $codeblock = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $codeblock); 1114 | 1115 | $result = "\n\n".$this->hashBlock("" . $codeblock . "\n")."\n\n"; 1116 | 1117 | return $result; 1118 | } 1119 | 1120 | 1121 | function doCodeSpans($text) { 1122 | # 1123 | # * Backtick quotes are used for spans. 1124 | # 1125 | # * You can use multiple backticks as the delimiters if you want to 1126 | # include literal backticks in the code span. So, this input: 1127 | # 1128 | # Just type ``foo `bar` baz`` at the prompt. 1129 | # 1130 | # Will translate to: 1131 | # 1132 | # Just type foo `bar` baz at the prompt. 1133 | # 1134 | # There's no arbitrary limit to the number of backticks you 1135 | # can use as delimters. If you need three consecutive backticks 1136 | # in your code, use four for delimiters, etc. 1137 | # 1138 | # * You can use spaces to get literal backticks at the edges: 1139 | # 1140 | # ... type `` `bar` `` ... 1141 | # 1142 | # Turns to: 1143 | # 1144 | # ... type `bar` ... 1145 | # 1146 | $text = preg_replace_callback('@ 1147 | (?encodeCode($c); 1163 | return $this->hashSpan("$c"); 1164 | } 1165 | 1166 | 1167 | function encodeCode($_) { 1168 | # 1169 | # Encode/escape certain characters inside Markdown code runs. 1170 | # The point is that in code, these characters are literals, 1171 | # and lose their special Markdown meanings. 1172 | # 1173 | # Encode all ampersands; HTML entities are not 1174 | # entities within a Markdown code span. 1175 | $_ = str_replace('&', '&', $_); 1176 | 1177 | # Do the angle bracket song and dance: 1178 | $_ = str_replace(array('<', '>'), 1179 | array('<', '>'), $_); 1180 | 1181 | # Now, escape characters that are magic in Markdown: 1182 | // $_ = str_replace(array_keys($this->escape_table), 1183 | // array_values($this->escape_table), $_); 1184 | 1185 | return $_; 1186 | } 1187 | 1188 | 1189 | function doItalicsAndBold($text) { 1190 | # must go first: 1191 | $text = preg_replace_callback('{ 1192 | ( # $1: Marker 1193 | (? 1201 | [^*_]+? # Anthing not em markers. 1202 | | 1203 | # Balence any regular emphasis inside. 1204 | \1 (?=\S) .+? (?<=\S) \1 1205 | | 1206 | . # Allow unbalenced * and _. 1207 | )+? 1208 | ) 1209 | (?<=\S) \1\1 # End mark not preceded by whitespace. 1210 | }sx', 1211 | array(&$this, '_doItalicAndBold_strong_callback'), $text); 1212 | # Then : 1213 | $text = preg_replace_callback( 1214 | '{ ( (?runSpanGamut($text); 1222 | return $this->hashSpan("$text"); 1223 | } 1224 | function _doItalicAndBold_strong_callback($matches) { 1225 | $text = $matches[2]; 1226 | $text = $this->runSpanGamut($text); 1227 | return $this->hashSpan("$text"); 1228 | } 1229 | 1230 | 1231 | function doBlockQuotes($text) { 1232 | $text = preg_replace_callback('/ 1233 | ( # Wrap whole match in $1 1234 | ( 1235 | ^[ ]*>[ ]? # ">" at the start of a line 1236 | .+\n # rest of the first line 1237 | (.+\n)* # subsequent consecutive lines 1238 | \n* # blanks 1239 | )+ 1240 | ) 1241 | /xm', 1242 | array(&$this, '_doBlockQuotes_callback'), $text); 1243 | 1244 | return $text; 1245 | } 1246 | function _doBlockQuotes_callback($matches) { 1247 | $bq = $matches[1]; 1248 | # trim one level of quoting - trim whitespace-only lines 1249 | $bq = preg_replace(array('/^[ ]*>[ ]?/m', '/^[ ]+$/m'), '', $bq); 1250 | $bq = $this->runBlockGamut($bq); # recurse 1251 | 1252 | $bq = preg_replace('/^/m', " ", $bq); 1253 | # These leading spaces cause problem with content, 1254 | # so we need to fix that: 1255 | $bq = preg_replace_callback('{(\s*.+?)}sx', 1256 | array(&$this, '_DoBlockQuotes_callback2'), $bq); 1257 | 1258 | return "\n". $this->hashBlock("\n$bq\n")."\n\n"; 1259 | } 1260 | function _doBlockQuotes_callback2($matches) { 1261 | $pre = $matches[1]; 1262 | $pre = preg_replace('/^ /m', '', $pre); 1263 | return $pre; 1264 | } 1265 | 1266 | 1267 | function formParagraphs($text) { 1268 | # 1269 | # Params: 1270 | # $text - string to process with html tags 1271 | # 1272 | # Strip leading and trailing lines: 1273 | $text = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $text); 1274 | 1275 | $grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY); 1276 | 1277 | # 1278 | # Wrap tags. 1279 | # 1280 | foreach ($grafs as $key => $value) { 1281 | if (!isset( $this->html_blocks[$value] )) { 1282 | $value = $this->runSpanGamut($value); 1283 | $value = preg_replace('/^([ ]*)/', " ", $value); 1284 | $value .= ""; 1285 | $grafs[$key] = $this->unhash($value); 1286 | } 1287 | } 1288 | 1289 | # 1290 | # Unhashify HTML blocks 1291 | # 1292 | foreach ($grafs as $key => $graf) { 1293 | # Modify elements of @grafs in-place... 1294 | if (isset($this->html_blocks[$graf])) { 1295 | $block = $this->html_blocks[$graf]; 1296 | $graf = $block; 1297 | // if (preg_match('{ 1298 | // \A 1299 | // ( # $1 = tag 1300 | // ]* 1302 | // \b 1303 | // markdown\s*=\s* ([\'"]) # $2 = attr quote char 1304 | // 1 1305 | // \2 1306 | // [^>]* 1307 | // > 1308 | // ) 1309 | // ( # $3 = contents 1310 | // .* 1311 | // ) 1312 | // () # $4 = closing tag 1313 | // \z 1314 | // }xs', $block, $matches)) 1315 | // { 1316 | // list(, $div_open, , $div_content, $div_close) = $matches; 1317 | // 1318 | // # We can't call Markdown(), because that resets the hash; 1319 | // # that initialization code should be pulled into its own sub, though. 1320 | // $div_content = $this->hashHTMLBlocks($div_content); 1321 | // 1322 | // # Run document gamut methods on the content. 1323 | // foreach ($this->document_gamut as $method => $priority) { 1324 | // $div_content = $this->$method($div_content); 1325 | // } 1326 | // 1327 | // $div_open = preg_replace( 1328 | // '{\smarkdown\s*=\s*([\'"]).+?\1}', '', $div_open); 1329 | // 1330 | // $graf = $div_open . "\n" . $div_content . "\n" . $div_close; 1331 | // } 1332 | $grafs[$key] = $graf; 1333 | } 1334 | } 1335 | 1336 | return implode("\n\n", $grafs); 1337 | } 1338 | 1339 | 1340 | function encodeAmpsAndAngles($text) { 1341 | # Smart processing for ampersands and angle brackets that need to be encoded. 1342 | 1343 | # Ampersand-encoding based entirely on Nat Irons's Amputator MT plugin: 1344 | # http://bumppo.net/projects/amputator/ 1345 | $text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/', 1346 | '&', $text);; 1347 | 1348 | # Encode naked <'s 1349 | $text = preg_replace('{<(?![a-z/?\$!%])}i', '<', $text); 1350 | 1351 | return $text; 1352 | } 1353 | 1354 | 1355 | function encodeBackslashEscapes($text) { 1356 | # 1357 | # Parameter: String. 1358 | # Returns: The string, with after processing the following backslash 1359 | # escape sequences. 1360 | # 1361 | # Must process escaped backslashes first. 1362 | return str_replace(array_keys($this->backslash_escape_table), 1363 | array_values($this->backslash_escape_table), $text); 1364 | } 1365 | 1366 | 1367 | function doAutoLinks($text) { 1368 | $text = preg_replace_callback('{<((https?|ftp|dict):[^\'">\s]+)>}', 1369 | array(&$this, '_doAutoLinks_url_callback'), $text); 1370 | 1371 | # Email addresses: 1372 | $text = preg_replace_callback('{ 1373 | < 1374 | (?:mailto:)? 1375 | ( 1376 | [-.\w\x80-\xFF]+ 1377 | \@ 1378 | [-a-z0-9\x80-\xFF]+(\.[-a-z0-9\x80-\xFF]+)*\.[a-z]+ 1379 | ) 1380 | > 1381 | }xi', 1382 | array(&$this, '_doAutoLinks_email_callback'), $text); 1383 | 1384 | return $text; 1385 | } 1386 | function _doAutoLinks_url_callback($matches) { 1387 | $url = $this->encodeAmpsAndAngles($matches[1]); 1388 | $link = "$url"; 1389 | return $this->hashSpan($link); 1390 | } 1391 | function _doAutoLinks_email_callback($matches) { 1392 | $address = $matches[1]; 1393 | $link = $this->encodeEmailAddress($address); 1394 | return $this->hashSpan($link); 1395 | } 1396 | 1397 | 1398 | function encodeEmailAddress($addr) { 1399 | # 1400 | # Input: an email address, e.g. "foo@example.com" 1401 | # 1402 | # Output: the email address as a mailto link, with each character 1403 | # of the address encoded as either a decimal or hex entity, in 1404 | # the hopes of foiling most address harvesting spam bots. E.g.: 1405 | # 1406 | # foo@exampl 1409 | # e.com 1410 | # 1411 | # Based by a filter by Matthew Wickline, posted to BBEdit-Talk. 1412 | # With some optimizations by Milian Wolff. 1413 | # 1414 | $addr = "mailto:" . $addr; 1415 | $chars = preg_split('/(? $char) { 1419 | $ord = ord($char); 1420 | # Ignore non-ascii chars. 1421 | if ($ord < 128) { 1422 | $r = ($seed * (1 + $key)) % 100; # Pseudo-random function. 1423 | # roughly 10% raw, 45% hex, 45% dec 1424 | # '@' *must* be encoded. I insist. 1425 | if ($r > 90 && $char != '@') /* do nothing */; 1426 | else if ($r < 45) $chars[$key] = '&#x'.dechex($ord).';'; 1427 | else $chars[$key] = '&#'.$ord.';'; 1428 | } 1429 | } 1430 | 1431 | $addr = implode('', $chars); 1432 | $text = implode('', array_slice($chars, 7)); # text without `mailto:` 1433 | $addr = "$text"; 1434 | 1435 | return $addr; 1436 | } 1437 | 1438 | 1439 | function tokenizeHTML($str) { 1440 | # 1441 | # Parameter: String containing HTML + Markdown markup. 1442 | # Returns: An array of the tokens comprising the input 1443 | # string. Each token is either a tag or a run of text 1444 | # between tags. Each element of the array is a 1445 | # two-element array; the first is either 'tag' or 'text'; 1446 | # the second is the actual value. 1447 | # Note: Markdown code spans are taken into account: no tag token is 1448 | # generated within a code span. 1449 | # 1450 | $tokens = array(); 1451 | 1452 | while ($str != "") { 1453 | # 1454 | # Each loop iteration seach for either the next tag or the next 1455 | # openning code span marker. If a code span marker is found, the 1456 | # code span is extracted in entierty and will result in an extra 1457 | # text token. 1458 | # 1459 | $parts = preg_split('{ 1460 | ( 1461 | (? # comment 1465 | | 1466 | <\?.*?\?> | <%.*?%> # processing instruction 1467 | | 1468 | <[/!$]?[-a-zA-Z0-9:]+ # regular tags 1469 | (?: 1470 | \s 1471 | (?>[^"\'>]+|"[^"]*"|\'[^\']*\')* 1472 | )? 1473 | > 1474 | ) 1475 | }xs', $str, 2, PREG_SPLIT_DELIM_CAPTURE); 1476 | 1477 | # Create token from text preceding tag. 1478 | if ($parts[0] != "") { 1479 | $tokens[] = array('text', $parts[0]); 1480 | } 1481 | 1482 | # Check if we reach the end. 1483 | if (count($parts) < 3) { 1484 | break; 1485 | } 1486 | 1487 | # Create token from tag or code span. 1488 | if ($parts[1]{0} == "`") { 1489 | $tokens[] = array('text', $parts[1]); 1490 | $str = $parts[2]; 1491 | 1492 | # Skip the whole code span, pass as text token. 1493 | if (preg_match('/^(.*(?tab_width})/m", "", $text); 1514 | } 1515 | 1516 | 1517 | # String length function for detab. `_initDetab` will create a function to 1518 | # hanlde UTF-8 if the default function does not exist. 1519 | var $utf8_strlen = 'mb_strlen'; 1520 | 1521 | function detab($text) { 1522 | # 1523 | # Replace tabs with the appropriate amount of space. 1524 | # 1525 | # For each line we separate the line in blocks delemited by 1526 | # tab characters. Then we reconstruct every line by adding the 1527 | # appropriate number of space between each blocks. 1528 | 1529 | $strlen = $this->utf8_strlen; # strlen function for UTF-8. 1530 | $lines = explode("\n", $text); 1531 | $text = ""; 1532 | 1533 | foreach ($lines as $line) { 1534 | # Split in blocks. 1535 | $blocks = explode("\t", $line); 1536 | # Add each blocks to the line. 1537 | $line = $blocks[0]; 1538 | unset($blocks[0]); # Do not add first block twice. 1539 | foreach ($blocks as $block) { 1540 | # Calculate amount of space, insert spaces, insert block. 1541 | $amount = $this->tab_width - 1542 | $strlen($line, 'UTF-8') % $this->tab_width; 1543 | $line .= str_repeat(" ", $amount) . $block; 1544 | } 1545 | $text .= "$line\n"; 1546 | } 1547 | return $text; 1548 | } 1549 | function _initDetab() { 1550 | # 1551 | # Check for the availability of the function in the `utf8_strlen` property 1552 | # (initially `mb_strlen`). If the function is not available, create a 1553 | # function that will loosely count the number of UTF-8 characters with a 1554 | # regular expression. 1555 | # 1556 | if (function_exists($this->utf8_strlen)) return; 1557 | $this->utf8_strlen = create_function('$text', 'return preg_match_all( 1558 | "/[\\\\x00-\\\\xBF]|[\\\\xC0-\\\\xFF][\\\\x80-\\\\xBF]*/", 1559 | $text, $m);'); 1560 | } 1561 | 1562 | 1563 | function unhash($text) { 1564 | # 1565 | # Swap back in all the tags hashed by _HashHTMLBlocks. 1566 | # 1567 | return str_replace(array_keys($this->html_hashes), 1568 | array_values($this->html_hashes), $text); 1569 | } 1570 | 1571 | } 1572 | 1573 | 1574 | # 1575 | # Markdown Extra Parser Class 1576 | # 1577 | 1578 | class MarkdownExtra_Parser extends Markdown_Parser { 1579 | 1580 | # Prefix for footnote ids. 1581 | var $fn_id_prefix = ""; 1582 | 1583 | # Optional title attribute for footnote links and backlinks. 1584 | var $fn_link_title = MARKDOWN_FN_LINK_TITLE; 1585 | var $fn_backlink_title = MARKDOWN_FN_BACKLINK_TITLE; 1586 | 1587 | # Optional class attribute for footnote links and backlinks. 1588 | var $fn_link_class = MARKDOWN_FN_LINK_CLASS; 1589 | var $fn_backlink_class = MARKDOWN_FN_BACKLINK_CLASS; 1590 | 1591 | 1592 | function MarkdownExtra_Parser() { 1593 | # 1594 | # Constructor function. Initialize the parser object. 1595 | # 1596 | # Add extra escapable characters before parent constructor 1597 | # initialize the table. 1598 | $this->escape_chars .= ':|'; 1599 | 1600 | # Insert extra document, block, and span transformations. 1601 | # Parent constructor will do the sorting. 1602 | $this->document_gamut += array( 1603 | "stripFootnotes" => 15, 1604 | "stripAbbreviations" => 25, 1605 | "appendFootnotes" => 50, 1606 | ); 1607 | $this->block_gamut += array( 1608 | "doTables" => 15, 1609 | "doDefLists" => 45, 1610 | ); 1611 | $this->span_gamut += array( 1612 | "doFootnotes" => 5, 1613 | "doAbbreviations" => 70, 1614 | ); 1615 | 1616 | parent::Markdown_Parser(); 1617 | } 1618 | 1619 | 1620 | # Extra hashes used during extra transformations. 1621 | var $footnotes = array(); 1622 | var $footnotes_ordered = array(); 1623 | var $abbr_desciptions = array(); 1624 | var $abbr_matches = array(); 1625 | var $html_cleans = array(); 1626 | 1627 | # Status flag to avoid invalid nesting. 1628 | var $in_footnote = false; 1629 | 1630 | 1631 | function transform($text) { 1632 | # 1633 | # Added clear to the new $html_hashes, reordered `hashHTMLBlocks` before 1634 | # blank line stripping and added extra parameter to `runBlockGamut`. 1635 | # 1636 | # Clear the global hashes. If we don't clear these, you get conflicts 1637 | # from other articles when generating a page which contains more than 1638 | # one article (e.g. an index page that shows the N most recent 1639 | # articles): 1640 | $this->footnotes = array(); 1641 | $this->footnotes_ordered = array(); 1642 | $this->abbr_desciptions = array(); 1643 | $this->abbr_matches = array(); 1644 | $this->html_cleans = array(); 1645 | 1646 | return parent::transform($text); 1647 | } 1648 | 1649 | 1650 | ### HTML Block Parser ### 1651 | 1652 | # Tags that are always treated as block tags: 1653 | var $block_tags = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|form|fieldset|iframe|hr|legend'; 1654 | 1655 | # Tags treated as block tags only if the opening tag is alone on it's line: 1656 | var $context_block_tags = 'script|noscript|math|ins|del'; 1657 | 1658 | # Tags where markdown="1" default to span mode: 1659 | var $contain_span_tags = 'p|h[1-6]|li|dd|dt|td|th|legend|address'; 1660 | 1661 | # Tags which must not have their contents modified, no matter where 1662 | # they appear: 1663 | var $clean_tags = 'script|math'; 1664 | 1665 | # Tags that do not need to be closed. 1666 | var $auto_close_tags = 'hr|img'; 1667 | 1668 | 1669 | function hashHTMLBlocks($text) { 1670 | # 1671 | # Hashify HTML Blocks and "clean tags". 1672 | # 1673 | # We only want to do this for block-level HTML tags, such as headers, 1674 | # lists, and tables. That's because we still want to wrap s around 1675 | # "paragraphs" that are wrapped in non-block-level tags, such as anchors, 1676 | # phrase emphasis, and spans. The list of tags we're looking for is 1677 | # hard-coded. 1678 | # 1679 | # This works by calling _HashHTMLBlocks_InMarkdown, which then calls 1680 | # _HashHTMLBlocks_InHTML when it encounter block tags. When the markdown="1" 1681 | # attribute is found whitin a tag, _HashHTMLBlocks_InHTML calls back 1682 | # _HashHTMLBlocks_InMarkdown to handle the Markdown syntax within the tag. 1683 | # These two functions are calling each other. It's recursive! 1684 | # 1685 | # 1686 | # Call the HTML-in-Markdown hasher. 1687 | # 1688 | list($text, ) = $this->_hashHTMLBlocks_inMarkdown($text); 1689 | 1690 | return $text; 1691 | } 1692 | function _hashHTMLBlocks_inMarkdown($text, $indent = 0, 1693 | $enclosing_tag = '', $span = false) 1694 | { 1695 | # 1696 | # Parse markdown text, calling _HashHTMLBlocks_InHTML for block tags. 1697 | # 1698 | # * $indent is the number of space to be ignored when checking for code 1699 | # blocks. This is important because if we don't take the indent into 1700 | # account, something like this (which looks right) won't work as expected: 1701 | # 1702 | # 1703 | # 1704 | # Hello World. <-- Is this a Markdown code block or text? 1705 | # <-- Is this a Markdown code block or a real tag? 1706 | # 1707 | # 1708 | # If you don't like this, just don't indent the tag on which 1709 | # you apply the markdown="1" attribute. 1710 | # 1711 | # * If $enclosing_tag is not empty, stops at the first unmatched closing 1712 | # tag with that name. Nested tags supported. 1713 | # 1714 | # * If $span is true, text inside must treated as span. So any double 1715 | # newline will be replaced by a single newline so that it does not create 1716 | # paragraphs. 1717 | # 1718 | # Returns an array of that form: ( processed text , remaining text ) 1719 | # 1720 | if ($text === '') return array('', ''); 1721 | 1722 | # Regex to check for the presense of newlines around a block tag. 1723 | $newline_match_before = '/(?:^\n?|\n\n)*$/'; 1724 | $newline_match_after = 1725 | '{ 1726 | ^ # Start of text following the tag. 1727 | (?:[ ]*)? # Optional comment. 1728 | [ ]*\n # Must be followed by newline. 1729 | }xs'; 1730 | 1731 | # Regex to match any tag. 1732 | $block_tag_match = 1733 | '{ 1734 | ( # $2: Capture hole tag. 1735 | block_tags.' | 1738 | '.$this->context_block_tags.' | 1739 | '.$this->clean_tags.' | 1740 | (?!\s)'.$enclosing_tag.' 1741 | ) 1742 | \s* # Whitespace. 1743 | (?> 1744 | ".*?" | # Double quotes (can contain `>`) 1745 | \'.*?\' | # Single quotes (can contain `>`) 1746 | .+? # Anything but quotes and `>`. 1747 | )*? 1748 | > # End of tag. 1749 | | 1750 | # HTML Comment 1751 | | 1752 | <\?.*?\?> | <%.*?%> # Processing instruction 1753 | | 1754 | # CData Block 1755 | ) 1756 | }xs'; 1757 | 1758 | 1759 | $depth = 0; # Current depth inside the tag tree. 1760 | $parsed = ""; # Parsed text that will be returned. 1761 | 1762 | # 1763 | # Loop through every tag until we find the closing tag of the parent 1764 | # or loop until reaching the end of text if no parent tag specified. 1765 | # 1766 | do { 1767 | # 1768 | # Split the text using the first $tag_match pattern found. 1769 | # Text before pattern will be first in the array, text after 1770 | # pattern will be at the end, and between will be any catches made 1771 | # by the pattern. 1772 | # 1773 | $parts = preg_split($block_tag_match, $text, 2, 1774 | PREG_SPLIT_DELIM_CAPTURE); 1775 | 1776 | # If in Markdown span mode, add a empty-string span-level hash 1777 | # after each newline to prevent triggering any block element. 1778 | if ($span) { 1779 | $void = $this->hashSpan("", true) ; 1780 | $newline = $this->hashSpan("", true) . "\n"; 1781 | $parts[0] = $void . str_replace("\n", $newline, $parts[0]) . $void; 1782 | } 1783 | 1784 | $parsed .= $parts[0]; # Text before current tag. 1785 | 1786 | # If end of $text has been reached. Stop loop. 1787 | if (count($parts) < 3) { 1788 | $text = ""; 1789 | break; 1790 | } 1791 | 1792 | $tag = $parts[1]; # Tag to handle. 1793 | $text = $parts[2]; # Remaining text after current tag. 1794 | 1795 | # 1796 | # Check for: Tag inside code block or span 1797 | # 1798 | if (# Find current paragraph 1799 | preg_match('/(?>^\n?|\n\n)((?>.\n?)+?)$/', $parsed, $matches) && 1800 | ( 1801 | # Then match in it either a code block... 1802 | preg_match('/^ {'.($indent+4).'}.*(?>\n {'.($indent+4).'}.*)*'. 1803 | '(?!\n)$/', $matches[1], $x) || 1804 | # ...or unbalenced code span markers. (the regex matches balenced) 1805 | !preg_match('/^(?>[^`]+|(`+)(?>[^`]+|(?!\1[^`])`)*?\1(?!`))*$/s', 1806 | $matches[1]) 1807 | )) 1808 | { 1809 | # Tag is in code block or span and may not be a tag at all. So we 1810 | # simply skip the first char (should be a `<`). 1811 | $parsed .= $tag{0}; 1812 | $text = substr($tag, 1) . $text; # Put back $tag minus first char. 1813 | } 1814 | # 1815 | # Check for: Opening Block level tag or 1816 | # Opening Content Block tag (like ins and del) 1817 | # used as a block tag (tag is alone on it's line). 1818 | # 1819 | else if (preg_match("{^<(?:$this->block_tags)\b}", $tag) || 1820 | ( preg_match("{^<(?:$this->context_block_tags)\b}", $tag) && 1821 | preg_match($newline_match_before, $parsed) && 1822 | preg_match($newline_match_after, $text) ) 1823 | ) 1824 | { 1825 | # Need to parse tag and following text using the HTML parser. 1826 | list($block_text, $text) = 1827 | $this->_hashHTMLBlocks_inHTML($tag . $text, "hashBlock", true); 1828 | 1829 | # Make sure it stays outside of any paragraph by adding newlines. 1830 | $parsed .= "\n\n$block_text\n\n"; 1831 | } 1832 | # 1833 | # Check for: Clean tag (like script, math) 1834 | # HTML Comments, processing instructions. 1835 | # 1836 | else if (preg_match("{^<(?:$this->clean_tags)\b}", $tag) || 1837 | $tag{1} == '!' || $tag{1} == '?') 1838 | { 1839 | # Need to parse tag and following text using the HTML parser. 1840 | # (don't check for markdown attribute) 1841 | list($block_text, $text) = 1842 | $this->_hashHTMLBlocks_inHTML($tag . $text, "hashClean", false); 1843 | 1844 | $parsed .= $block_text; 1845 | } 1846 | # 1847 | # Check for: Tag with same name as enclosing tag. 1848 | # 1849 | else if ($enclosing_tag !== '' && 1850 | # Same name as enclosing tag. 1851 | preg_match("{^= 0); 1874 | 1875 | return array($parsed, $text); 1876 | } 1877 | function _hashHTMLBlocks_inHTML($text, $hash_method, $md_attr) { 1878 | # 1879 | # Parse HTML, calling _HashHTMLBlocks_InMarkdown for block tags. 1880 | # 1881 | # * Calls $hash_method to convert any blocks. 1882 | # * Stops when the first opening tag closes. 1883 | # * $md_attr indicate if the use of the `markdown="1"` attribute is allowed. 1884 | # (it is not inside clean tags) 1885 | # 1886 | # Returns an array of that form: ( processed text , remaining text ) 1887 | # 1888 | if ($text === '') return array('', ''); 1889 | 1890 | # Regex to match `markdown` attribute inside of a tag. 1891 | $markdown_attr_match = ' 1892 | { 1893 | \s* # Eat whitespace before the `markdown` attribute 1894 | markdown 1895 | \s*=\s* 1896 | (?: 1897 | (["\']) # $1: quote delimiter 1898 | (.*?) # $2: attribute value 1899 | \1 # matching delimiter 1900 | | 1901 | ([^\s>]*) # $3: unquoted attribute value 1902 | ) 1903 | () # $4: make $3 always defined (avoid warnings) 1904 | }xs'; 1905 | 1906 | # Regex to match any tag. 1907 | $tag_match = '{ 1908 | ( # $2: Capture hole tag. 1909 | 1913 | ".*?" | # Double quotes (can contain `>`) 1914 | \'.*?\' | # Single quotes (can contain `>`) 1915 | .+? # Anything but quotes and `>`. 1916 | )*? 1917 | > # End of tag. 1918 | | 1919 | # HTML Comment 1920 | | 1921 | <\?.*?\?> | <%.*?%> # Processing instruction 1922 | | 1923 | # CData Block 1924 | ) 1925 | }xs'; 1926 | 1927 | $original_text = $text; # Save original text in case of faliure. 1928 | 1929 | $depth = 0; # Current depth inside the tag tree. 1930 | $block_text = ""; # Temporary text holder for current text. 1931 | $parsed = ""; # Parsed text that will be returned. 1932 | 1933 | # 1934 | # Get the name of the starting tag. 1935 | # 1936 | if (preg_match("/^<([\w:$]*)\b/", $text, $matches)) 1937 | $base_tag_name = $matches[1]; 1938 | 1939 | # 1940 | # Loop through every tag until we find the corresponding closing tag. 1941 | # 1942 | do { 1943 | # 1944 | # Split the text using the first $tag_match pattern found. 1945 | # Text before pattern will be first in the array, text after 1946 | # pattern will be at the end, and between will be any catches made 1947 | # by the pattern. 1948 | # 1949 | $parts = preg_split($tag_match, $text, 2, PREG_SPLIT_DELIM_CAPTURE); 1950 | 1951 | if (count($parts) < 3) { 1952 | # 1953 | # End of $text reached with unbalenced tag(s). 1954 | # In that case, we return original text unchanged and pass the 1955 | # first character as filtered to prevent an infinite loop in the 1956 | # parent function. 1957 | # 1958 | return array($original_text{0}, substr($original_text, 1)); 1959 | } 1960 | 1961 | $block_text .= $parts[0]; # Text before current tag. 1962 | $tag = $parts[1]; # Tag to handle. 1963 | $text = $parts[2]; # Remaining text after current tag. 1964 | 1965 | # 1966 | # Check for: Auto-close tag (like ) 1967 | # Comments and Processing Instructions. 1968 | # 1969 | if (preg_match("{^auto_close_tags)\b}", $tag) || 1970 | $tag{1} == '!' || $tag{1} == '?') 1971 | { 1972 | # Just add the tag to the block as if it was text. 1973 | $block_text .= $tag; 1974 | } 1975 | else { 1976 | # 1977 | # Increase/decrease nested tag count. Only do so if 1978 | # the tag's name match base tag's. 1979 | # 1980 | if (preg_match("{^mode = $attr_m[2] . $attr_m[3]; 1997 | $span_mode = $this->mode == 'span' || $this->mode != 'block' && 1998 | preg_match("{^<(?:$this->contain_span_tags)\b}", $tag); 1999 | 2000 | # Calculate indent before tag. 2001 | preg_match('/(?:^|\n)( *?)(?! ).*?$/', $block_text, $matches); 2002 | $indent = strlen($matches[1]); 2003 | 2004 | # End preceding block with this tag. 2005 | $block_text .= $tag; 2006 | $parsed .= $this->$hash_method($block_text); 2007 | 2008 | # Get enclosing tag name for the ParseMarkdown function. 2009 | preg_match('/^<([\w:$]*)\b/', $tag, $matches); 2010 | $tag_name = $matches[1]; 2011 | 2012 | # Parse the content using the HTML-in-Markdown parser. 2013 | list ($block_text, $text) 2014 | = $this->_hashHTMLBlocks_inMarkdown($text, $indent, 2015 | $tag_name, $span_mode); 2016 | 2017 | # Outdent markdown text. 2018 | if ($indent > 0) { 2019 | $block_text = preg_replace("/^[ ]{1,$indent}/m", "", 2020 | $block_text); 2021 | } 2022 | 2023 | # Append tag content to parsed text. 2024 | if (!$span_mode) $parsed .= "\n\n$block_text\n\n"; 2025 | else $parsed .= "$block_text"; 2026 | 2027 | # Start over a new block. 2028 | $block_text = ""; 2029 | } 2030 | else $block_text .= $tag; 2031 | } 2032 | 2033 | } while ($depth > 0); 2034 | 2035 | # 2036 | # Hash last block text that wasn't processed inside the loop. 2037 | # 2038 | $parsed .= $this->$hash_method($block_text); 2039 | 2040 | return array($parsed, $text); 2041 | } 2042 | 2043 | 2044 | function hashClean($text) { 2045 | # 2046 | # Called whenever a tag must be hashed when a function insert a "clean" tag 2047 | # in $text, it pass through this function and is automaticaly escaped, 2048 | # blocking invalid nested overlap. 2049 | # 2050 | # Swap back any tag hash found in $text so we do not have to `unhash` 2051 | # multiple times at the end. 2052 | $text = $this->unhash($text); 2053 | 2054 | # Then hash the tag. 2055 | $key = "C\x1A". md5($text); 2056 | $this->html_cleans[$key] = $text; 2057 | $this->html_hashes[$key] = $text; 2058 | return $key; # String that will replace the clean tag. 2059 | } 2060 | 2061 | 2062 | function doHeaders($text) { 2063 | # 2064 | # Redefined to add id attribute support. 2065 | # 2066 | # Setext-style headers: 2067 | # Header 1 {#header1} 2068 | # ======== 2069 | # 2070 | # Header 2 {#header2} 2071 | # -------- 2072 | # 2073 | $text = preg_replace_callback( 2074 | '{ (^.+?) (?:[ ]+\{\#([-_:a-zA-Z0-9]+)\})? [ ]*\n=+[ ]*\n+ }mx', 2075 | array(&$this, '_doHeaders_callback_setext_h1'), $text); 2076 | $text = preg_replace_callback( 2077 | '{ (^.+?) (?:[ ]+\{\#([-_:a-zA-Z0-9]+)\})? [ ]*\n-+[ ]*\n+ }mx', 2078 | array(&$this, '_doHeaders_callback_setext_h2'), $text); 2079 | 2080 | # atx-style headers: 2081 | # # Header 1 {#header1} 2082 | # ## Header 2 {#header2} 2083 | # ## Header 2 with closing hashes ## {#header3} 2084 | # ... 2085 | # ###### Header 6 {#header2} 2086 | # 2087 | $text = preg_replace_callback('{ 2088 | ^(\#{1,6}) # $1 = string of #\'s 2089 | [ ]* 2090 | (.+?) # $2 = Header text 2091 | [ ]* 2092 | \#* # optional closing #\'s (not counted) 2093 | (?:[ ]+\{\#([-_:a-zA-Z0-9]+)\})? # id attribute 2094 | [ ]* 2095 | \n+ 2096 | }xm', 2097 | array(&$this, '_doHeaders_callback_atx'), $text); 2098 | 2099 | return $text; 2100 | } 2101 | function _doHeaders_attr($attr) { 2102 | if (empty($attr)) return ""; 2103 | return " id=\"$attr\""; 2104 | } 2105 | function _doHeaders_callback_setext_h1($matches) { 2106 | $attr = $this->_doHeaders_attr($id =& $matches[2]); 2107 | $block = "".$this->runSpanGamut($matches[1]).""; 2108 | return "\n" . $this->hashBlock($block) . "\n\n"; 2109 | } 2110 | function _doHeaders_callback_setext_h2($matches) { 2111 | $attr = $this->_doHeaders_attr($id =& $matches[2]); 2112 | $block = "".$this->runSpanGamut($matches[1]).""; 2113 | return "\n" . $this->hashBlock($block) . "\n\n"; 2114 | } 2115 | function _doHeaders_callback_atx($matches) { 2116 | $level = strlen($matches[1]); 2117 | $attr = $this->_doHeaders_attr($id =& $matches[3]); 2118 | $block = "".$this->runSpanGamut($matches[2]).""; 2119 | return "\n" . $this->hashBlock($block) . "\n\n"; 2120 | } 2121 | 2122 | 2123 | function doTables($text) { 2124 | # 2125 | # Form HTML tables. 2126 | # 2127 | $less_than_tab = $this->tab_width - 1; 2128 | # 2129 | # Find tables with leading pipe. 2130 | # 2131 | # | Header 1 | Header 2 2132 | # | -------- | -------- 2133 | # | Cell 1 | Cell 2 2134 | # | Cell 3 | Cell 4 2135 | # 2136 | $text = preg_replace_callback(' 2137 | { 2138 | ^ # Start of a line 2139 | [ ]{0,'.$less_than_tab.'} # Allowed whitespace. 2140 | [|] # Optional leading pipe (present) 2141 | (.+) \n # $1: Header row (at least one pipe) 2142 | 2143 | [ ]{0,'.$less_than_tab.'} # Allowed whitespace. 2144 | [|] ([ ]*[-:]+[-| :]*) \n # $2: Header underline 2145 | 2146 | ( # $3: Cells 2147 | (?: 2148 | [ ]* # Allowed whitespace. 2149 | [|] .* \n # Row content. 2150 | )* 2151 | ) 2152 | (?=\n|\Z) # Stop at final double newline. 2153 | }xm', 2154 | array(&$this, '_doTable_leadingPipe_callback'), $text); 2155 | 2156 | # 2157 | # Find tables without leading pipe. 2158 | # 2159 | # Header 1 | Header 2 2160 | # -------- | -------- 2161 | # Cell 1 | Cell 2 2162 | # Cell 3 | Cell 4 2163 | # 2164 | $text = preg_replace_callback(' 2165 | { 2166 | ^ # Start of a line 2167 | [ ]{0,'.$less_than_tab.'} # Allowed whitespace. 2168 | (\S.*[|].*) \n # $1: Header row (at least one pipe) 2169 | 2170 | [ ]{0,'.$less_than_tab.'} # Allowed whitespace. 2171 | ([-:]+[ ]*[|][-| :]*) \n # $2: Header underline 2172 | 2173 | ( # $3: Cells 2174 | (?: 2175 | .* [|] .* \n # Row content 2176 | )* 2177 | ) 2178 | (?=\n|\Z) # Stop at final double newline. 2179 | }xm', 2180 | array(&$this, '_DoTable_callback'), $text); 2181 | 2182 | return $text; 2183 | } 2184 | function _doTable_leadingPipe_callback($matches) { 2185 | $head = $matches[1]; 2186 | $underline = $matches[2]; 2187 | $content = $matches[3]; 2188 | 2189 | # Remove leading pipe for each row. 2190 | $content = preg_replace('/^ *[|]/m', '', $content); 2191 | 2192 | return $this->_doTable_callback(array($matches[0], $head, $underline, $content)); 2193 | } 2194 | function _doTable_callback($matches) { 2195 | $head = $matches[1]; 2196 | $underline = $matches[2]; 2197 | $content = $matches[3]; 2198 | 2199 | # Remove any tailing pipes for each line. 2200 | $head = preg_replace('/[|] *$/m', '', $head); 2201 | $underline = preg_replace('/[|] *$/m', '', $underline); 2202 | $content = preg_replace('/[|] *$/m', '', $content); 2203 | 2204 | # Reading alignement from header underline. 2205 | $separators = preg_split('/ *[|] */', $underline); 2206 | foreach ($separators as $n => $s) { 2207 | if (preg_match('/^ *-+: *$/', $s)) $attr[$n] = ' align="right"'; 2208 | else if (preg_match('/^ *:-+: *$/', $s))$attr[$n] = ' align="center"'; 2209 | else if (preg_match('/^ *:-+ *$/', $s)) $attr[$n] = ' align="left"'; 2210 | else $attr[$n] = ''; 2211 | } 2212 | 2213 | # Creating code spans before splitting the row is an easy way to 2214 | # handle a code span containg pipes. 2215 | $head = $this->doCodeSpans($head); 2216 | $headers = preg_split('/ *[|] */', $head); 2217 | $col_count = count($headers); 2218 | 2219 | # Write column headers. 2220 | $text = "\n"; 2221 | $text .= "\n"; 2222 | $text .= "\n"; 2223 | foreach ($headers as $n => $header) 2224 | $text .= " ".$this->runSpanGamut(trim($header))."\n"; 2225 | $text .= "\n"; 2226 | $text .= "\n"; 2227 | 2228 | # Split content by row. 2229 | $rows = explode("\n", trim($content, "\n")); 2230 | 2231 | $text .= "\n"; 2232 | foreach ($rows as $row) { 2233 | # Creating code spans before splitting the row is an easy way to 2234 | # handle a code span containg pipes. 2235 | $row = $this->doCodeSpans($row); 2236 | 2237 | # Split row by cell. 2238 | $row_cells = preg_split('/ *[|] */', $row, $col_count); 2239 | $row_cells = array_pad($row_cells, $col_count, ''); 2240 | 2241 | $text .= "\n"; 2242 | foreach ($row_cells as $n => $cell) 2243 | $text .= " ".$this->runSpanGamut(trim($cell))."\n"; 2244 | $text .= "\n"; 2245 | } 2246 | $text .= "\n"; 2247 | $text .= " "; 2248 | 2249 | return $this->hashBlock($text) . "\n"; 2250 | } 2251 | 2252 | 2253 | function doDefLists($text) { 2254 | # 2255 | # Form HTML definition lists. 2256 | # 2257 | $less_than_tab = $this->tab_width - 1; 2258 | 2259 | # Re-usable pattern to match any entire dl list: 2260 | $whole_list = ' 2261 | ( # $1 = whole list 2262 | ( # $2 2263 | [ ]{0,'.$less_than_tab.'} 2264 | ((?>.*\S.*\n)+) # $3 = defined term 2265 | \n? 2266 | [ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition 2267 | ) 2268 | (?s:.+?) 2269 | ( # $4 2270 | \z 2271 | | 2272 | \n{2,} 2273 | (?=\S) 2274 | (?! # Negative lookahead for another term 2275 | [ ]{0,'.$less_than_tab.'} 2276 | (?: \S.*\n )+? # defined term 2277 | \n? 2278 | [ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition 2279 | ) 2280 | (?! # Negative lookahead for another definition 2281 | [ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition 2282 | ) 2283 | ) 2284 | ) 2285 | '; // mx 2286 | 2287 | $text = preg_replace_callback('{ 2288 | (?:(?<=\n\n)|\A\n?) 2289 | '.$whole_list.' 2290 | }mx', 2291 | array(&$this, '_doDefLists_callback'), $text); 2292 | 2293 | return $text; 2294 | } 2295 | function _doDefLists_callback($matches) { 2296 | # Re-usable patterns to match list item bullets and number markers: 2297 | $list = $matches[1]; 2298 | 2299 | # Turn double returns into triple returns, so that we can make a 2300 | # paragraph for the last item in a list, if necessary: 2301 | $result = trim($this->processDefListItems($list)); 2302 | $result = "\n" . $result . "\n"; 2303 | return $this->hashBlock($result) . "\n\n"; 2304 | } 2305 | 2306 | 2307 | function processDefListItems($list_str) { 2308 | # 2309 | # Process the contents of a single definition list, splitting it 2310 | # into individual term and definition list items. 2311 | # 2312 | $less_than_tab = $this->tab_width - 1; 2313 | 2314 | # trim trailing blank lines: 2315 | $list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str); 2316 | 2317 | # Process definition terms. 2318 | $list_str = preg_replace_callback('{ 2319 | (?:\n\n+|\A\n?) # leading line 2320 | ( # definition terms = $1 2321 | [ ]{0,'.$less_than_tab.'} # leading whitespace 2322 | (?![:][ ]|[ ]) # negative lookahead for a definition 2323 | # mark (colon) or more whitespace. 2324 | (?: \S.* \n)+? # actual term (not whitespace). 2325 | ) 2326 | (?=\n?[ ]{0,3}:[ ]) # lookahead for following line feed 2327 | # with a definition mark. 2328 | }xm', 2329 | array(&$this, '_processDefListItems_callback_dt'), $list_str); 2330 | 2331 | # Process actual definitions. 2332 | $list_str = preg_replace_callback('{ 2333 | \n(\n+)? # leading line = $1 2334 | [ ]{0,'.$less_than_tab.'} # whitespace before colon 2335 | [:][ ]+ # definition mark (colon) 2336 | ((?s:.+?)) # definition text = $2 2337 | (?= \n+ # stop at next definition mark, 2338 | (?: # next term or end of text 2339 | [ ]{0,'.$less_than_tab.'} [:][ ] | 2340 | | \z 2341 | ) 2342 | ) 2343 | }xm', 2344 | array(&$this, '_processDefListItems_callback_dd'), $list_str); 2345 | 2346 | return $list_str; 2347 | } 2348 | function _processDefListItems_callback_dt($matches) { 2349 | $terms = explode("\n", trim($matches[1])); 2350 | $text = ''; 2351 | foreach ($terms as $term) { 2352 | $term = $this->runSpanGamut(trim($term)); 2353 | $text .= "\n " . $term . ""; 2354 | } 2355 | return $text . "\n"; 2356 | } 2357 | function _processDefListItems_callback_dd($matches) { 2358 | $leading_line = $matches[1]; 2359 | $def = $matches[2]; 2360 | 2361 | if ($leading_line || preg_match('/\n{2,}/', $def)) { 2362 | $def = $this->runBlockGamut($this->outdent($def . "\n\n")); 2363 | $def = "\n". $def ."\n"; 2364 | } 2365 | else { 2366 | $def = rtrim($def); 2367 | $def = $this->runSpanGamut($this->outdent($def)); 2368 | } 2369 | 2370 | return "\n" . $def . "\n"; 2371 | } 2372 | 2373 | 2374 | function doItalicsAndBold($text) { 2375 | # 2376 | # Redefined to change emphasis by underscore behaviour so that it does not 2377 | # work in the middle of a word. 2378 | # 2379 | # must go first: 2380 | $text = preg_replace_callback(array( 2381 | '{ 2382 | ( # $1: Marker 2383 | (? 2391 | [^_]+? # Anthing not em markers. 2392 | | 2393 | # Balence any regular _ emphasis inside. 2394 | (? 2410 | [^*]+? # Anthing not em markers. 2411 | | 2412 | # Balence any regular * emphasis inside. 2413 | \* (?=\S) (.+?) (?<=\S) \* 2414 | | 2415 | \* # Allow unbalenced as last resort. 2416 | )+? 2417 | ) 2418 | (?<=\S) \*\* # End mark not preceded by whitespace. 2419 | }sx', 2420 | ), 2421 | array(&$this, '_doItalicAndBold_strong_callback'), $text); 2422 | # Then : 2423 | $text = preg_replace_callback(array( 2424 | '{ ( (? tags 2437 | # 2438 | # Strip leading and trailing lines: 2439 | $text = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $text); 2440 | 2441 | $grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY); 2442 | 2443 | # 2444 | # Wrap tags and unhashify HTML blocks 2445 | # 2446 | foreach ($grafs as $key => $value) { 2447 | $value = trim($this->runSpanGamut($value)); 2448 | 2449 | # Check if this should be enclosed in a paragraph. 2450 | # Clean tag hashes & block tag hashes are left alone. 2451 | $clean_key = $value; 2452 | $block_key = substr($value, 0, 34); 2453 | 2454 | $is_p = (!isset($this->html_blocks[$block_key]) && 2455 | !isset($this->html_cleans[$clean_key])); 2456 | 2457 | if ($is_p) { 2458 | $value = " $value"; 2459 | } 2460 | $grafs[$key] = $value; 2461 | } 2462 | 2463 | # Join grafs in one text, then unhash HTML tags. 2464 | $text = implode("\n\n", $grafs); 2465 | 2466 | # Finish by removing any tag hashes still present in $text. 2467 | $text = $this->unhash($text); 2468 | 2469 | return $text; 2470 | } 2471 | 2472 | 2473 | ### Footnotes 2474 | 2475 | function stripFootnotes($text) { 2476 | # 2477 | # Strips link definitions from text, stores the URLs and titles in 2478 | # hash references. 2479 | # 2480 | $less_than_tab = $this->tab_width - 1; 2481 | 2482 | # Link defs are in the form: [^id]: url "optional title" 2483 | $text = preg_replace_callback('{ 2484 | ^[ ]{0,'.$less_than_tab.'}\[\^(.+?)\][ ]?: # note_id = $1 2485 | [ ]* 2486 | \n? # maybe *one* newline 2487 | ( # text = $2 (no blank lines allowed) 2488 | (?: 2489 | .+ # actual text 2490 | | 2491 | \n # newlines but 2492 | (?!\[\^.+?\]:\s)# negative lookahead for footnote marker. 2493 | (?!\n+[ ]{0,3}\S)# ensure line is not blank and followed 2494 | # by non-indented content 2495 | )* 2496 | ) 2497 | }xm', 2498 | array(&$this, '_stripFootnotes_callback'), 2499 | $text); 2500 | return $text; 2501 | } 2502 | function _stripFootnotes_callback($matches) { 2503 | $note_id = $this->fn_id_prefix . $matches[1]; 2504 | $this->footnotes[$note_id] = $this->outdent($matches[2]); 2505 | return ''; # String that will replace the block 2506 | } 2507 | 2508 | 2509 | function doFootnotes($text) { 2510 | # 2511 | # Replace footnote references in $text [^id] with a special text-token 2512 | # which will be can be 2513 | # 2514 | if (!$this->in_footnote && !$this->in_anchor) { 2515 | $text = preg_replace('{\[\^(.+?)\]}', "F\x1Afn:\\1\x1A:", $text); 2516 | } 2517 | return $text; 2518 | } 2519 | 2520 | 2521 | function appendFootnotes($text) { 2522 | # 2523 | # Append footnote list to text. 2524 | # 2525 | 2526 | $text = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}', 2527 | array(&$this, '_appendFootnotes_callback'), $text); 2528 | 2529 | if (!empty($this->footnotes_ordered)) { 2530 | $text .= "\n\n"; 2531 | $text .= "\n"; 2532 | $text .= "fn_backlink_class != "") { 2537 | $class = $this->fn_backlink_class; 2538 | $class = $this->encodeAmpsAndAngles($class); 2539 | $class = str_replace('"', '"', $class); 2540 | $attr .= " class=\"$class\""; 2541 | } 2542 | if ($this->fn_backlink_title != "") { 2543 | $title = $this->fn_backlink_title; 2544 | $title = $this->encodeAmpsAndAngles($title); 2545 | $title = str_replace('"', '"', $title); 2546 | $attr .= " title=\"$title\""; 2547 | } 2548 | $num = 0; 2549 | 2550 | $this->in_footnote = true; 2551 | 2552 | foreach ($this->footnotes_ordered as $note_id => $footnote) { 2553 | $footnote .= "\n"; # Need to append newline before parsing. 2554 | $footnote = $this->runBlockGamut("$footnote\n"); 2555 | 2556 | $attr2 = str_replace("%%", ++$num, $attr); 2557 | 2558 | # Add backlink to last paragraph; create new paragraph if needed. 2559 | $backlink = "↩"; 2560 | if (preg_match('{$}', $footnote)) { 2561 | $footnote = substr($footnote, 0, -4) . " $backlink"; 2562 | } else { 2563 | $footnote .= "\n\n$backlink"; 2564 | } 2565 | 2566 | $text .= "\n"; 2567 | $text .= $footnote . "\n"; 2568 | $text .= "\n\n"; 2569 | } 2570 | 2571 | $this->in_footnote = false; 2572 | 2573 | $text .= "\n"; 2574 | $text .= ""; 2575 | } 2576 | return $text; 2577 | } 2578 | function _appendFootnotes_callback($matches) { 2579 | $node_id = $this->fn_id_prefix . $matches[1]; 2580 | 2581 | # Create footnote marker only if it has a corresponding footnote *and* 2582 | # the footnote hasn't been used by another marker. 2583 | if (isset($this->footnotes[$node_id])) { 2584 | # Transfert footnote content to the ordered list. 2585 | $this->footnotes_ordered[$node_id] = $this->footnotes[$node_id]; 2586 | unset($this->footnotes[$node_id]); 2587 | 2588 | $num = count($this->footnotes_ordered); 2589 | $attr = " rel=\"footnote\""; 2590 | if ($this->fn_link_class != "") { 2591 | $class = $this->fn_link_class; 2592 | $class = $this->encodeAmpsAndAngles($class); 2593 | $class = str_replace('"', '"', $class); 2594 | $attr .= " class=\"$class\""; 2595 | } 2596 | if ($this->fn_link_title != "") { 2597 | $title = $this->fn_link_title; 2598 | $title = $this->encodeAmpsAndAngles($title); 2599 | $title = str_replace('"', '"', $title); 2600 | $attr .= " title=\"$title\""; 2601 | } 2602 | $attr = str_replace("%%", $num, $attr); 2603 | 2604 | return 2605 | "^{". 2606 | "$num". 2607 | "}"; 2608 | } 2609 | 2610 | return "[^".$matches[1]."]"; 2611 | } 2612 | 2613 | 2614 | ### Abbreviations ### 2615 | 2616 | function stripAbbreviations($text) { 2617 | # 2618 | # Strips abbreviations from text, stores titles in hash references. 2619 | # 2620 | $less_than_tab = $this->tab_width - 1; 2621 | 2622 | # Link defs are in the form: [id]*: url "optional title" 2623 | $text = preg_replace_callback('{ 2624 | ^[ ]{0,'.$less_than_tab.'}\*\[(.+?)\][ ]?: # abbr_id = $1 2625 | (.*) # text = $2 (no blank lines allowed) 2626 | }xm', 2627 | array(&$this, '_stripAbbreviations_callback'), 2628 | $text); 2629 | return $text; 2630 | } 2631 | function _stripAbbreviations_callback($matches) { 2632 | $abbr_word = $matches[1]; 2633 | $abbr_desc = $matches[2]; 2634 | $this->abbr_matches[] = preg_quote($abbr_word); 2635 | $this->abbr_desciptions[$abbr_word] = trim($abbr_desc); 2636 | return ''; # String that will replace the block 2637 | } 2638 | 2639 | 2640 | function doAbbreviations($text) { 2641 | # 2642 | # Find defined abbreviations in text and wrap them in elements. 2643 | # 2644 | if ($this->abbr_matches) { 2645 | // cannot use the /x modifier because abbr_matches may 2646 | // contain spaces: 2647 | $text = preg_replace_callback('{'. 2648 | '(?abbr_matches) .')'. 2650 | '(?![\w\x1A])'. 2651 | '}', 2652 | array(&$this, '_doAbbreviations_callback'), $text); 2653 | } 2654 | return $text; 2655 | } 2656 | function _doAbbreviations_callback($matches) { 2657 | $abbr = $matches[0]; 2658 | if (isset($this->abbr_desciptions[$abbr])) { 2659 | $desc = $this->abbr_desciptions[$abbr]; 2660 | if (empty($desc)) { 2661 | return $this->hashSpan("$abbr"); 2662 | } else { 2663 | $desc = $this->escapeSpecialCharsWithinTagAttributes($desc); 2664 | return $this->hashSpan("$abbr"); 2665 | } 2666 | } else { 2667 | return $matches[0]; 2668 | } 2669 | } 2670 | 2671 | } 2672 | 2673 | 2674 | /* 2675 | 2676 | PHP Markdown Extra 2677 | ================== 2678 | 2679 | Description 2680 | ----------- 2681 | 2682 | This is a PHP port of the original Markdown formatter written in Perl 2683 | by John Gruber. This special "Extra" version of PHP Markdown features 2684 | further enhancements to the syntax for making additional constructs 2685 | such as tables and definition list. 2686 | 2687 | Markdown is a text-to-HTML filter; it translates an easy-to-read / 2688 | easy-to-write structured text format into HTML. Markdown's text format 2689 | is most similar to that of plain text email, and supports features such 2690 | as headers, *emphasis*, code blocks, blockquotes, and links. 2691 | 2692 | Markdown's syntax is designed not as a generic markup language, but 2693 | specifically to serve as a front-end to (X)HTML. You can use span-level 2694 | HTML tags anywhere in a Markdown document, and you can use block level 2695 | HTML tags (like and as well). 2696 | 2697 | For more information about Markdown's syntax, see: 2698 | 2699 | 2700 | 2701 | 2702 | Bugs 2703 | ---- 2704 | 2705 | To file bug reports please send email to: 2706 | 2707 | 2708 | 2709 | Please include with your report: (1) the example input; (2) the output you 2710 | expected; (3) the output Markdown actually produced. 2711 | 2712 | 2713 | Version History 2714 | --------------- 2715 | 2716 | See Readme file for details. 2717 | 2718 | Extra 1.1.3 (3 Jul 2007): 2719 | 2720 | Extra 1.1.2 (7 Feb 2007) 2721 | 2722 | Extra 1.1.1 (28 Dec 2006) 2723 | 2724 | Extra 1.1 (1 Dec 2006) 2725 | 2726 | Extra 1.0.1 (9 Dec 2005) 2727 | 2728 | Extra 1.0 (5 Sep 2005) 2729 | 2730 | 2731 | Copyright and License 2732 | --------------------- 2733 | 2734 | PHP Markdown & Extra 2735 | Copyright (c) 2004-2007 Michel Fortin 2736 | 2737 | All rights reserved. 2738 | 2739 | Based on Markdown 2740 | Copyright (c) 2003-2006 John Gruber 2741 | 2742 | All rights reserved. 2743 | 2744 | Redistribution and use in source and binary forms, with or without 2745 | modification, are permitted provided that the following conditions are 2746 | met: 2747 | 2748 | * Redistributions of source code must retain the above copyright notice, 2749 | this list of conditions and the following disclaimer. 2750 | 2751 | * Redistributions in binary form must reproduce the above copyright 2752 | notice, this list of conditions and the following disclaimer in the 2753 | documentation and/or other materials provided with the distribution. 2754 | 2755 | * Neither the name "Markdown" nor the names of its contributors may 2756 | be used to endorse or promote products derived from this software 2757 | without specific prior written permission. 2758 | 2759 | This software is provided by the copyright holders and contributors "as 2760 | is" and any express or implied warranties, including, but not limited 2761 | to, the implied warranties of merchantability and fitness for a 2762 | particular purpose are disclaimed. In no event shall the copyright owner 2763 | or contributors be liable for any direct, indirect, incidental, special, 2764 | exemplary, or consequential damages (including, but not limited to, 2765 | procurement of substitute goods or services; loss of use, data, or 2766 | profits; or business interruption) however caused and on any theory of 2767 | liability, whether in contract, strict liability, or tort (including 2768 | negligence or otherwise) arising in any way out of the use of this 2769 | software, even if advised of the possibility of such damage. 2770 | 2771 | */ 2772 | ?> --------------------------------------------------------------------------------

Login

Blog New Post

Contact

History

".$this->runSpanGamut($matches[1])."

".$this->runSpanGamut($matches[1])."