/Markdown.class.php
PHP | 1221 lines | 814 code | 130 blank | 277 comment | 65 complexity | de47dc224306da991342b0709aeb6442 MD5 | raw file
Possible License(s): GPL-3.0
- <?php
- # A php class for parsing Markdown markup in comments, without authorizing
- # html tags. This is an adaptation of PHP Markdown by Michel Fortin.
- #
- # PHP Markdown for comments
- # Adaptation of PHP Markdown by Pierre Monchalin
- # <http://github.com/piero-la-lune/PHP-Markdown-for-comments>
- #
- # PHP Markdown
- # Copyright (c) 2004-2012 Michel Fortin
- # <http://michelf.com/projects/php-markdown/>
- #
- # Original Markdown
- # Copyright (c) 2004-2006 John Gruber
- # <http://daringfireball.net/projects/markdown/>
- #
- #
- #
- # This program is free software: you can redistribute it and/or modify
- # it under the terms of the GNU General Public License as published by
- # the Free Software Foundation, either version 3 of the License, or
- # (at your option) any later version.
- #
- # This program is distributed in the hope that it will be useful,
- # but WITHOUT ANY WARRANTY; without even the implied warranty of
- # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- # GNU General Public License for more details.
- #
- # You should have received a copy of the GNU General Public License
- # along with this program. If not, see <http://www.gnu.org/licenses/>.
- class Markdown {
- # Regex to match balanced [brackets].
- # Needed to insert a maximum bracked depth while converting to PHP.
- protected $nested_brackets_depth = 6;
- protected $nested_brackets_re;
- protected $nested_url_parenthesis_depth = 4;
- protected $nested_url_parenthesis_re;
- # Table of hash values for escaped characters:
- protected $escape_chars = '\`*_{}[]()>#+-.!';
- protected $escape_chars_re;
- # Predefined urls and titles for reference links and images.
- protected $predef_urls = array();
- protected $predef_titles = array();
- protected $tab_width = 4;
- protected $document_gamut = array(
- 'stripLinkDefinitions',
- 'runBlockGamut'
- );
- # These are all the transformations that form block-level tags like
- # paragraphs, headers, and list items.
- protected $block_gamut = array(
- 'doHeaders',
- 'doHorizontalRules',
- 'doLists',
- 'doCodeBlocks',
- 'doBlockQuotes'
- );
- # These are all the transformations that occur *within* block-level tags
- # like paragraphs, headers, and list items.
- protected $span_gamut = array(
- # Process character escapes, code spans, and inline HTML in one shot.
- 'parseSpan',
- # Process anchor and image tags. Images must come first, because
- # ![foo][f] looks like an anchor.
- 'doImages',
- 'doAnchors',
- # Make links out of things like `<http://example.com/>`
- # Must come after doAnchors, because you can use < and > delimiters in
- # inline links like [this](<url>).
- 'doAutoLinks',
- 'encodeAmpsAndAngles',
- 'doItalicsAndBold',
- 'doHardBreaks'
- );
- public function __construct() {
- $this->prepareItalicsAndBold();
- $this->nested_brackets_re =
- str_repeat('(?>[^\[\]]+|\[', $this->nested_brackets_depth).
- str_repeat('\])*', $this->nested_brackets_depth);
- $this->nested_url_parenthesis_re =
- str_repeat('(?>[^()\s]+|\(', $this->nested_url_parenthesis_depth).
- str_repeat('(?>\)))*', $this->nested_url_parenthesis_depth);
- $this->escape_chars_re = '['.preg_quote($this->escape_chars).']';
- }
- #
- # Main function. Performs some preprocessing on the input text and pass it
- # through the document gamut.
- #
- public function transform($text) {
- $this->setup();
- # Remove UTF-8 BOM and marker character in input, if present.
- $text = preg_replace('{^\xEF\xBB\xBF|\x1A}', '', $text);
- # Standardize line endings: DOS to Unix and Mac to Unix
- $text = preg_replace('{\r\n?}', "\n", $text);
- # Make sure $text ends with a couple of newlines:
- $text .= "\n\n";
- # Convert all tabs to spaces.
- $text = $this->detab($text);
- # Strip any lines consisting only of spaces and tabs.
- # This makes subsequent regexen easier to write, because we can match
- # consecutive blank lines with /\n+/ instead of something contorted
- # like /[ ]*\n+/ .
- $text = preg_replace('/^[ ]+$/m', '', $text);
- # Run document gamut methods.
- foreach ($this->document_gamut as $method) {
- $text = $this->$method($text);
- }
- $this->teardown();
- return $text . "\n";
- }
- #
- # Prepare regular expressions for searching emphasis tokens in any context.
- #
- private $em_relist = array(
- '' => '(?:(?<!\*)\*(?!\*)|(?<!_)_(?!_))(?=\S|$)(?![\.,:;]\s)',
- '*' => '(?<=\S|^)(?<!\*)\*(?!\*)',
- '_' => '(?<=\S|^)(?<!_)_(?!_)',
- );
- private $strong_relist = array(
- '' => '(?:(?<!\*)\*\*(?!\*)|(?<!_)__(?!_))(?=\S|$)(?![\.,:;]\s)',
- '**' => '(?<=\S|^)(?<!\*)\*\*(?!\*)',
- '__' => '(?<=\S|^)(?<!_)__(?!_)',
- );
- private $em_strong_relist = array(
- '' => '(?:(?<!\*)\*\*\*(?!\*)|(?<!_)___(?!_))(?=\S|$)(?![\.,:;]\s)',
- '***' => '(?<=\S|^)(?<!\*)\*\*\*(?!\*)',
- '___' => '(?<=\S|^)(?<!_)___(?!_)',
- );
- private $em_strong_prepared_relist;
- protected function prepareItalicsAndBold() {
- foreach ($this->em_relist as $em => $em_re) {
- foreach ($this->strong_relist as $strong => $strong_re) {
- # Construct list of allowed token expressions.
- $token_relist = array();
- if (isset($this->em_strong_relist["$em$strong"])) {
- $token_relist[] = $this->em_strong_relist["$em$strong"];
- }
- $token_relist[] = $em_re;
- $token_relist[] = $strong_re;
- # Construct master expression from list.
- $token_re = '{('. implode('|', $token_relist) .')}';
- $this->em_strong_prepared_relist["$em$strong"] = $token_re;
- }
- }
- }
- #
- # Called before the transformation process starts to setup parser states.
- #
- # Internal hashes used during transformation.
- private $urls = array();
- private $titles = array();
- private $html_hashes = array();
- # Status flag to avoid invalid nesting.
- private $in_anchor = false;
- protected function setup() {
- # Clear global hashes.
- $this->urls = $this->predef_urls;
- $this->titles = $this->predef_titles;
- $this->html_hashes = array();
- $in_anchor = false;
- }
- #
- # Called after the transformation process to clear any variable which may be
- # taking up memory unnecessarly.
- #
- protected function teardown() {
- $this->urls = array();
- $this->titles = array();
- $this->html_hashes = array();
- }
- #
- # Remove one level of line-leading tabs or spaces
- #
- protected function outdent($text) {
- return preg_replace('/^(\t|[ ]{1,'.$this->tab_width.'})/m', '', $text);
- }
- #
- # Replace tabs with the appropriate amount of space.
- #
- protected function detab($text) {
- # For each line we separate the line in blocks delemited by tab
- # characters. Then we reconstruct every line by adding the appropriate
- # number of space between each blocks.
- $text = preg_replace_callback('/^.*\t.*$/m',
- array(&$this, '_detab_callback'), $text);
- return $text;
- }
- protected function _detab_callback($matches) {
- $line = $matches[0];
- # Split in blocks.
- $blocks = explode("\t", $line);
- # Add each blocks to the line.
- $line = $blocks[0];
- unset($blocks[0]); # Do not add first block twice.
- foreach ($blocks as $block) {
- # Calculate amount of space, insert spaces, insert block.
- $amount = $this->tab_width -
- mb_strlen($line, 'UTF-8') % $this->tab_width;
- $line .= str_repeat(" ", $amount) . $block;
- }
- return $line;
- }
- #
- # Strips link definitions from text, stores the URLs and titles in hash
- # references.
- #
- protected function stripLinkDefinitions($text) {
- $less_than_tab = $this->tab_width - 1;
- # Link defs are in the form: ^[id]: url "optional title"
- $text = preg_replace_callback('{
- ^[ ]{0,'.$less_than_tab.'}\[(.+)\][ ]?: # id = $1
- [ ]*
- \n? # maybe *one* newline
- [ ]*
- (?:
- <(.+?)> # url = $2
- |
- (\S+?) # url = $3
- )
- [ ]*
- \n? # maybe one newline
- [ ]*
- (?:
- (?<=\s) # lookbehind for whitespace
- ["(]
- (.*?) # title = $4
- [")]
- [ ]*
- )? # title is optional
- (?:\n+|\Z)
- }xm',
- array(&$this, '_stripLinkDefinitions_callback'),
- $text);
- return $text;
- }
- protected function _stripLinkDefinitions_callback($matches) {
- $link_id = strtolower($matches[1]);
- $url = $matches[2] == '' ? $matches[3] : $matches[2];
- $this->urls[$link_id] = $url;
- $this->titles[$link_id] =& $matches[4];
- return ''; # String that will replace the block
- }
- #
- # Run block gamut tranformations, without hashing HTML blocks.
- #
- protected function runBlockGamut($text) {
- foreach ($this->block_gamut as $method) {
- $text = $this->$method($text);
- }
- # Finally form paragraph and restore hashed blocks.
- $text = $this->formParagraphs($text);
- return $text;
- }
- #
- # Run span gamut tranformations.
- #
- protected function runSpanGamut($text) {
- foreach ($this->span_gamut as $method) {
- $text = $this->$method($text);
- }
- return $text;
- }
- #
- # Params:
- # $text - string to process with html <p> tags
- #
- protected function formParagraphs($text) {
- # Strip leading and trailing lines:
- $text = preg_replace('/\A\n+|\n+\z/', '', $text);
- $grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
- #
- # Wrap <p> tags and unhashify HTML blocks
- #
- foreach ($grafs as $key => $value) {
- if (!preg_match('/^B\x1A[0-9]+B$/', $value)) {
- # Is a paragraph.
- $value = $this->runSpanGamut($value);
- $value = preg_replace('/^([ ]*)/', "<p>", $value);
- $value .= "</p>";
- $grafs[$key] = $this->unhash($value);
- }
- else {
- # Is a block.
- # Modify elements of @grafs in-place...
- $graf = $value;
- $block = $this->html_hashes[$graf];
- $graf = $block;
- $grafs[$key] = $graf;
- }
- }
- return implode("\n\n", $grafs);
- }
- #
- # Process headers
- # Note : this function was adapted, not allowing levels 1 and 2 for titles.
- #
- protected function doHeaders($text) {
- # Setext-style headers:
- # Header 1
- # ========
- #
- # Header 2
- # --------
- #
- $text = preg_replace_callback('{ ^(.+?)[ ]*\n(=+|-+)[ ]*\n+ }mx',
- array(&$this, '_doHeaders_callback_setext'), $text);
- # atx-style headers:
- # # Header 1
- # ## Header 2
- # ## Header 2 with closing hashes ##
- # ...
- # #### Header 4
- #
- $text = preg_replace_callback('{
- ^(\#{1,4}) # $1 = string of #\'s
- [ ]*
- (.+?) # $2 = Header text
- [ ]*
- \#* # optional closing #\'s (not counted)
- \n+
- }xm',
- array(&$this, '_doHeaders_callback_atx'), $text);
- return $text;
- }
- protected function _doHeaders_callback_setext($matches) {
- # Terrible hack to check we haven't found an empty list item.
- if ($matches[2] == '-' && preg_match('{^-(?: |$)}', $matches[1]))
- return $matches[0];
- $level = $matches[2]{0} == '=' ? 3 : 4;
- $block = "<h$level>".$this->runSpanGamut($matches[1])."</h$level>";
- return "\n" . $this->hashBlock($block) . "\n\n";
- }
- protected function _doHeaders_callback_atx($matches) {
- $level = strlen($matches[1])+2;
- $block = "<h$level>".$this->runSpanGamut($matches[2])."</h$level>";
- return "\n" . $this->hashBlock($block) . "\n\n";
- }
- #
- # Process horizontal rules
- #
- protected function doHorizontalRules($text) {
- return preg_replace(
- '{
- ^[ ]{0,3} # Leading space
- ([-*_]) # $1: First marker
- (?> # Repeated marker group
- [ ]{0,2} # Zero, one, or two spaces.
- \1 # Marker character
- ){2,} # Group repeated at least twice
- [ ]* # Tailing spaces
- $ # End of line.
- }mx',
- "\n".$this->hashBlock("<hr />")."\n",
- $text);
- }
- #
- # Form HTML ordered (numbered) and unordered (bulleted) lists.
- #
- protected function doLists($text) {
- $less_than_tab = $this->tab_width - 1;
- # Re-usable patterns to match list item bullets and number markers:
- $marker_ul_re = '[*+-]';
- $marker_ol_re = '\d+[\.]';
- $marker_any_re = "(?:$marker_ul_re|$marker_ol_re)";
- $markers_relist = array(
- $marker_ul_re => $marker_ol_re,
- $marker_ol_re => $marker_ul_re,
- );
- foreach ($markers_relist as $marker_re => $other_marker_re) {
- # Re-usable pattern to match any entirel ul or ol list:
- $whole_list_re = '
- ( # $1 = whole list
- ( # $2
- ([ ]{0,'.$less_than_tab.'}) # $3 = number of spaces
- ('.$marker_re.') # $4 = first list item marker
- [ ]+
- )
- (?s:.+?)
- ( # $5
- \z
- |
- \n{2,}
- (?=\S)
- (?! # Negative lookahead for another list item marker
- [ ]*
- '.$marker_re.'[ ]+
- )
- |
- (?= # Lookahead for another kind of list
- \n
- \3 # Must have the same indentation
- '.$other_marker_re.'[ ]+
- )
- )
- )
- '; // mx
- # We use a different prefix before nested lists than top-level lists.
- # See extended comment in _ProcessListItems().
- if ($this->list_level) {
- $text = preg_replace_callback('{
- ^
- '.$whole_list_re.'
- }mx',
- array(&$this, '_doLists_callback'), $text);
- }
- else {
- $text = preg_replace_callback('{
- (?:(?<=\n)\n|\A\n?) # Must eat the newline
- '.$whole_list_re.'
- }mx',
- array(&$this, '_doLists_callback'), $text);
- }
- }
- return $text;
- }
- protected function _doLists_callback($matches) {
- # Re-usable patterns to match list item bullets and number markers:
- $marker_ul_re = '[*+-]';
- $marker_ol_re = '\d+[\.]';
- $marker_any_re = "(?:$marker_ul_re|$marker_ol_re)";
- $list = $matches[1];
- $list_type = preg_match("/$marker_ul_re/", $matches[4]) ? "ul" : "ol";
- $marker_any_re = ( $list_type == "ul" ? $marker_ul_re : $marker_ol_re );
- $list .= "\n";
- $result = $this->processListItems($list, $marker_any_re);
- $result = $this->hashBlock("<$list_type>\n" . $result . "</$list_type>");
- return "\n". $result ."\n\n";
- }
- #
- # Process the contents of a single ordered or unordered list, splitting it
- # into individual list items.
- #
- # The $this->list_level global keeps track of when we're inside a list.
- # Each time we enter a list, we increment it; when we leave a list,
- # we decrement. If it's zero, we're not in a list anymore.
- #
- # We do this because when we're not inside a list, we want to treat
- # something like this:
- #
- # I recommend upgrading to version
- # 8. Oops, now this line is treated
- # as a sub-list.
- #
- # As a single paragraph, despite the fact that the second line starts
- # with a digit-period-space sequence.
- #
- # Whereas when we're inside a list (or sub-list), that line will be
- # treated as the start of a sub-list. What a kludge, huh? This is
- # an aspect of Markdown's syntax that's hard to parse perfectly
- # without resorting to mind-reading. Perhaps the solution is to
- # change the syntax rules such that sub-lists must start with a
- # starting cardinal number; e.g. "1." or "a.".
- private $list_level = 0;
- protected function processListItems($list_str, $marker_any_re) {
- $this->list_level++;
- # trim trailing blank lines:
- $list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str);
- $list_str = preg_replace_callback('{
- (\n)? # leading line = $1
- (^[ ]*) # leading whitespace = $2
- ('.$marker_any_re.' # list marker and space = $3
- (?:[ ]+|(?=\n)) # space only required if item is not empty
- )
- ((?s:.*?)) # list item text = $4
- (?:(\n+(?=\n))|\n) # tailing blank line = $5
- (?= \n* (\z | \2 ('.$marker_any_re.') (?:[ ]+|(?=\n))))
- }xm',
- array(&$this, '_processListItems_callback'), $list_str);
- $this->list_level--;
- return $list_str;
- }
- protected function _processListItems_callback($matches) {
- $item = $matches[4];
- $leading_line =& $matches[1];
- $leading_space =& $matches[2];
- $marker_space = $matches[3];
- $tailing_blank_line =& $matches[5];
- if ($leading_line || $tailing_blank_line ||
- preg_match('/\n{2,}/', $item))
- {
- # Replace marker with the appropriate whitespace indentation
- $item = $leading_space . str_repeat(' ', strlen($marker_space)) . $item;
- $item = $this->runBlockGamut($this->outdent($item)."\n");
- }
- else {
- # Recursion for sub-lists:
- $item = $this->doLists($this->outdent($item));
- $item = preg_replace('/\n+$/', '', $item);
- $item = $this->runSpanGamut($item);
- }
- return "<li>" . $item . "</li>\n";
- }
- #
- # Process Markdown `<pre><code>` blocks.
- #
- protected function doCodeBlocks($text) {
- $text = preg_replace_callback('{
- (?:\n\n|\A\n?)
- ( # $1 = the code block -- one or more lines, starting with a space/tab
- (?>
- [ ]{'.$this->tab_width.'} # Lines must start with a tab or a tab-width of spaces
- .*\n+
- )+
- )
- ((?=^[ ]{0,'.$this->tab_width.'}\S)|\Z) # Lookahead for non-space at line-start, or end of doc
- }xm',
- array(&$this, '_doCodeBlocks_callback'), $text);
- return $text;
- }
- protected function _doCodeBlocks_callback($matches) {
- $codeblock = $matches[1];
- $codeblock = $this->outdent($codeblock);
- $codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES);
- # trim leading newlines and trailing newlines
- $codeblock = preg_replace('/\A\n+|\n+\z/', '', $codeblock);
- $codeblock = "<pre><code>$codeblock\n</code></pre>";
- return "\n\n".$this->hashBlock($codeblock)."\n\n";
- }
- #
- # Process blockquotes.
- #
- protected function doBlockQuotes($text) {
- $text = preg_replace_callback('/
- ( # Wrap whole match in $1
- (?>
- ^[ ]*>[ ]? # ">" at the start of a line
- .+\n # rest of the first line
- (.+\n)* # subsequent consecutive lines
- \n* # blanks
- )+
- )
- /xm',
- array(&$this, '_doBlockQuotes_callback'), $text);
- return $text;
- }
- protected function _doBlockQuotes_callback($matches) {
- $bq = $matches[1];
- # trim one level of quoting - trim whitespace-only lines
- $bq = preg_replace('/^[ ]*>[ ]?|^[ ]+$/m', '', $bq);
- $bq = $this->runBlockGamut($bq); # recurse
- $bq = preg_replace('/^/m', " ", $bq);
- # These leading spaces cause problem with <pre> content,
- # so we need to fix that:
- $bq = preg_replace_callback('{(\s*<pre>.+?</pre>)}sx',
- array(&$this, '_doBlockQuotes_callback2'), $bq);
- return "\n". $this->hashBlock("<blockquote>\n$bq\n</blockquote>")."\n\n";
- }
- protected function _doBlockQuotes_callback2($matches) {
- $pre = $matches[1];
- $pre = preg_replace('/^ /m', '', $pre);
- return $pre;
- }
- #
- # Take the string $str and parse it into tokens, hashing embeded HTML, escaped
- # characters and handling code spans.
- #
- protected function parseSpan($str) {
- $output = '';
- $span_re = '{
- (
- \\\\'.$this->escape_chars_re.'
- |
- (?<![`\\\\])
- `+ # code span marker
- )
- }xs';
- while (1) {
- #
- # Each loop iteration seach for either the next tag, the next
- # openning code span marker, or the next escaped character.
- # Each token is then passed to handleSpanToken.
- #
- $parts = preg_split($span_re, $str, 2, PREG_SPLIT_DELIM_CAPTURE);
- # Create token from text preceding tag.
- if ($parts[0] != "") {
- $output .= $parts[0];
- }
- # Check if we reach the end.
- if (isset($parts[1])) {
- $output .= $this->handleSpanToken($parts[1], $parts[2]);
- $str = $parts[2];
- }
- else {
- break;
- }
- }
- return $output;
- }
- #
- # Handle $token provided by parseSpan by determining its nature and returning
- # the corresponding value that should replace it.
- #
- protected function handleSpanToken($token, &$str) {
- switch ($token{0}) {
- case "\\":
- return $this->hashPart("&#". ord($token{1}). ";");
- case "`":
- # Search for end marker in remaining text.
- if (preg_match('/^(.*?[^`])'.preg_quote($token).'(?!`)(.*)$/sm',
- $str, $matches))
- {
- $str = $matches[2];
- $codespan = $this->makeCodeSpan($matches[1]);
- return $this->hashPart($codespan);
- }
- return $token; // return as text since no ending marker found.
- default:
- return $this->hashPart($token);
- }
- }
- #
- # Create a code span markup for $code.
- #
- protected function makeCodeSpan($code) {
- $code = htmlspecialchars(trim($code), ENT_NOQUOTES);
- return $this->hashPart("<code>$code</code>");
- }
- #
- # Turn Markdown image shortcuts into <img> tags.
- #
- protected function doImages($text) {
- #
- # First, handle reference-style labeled images: ![alt text][id]
- #
- $text = preg_replace_callback('{
- ( # wrap whole match in $1
- !\[
- ('.$this->nested_brackets_re.') # alt text = $2
- \]
- [ ]? # one optional space
- (?:\n[ ]*)? # one optional newline followed by spaces
- \[
- (.*?) # id = $3
- \]
- )
- }xs',
- array(&$this, '_doImages_reference_callback'), $text);
- #
- # Next, handle inline images: ![alt text](url "optional title")
- # Don't forget: encode * and _
- #
- $text = preg_replace_callback('{
- ( # wrap whole match in $1
- !\[
- ('.$this->nested_brackets_re.') # alt text = $2
- \]
- \s? # One optional whitespace character
- \( # literal paren
- [ \n]*
- (?:
- <(\S*)> # src url = $3
- |
- ('.$this->nested_url_parenthesis_re.') # src url = $4
- )
- [ \n]*
- ( # $5
- ([\'"]) # quote char = $6
- (.*?) # title = $7
- \6 # matching quote
- [ \n]*
- )? # title is optional
- \)
- )
- }xs',
- array(&$this, '_doImages_inline_callback'), $text);
- return $text;
- }
- protected function _doImages_reference_callback($matches) {
- $whole_match = $matches[1];
- $alt_text = $matches[2];
- $link_id = strtolower($matches[3]);
- if ($link_id == "") {
- $link_id = strtolower($alt_text); # for shortcut links like ![this][].
- }
- $alt_text = $this->encodeAttribute($alt_text);
- if (isset($this->urls[$link_id])) {
- $url = $this->encodeAttribute($this->urls[$link_id]);
- $result = "<img src=\"$url\" alt=\"$alt_text\"";
- if (isset($this->titles[$link_id])) {
- $title = $this->titles[$link_id];
- $title = $this->encodeAttribute($title);
- $result .= " title=\"$title\"";
- }
- $result .= ' />';
- $result = $this->hashPart($result);
- }
- else {
- # If there's no such link ID, leave intact:
- $result = $whole_match;
- }
- return $result;
- }
- protected function _doImages_inline_callback($matches) {
- $whole_match = $matches[1];
- $alt_text = $matches[2];
- $url = $matches[3] == '' ? $matches[4] : $matches[3];
- $title =& $matches[7];
- $alt_text = $this->encodeAttribute($alt_text);
- $url = $this->encodeAttribute($url);
- $result = "<img src=\"$url\" alt=\"$alt_text\"";
- if (isset($title)) {
- $title = $this->encodeAttribute($title);
- $result .= " title=\"$title\""; # $title already quoted
- }
- $result .= ' />';
- return $this->hashPart($result);
- }
- #
- # Turn Markdown link shortcuts into XHTML <a> tags.
- #
- protected function doAnchors($text) {
- if ($this->in_anchor) return $text;
- $this->in_anchor = true;
- #
- # First, handle reference-style links: [link text] [id]
- #
- $text = preg_replace_callback('{
- ( # wrap whole match in $1
- \[
- ('.$this->nested_brackets_re.') # link text = $2
- \]
- [ ]? # one optional space
- (?:\n[ ]*)? # one optional newline followed by spaces
- \[
- (.*?) # id = $3
- \]
- )
- }xs',
- array(&$this, '_doAnchors_reference_callback'), $text);
- #
- # Next, inline-style links: [link text](url "optional title")
- #
- $text = preg_replace_callback('{
- ( # wrap whole match in $1
- \[
- ('.$this->nested_brackets_re.') # link text = $2
- \]
- \( # literal paren
- [ \n]*
- (?:
- <(.+?)> # href = $3
- |
- ('.$this->nested_url_parenthesis_re.') # href = $4
- )
- [ \n]*
- ( # $5
- ([\'"]) # quote char = $6
- (.*?) # Title = $7
- \6 # matching quote
- [ \n]* # ignore any spaces/tabs between closing quote and )
- )? # title is optional
- \)
- )
- }xs',
- array(&$this, '_doAnchors_inline_callback'), $text);
- #
- # Last, handle reference-style shortcuts: [link text]
- # These must come last in case you've also got [link text][1]
- # or [link text](/foo)
- #
- $text = preg_replace_callback('{
- ( # wrap whole match in $1
- \[
- ([^\[\]]+) # link text = $2; can\'t contain [ or ]
- \]
- )
- }xs',
- array(&$this, '_doAnchors_reference_callback'), $text);
- $this->in_anchor = false;
- return $text;
- }
- protected function _doAnchors_reference_callback($matches) {
- $whole_match = $matches[1];
- $link_text = $matches[2];
- $link_id =& $matches[3];
- if ($link_id == "") {
- # for shortcut links like [this][] or [this].
- $link_id = $link_text;
- }
- # lower-case and turn embedded newlines into spaces
- $link_id = strtolower($link_id);
- $link_id = preg_replace('{[ ]?\n}', ' ', $link_id);
- if (isset($this->urls[$link_id])) {
- $url = $this->urls[$link_id];
- $url = $this->encodeAttribute($url);
- $result = "<a href=\"$url\"";
- if ( isset( $this->titles[$link_id] ) ) {
- $title = $this->titles[$link_id];
- $title = $this->encodeAttribute($title);
- $result .= " title=\"$title\"";
- }
- $link_text = $this->runSpanGamut($link_text);
- $result .= ">$link_text</a>";
- $result = $this->hashPart($result);
- }
- else {
- $result = $whole_match;
- }
- return $result;
- }
- protected function _doAnchors_inline_callback($matches) {
- $whole_match = $matches[1];
- $link_text = $this->runSpanGamut($matches[2]);
- $url = $matches[3] == '' ? $matches[4] : $matches[3];
- $title =& $matches[7];
- $url = $this->encodeAttribute($url);
- $result = "<a href=\"$url\"";
- if (isset($title)) {
- $title = $this->encodeAttribute($title);
- $result .= " title=\"$title\"";
- }
- $link_text = $this->runSpanGamut($link_text);
- $result .= ">$link_text</a>";
- return $this->hashPart($result);
- }
- #
- # Process <email> or <url>
- #
- protected function doAutoLinks($text) {
- $text = preg_replace_callback('{<((https?|ftp|dict):[^\'">\s]+)>}i',
- array(&$this, '_doAutoLinks_url_callback'), $text);
- # Email addresses: <address@domain.foo>
- $text = preg_replace_callback('{
- <
- (?:mailto:)?
- (
- (?:
- [-!#$%&\'*+/=?^_`.{|}~\w\x80-\xFF]+
- |
- ".*?"
- )
- \@
- (?:
- [-a-z0-9\x80-\xFF]+(\.[-a-z0-9\x80-\xFF]+)*\.[a-z]+
- |
- \[[\d.a-fA-F:]+\] # IPv4 & IPv6
- )
- )
- >
- }xi',
- array(&$this, '_doAutoLinks_email_callback'), $text);
- return $text;
- }
- protected function _doAutoLinks_url_callback($matches) {
- $url = $this->encodeAttribute($matches[1]);
- $link = "<a href=\"$url\">$url</a>";
- return $this->hashPart($link);
- }
- protected function _doAutoLinks_email_callback($matches) {
- $address = $matches[1];
- $link = $this->encodeEmailAddress($address);
- return $this->hashPart($link);
- }
- #
- # Input: an email address, e.g. "foo@example.com"
- #
- # Output: the email address as a mailto link, with each character
- # of the address encoded as either a decimal or hex entity, in
- # the hopes of foiling most address harvesting spam bots. E.g.:
- #
- # <p><a href="mailto:foo
- # @example.co
- # m">foo@exampl
- # e.com</a></p>
- #
- # Based by a filter by Matthew Wickline, posted to BBEdit-Talk.
- # With some optimizations by Milian Wolff.
- #
- protected function encodeEmailAddress($addr) {
- $addr = "mailto:" . $addr;
- $chars = preg_split('/(?<!^)(?!$)/', $addr);
- $seed = (int)abs(crc32($addr) / strlen($addr)); # Deterministic seed.
- foreach ($chars as $key => $char) {
- $ord = ord($char);
- # Ignore non-ascii chars.
- if ($ord < 128) {
- $r = ($seed * (1 + $key)) % 100; # Pseudo-random function.
- # roughly 10% raw, 45% hex, 45% dec
- # '@' *must* be encoded. I insist.
- if ($r > 90 && $char != '@') /* do nothing */;
- else if ($r < 45) $chars[$key] = '&#x'.dechex($ord).';';
- else $chars[$key] = '&#'.$ord.';';
- }
- }
- $addr = implode('', $chars);
- $text = implode('', array_slice($chars, 7)); # text without `mailto:`
- $addr = "<a href=\"$addr\">$text</a>";
- return $addr;
- }
- #
- # Process italic and bold
- #
- protected function doItalicsAndBold($text) {
- $token_stack = array('');
- $text_stack = array('');
- $em = '';
- $strong = '';
- $tree_char_em = false;
- while (1) {
- #
- # Get prepared regular expression for seraching emphasis tokens
- # in current context.
- #
- $token_re = $this->em_strong_prepared_relist["$em$strong"];
- #
- # Each loop iteration search for the next emphasis token.
- # Each token is then passed to handleSpanToken.
- #
- $parts = preg_split($token_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE);
- $text_stack[0] .= $parts[0];
- $token =& $parts[1];
- $text =& $parts[2];
- if (empty($token)) {
- # Reached end of text span: empty stack without emitting.
- # any more emphasis.
- while ($token_stack[0]) {
- $text_stack[1] .= array_shift($token_stack);
- $text_stack[0] .= array_shift($text_stack);
- }
- break;
- }
- $token_len = strlen($token);
- if ($tree_char_em) {
- # Reached closing marker while inside a three-char emphasis.
- if ($token_len == 3) {
- # Three-char closing marker, close em and strong.
- array_shift($token_stack);
- $span = array_shift($text_stack);
- $span = $this->runSpanGamut($span);
- $span = "<strong><em>$span</em></strong>";
- $text_stack[0] .= $this->hashPart($span);
- $em = '';
- $strong = '';
- } else {
- # Other closing marker: close one em or strong and
- # change current token state to match the other
- $token_stack[0] = str_repeat($token{0}, 3-$token_len);
- $tag = $token_len == 2 ? "strong" : "em";
- $span = $text_stack[0];
- $span = $this->runSpanGamut($span);
- $span = "<$tag>$span</$tag>";
- $text_stack[0] = $this->hashPart($span);
- $$tag = ''; # $$tag stands for $em or $strong
- }
- $tree_char_em = false;
- } else if ($token_len == 3) {
- if ($em) {
- # Reached closing marker for both em and strong.
- # Closing strong marker:
- for ($i = 0; $i < 2; ++$i) {
- $shifted_token = array_shift($token_stack);
- $tag = strlen($shifted_token) == 2 ? "strong" : "em";
- $span = array_shift($text_stack);
- $span = $this->runSpanGamut($span);
- $span = "<$tag>$span</$tag>";
- $text_stack[0] .= $this->hashPart($span);
- $$tag = ''; # $$tag stands for $em or $strong
- }
- } else {
- # Reached opening three-char emphasis marker. Push on token
- # stack; will be handled by the special condition above.
- $em = $token{0};
- $strong = "$em$em";
- array_unshift($token_stack, $token);
- array_unshift($text_stack, '');
- $tree_char_em = true;
- }
- } else if ($token_len == 2) {
- if ($strong) {
- # Unwind any dangling emphasis marker:
- if (strlen($token_stack[0]) == 1) {
- $text_stack[1] .= array_shift($token_stack);
- $text_stack[0] .= array_shift($text_stack);
- }
- # Closing strong marker:
- array_shift($token_stack);
- $span = array_shift($text_stack);
- $span = $this->runSpanGamut($span);
- $span = "<strong>$span</strong>";
- $text_stack[0] .= $this->hashPart($span);
- $strong = '';
- } else {
- array_unshift($token_stack, $token);
- array_unshift($text_stack, '');
- $strong = $token;
- }
- } else {
- # Here $token_len == 1
- if ($em) {
- if (strlen($token_stack[0]) == 1) {
- # Closing emphasis marker:
- array_shift($token_stack);
- $span = array_shift($text_stack);
- $span = $this->runSpanGamut($span);
- $span = "<em>$span</em>";
- $text_stack[0] .= $this->hashPart($span);
- $em = '';
- } else {
- $text_stack[0] .= $token;
- }
- } else {
- array_unshift($token_stack, $token);
- array_unshift($text_stack, '');
- $em = $token;
- }
- }
- }
- return $text_stack[0];
- }
- #
- # Process hard breaks
- #
- protected function doHardBreaks($text) {
- return preg_replace_callback('/ {2,}\n/',
- array(&$this, '_doHardBreaks_callback'), $text);
- }
- protected function _doHardBreaks_callback($matches) {
- return $this->hashPart("<br />\n");
- }
- #
- # Called whenever a tag must be hashed when a function insert an atomic
- # element in the text stream. Passing $text to through this function gives
- # a unique text-token which will be reverted back when calling unhash.
- #
- # The $boundary argument specify what character should be used to surround
- # the token. By convension, "B" is used for block elements that needs not
- # to be wrapped into paragraph tags at the end, ":" is used for elements
- # that are word separators and "X" is used in the general case.
- #
- protected function hashPart($text, $boundary = 'X') {
- # Swap back any tag hash found in $text so we do not have to `unhash`
- # multiple times at the end.
- $text = $this->unhash($text);
- # Then hash the block.
- static $i = 0;
- $key = "$boundary\x1A" . ++$i . $boundary;
- $this->html_hashes[$key] = $text;
- return $key; # String that will replace the tag.
- }
- #
- # Shortcut function for hashPart with block-level boundaries.
- #
- protected function hashBlock($text) {
- return $this->hashPart($text, 'B');
- }
- #
- # Swap back in all the tags hashed by _HashHTMLBlocks.
- # Is this function still usefull, considering _HashHTMLBlocks was removed (no
- # html tags authorized in comments)?
- #
- protected function unhash($text) {
- return preg_replace_callback('/(.)\x1A[0-9]+\1/',
- array(&$this, '_unhash_callback'), $text);
- }
- protected function _unhash_callback($matches) {
- return $this->html_hashes[$matches[0]];
- }
- #
- # Encode text for a double-quoted HTML attribute. This function is *not*
- # suitable for attributes enclosed in single quotes.
- #
- protected function encodeAttribute($text) {
- $text = $this->encodeAmpsAndAngles($text);
- $text = str_replace('"', '"', $text);
- return $text;
- }
- #
- # Smart processing for ampersands and angle brackets that need to be encoded.
- # Valid character entities are left alone unless the no-entities mode is set.
- #
- protected function encodeAmpsAndAngles($text) {
- # Ampersand-encoding based entirely on Nat Irons's Amputator
- # MT plugin: <http://bumppo.net/projects/amputator/>
- $text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/',
- '&', $text);;
- # Encode remaining <'s
- $text = str_replace('<', '<', $text);
- return $text;
- }
- }
- ?>