WordPress-C
/
WordPress
mirror of https://github.com/WordPress/WordPress.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Bump to Markdown 1.0 

git-svn-id: http://svn.automattic.com/wordpress/trunk@1593 1a063a9b-81f0-0310-95a4-ce76da25c4cd
This commit is contained in:
saxmatt 2004-09-04 20:48:57 +00:00
parent da9e7308b7
commit 336bd345c4
1 changed files with 284 additions and 217 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

501
wp-content/plugins/markdown.php
View File

@ -1,27 +1,66 @@
<?php <?php
/*
Plugin Name: MarkDown #
Plugin URI: http://daringfireball.net/projects/markdown/ # Markdown - A text-to-HTML conversion tool for web writers
Description: Markdown is a text-to-HTML conversion tool for web writers. <a href="http://daringfireball.net/projects/markdown/syntax">Markdown syntax</a> allows you to write using an easy-to-read, easy-to-write plain text format, then convert it to structurally valid XHTML. This plugin <strong>enables Markdown for your posts and comments</strong>. Written by <a href="http://daringfireball.net/">John Gruber</a> in Perl, translated to PHP by <a href="http://www.michelf.com/">Michel Fortin</a>, and made a WP plugin by <a href="http://photomatt.net/">Matt</a>. If you use this you should disable Textile 1 and 2 because the syntax conflicts. #
Version: 1.0b4 # Copyright (c) 2004 John Gruber
Author: John Gruber # <http://daringfireball.net/projects/markdown/>
Author URI: http://daringfireball.net/ #
*/ # Copyright (c) 2004 Michel Fortin - Translation to PHP
# <http://www.michelf.com/projects/php-markdown/>
#
# This version has been modified for inclusion in WordPress
# For the original please see Michel's site
global $MarkdownPHPVersion, $MarkdownSyntaxVersion,
$md_empty_element_suffix, $md_tab_width,
$md_nested_brackets_depth, $md_nested_brackets,
$md_escape_table, $md_backslash_escape_table;
$MarkdownPHPVersion = '1.0'; # Sat 21 Aug 2004
$MarkdownSyntaxVersion = '1.0'; # Fri 20 Aug 2004
#
# Global default settings:
#
$md_empty_element_suffix = " />"; # Change to ">" for HTML output
$md_tab_width = 4;
# -- WordPress Plugin Interface -----------------------------------------------
/* /*
Note to code readers: I've stripped most of the comments from the source, see the original at http://www.michelf.com/php-markdown/?code to get the unaltered version. --Matt Plugin Name: Markdown
Plugin URI: http://codex.wordpress.org/Plugin:Markdown
Description: <a href="http://daringfireball.net/projects/markdown/syntax">Markdown syntax</a> allows you to write using an easy-to-read, easy-to-write plain text format. Based on the original Perl version by <a href="http://daringfireball.net/">John Gruber</a>. <a href="http://www.michelf.com/projects/php-markdown/">More...</a>
Version: 1.0
Author: Michel Fortin
Author URI: http://www.michelf.com/
*/ */
if (isset($wp_version)) {
# Remove default WordPress auto-paragraph filter.
remove_filter('the_content', 'wpautop');
remove_filter('the_excerpt', 'wpautop');
remove_filter('comment_text', 'wpautop');
# Add Markdown filter with priority 6 (same as Textile).
add_filter('the_content', 'Markdown', 6);
add_filter('the_excerpt', 'Markdown', 6);
add_filter('comment_text', 'Markdown', 6);
}
$MarkdownPHPVersion = '1.0b4.1'; # Sun 4 Apr 2004 function smarty_modifier_markdown($text) {
$MarkdownSyntaxVersion = '1.0b4'; # Thu 25 Mar 2004 return Markdown($text);
$g_empty_element_suffix = " />"; # Change to ">" for HTML output }
$g_tab_width = 4;
$g_nested_brackets_depth = 6; $md_nested_brackets_depth = 6;
$g_nested_brackets = $md_nested_brackets =
str_repeat('(?>[^\[\]]+|\[', $g_nested_brackets_depth). str_repeat('(?>[^\[\]]+|\[', $md_nested_brackets_depth).
str_repeat('\])*', $g_nested_brackets_depth); str_repeat('\])*', $md_nested_brackets_depth);
$g_escape_table = array(
$md_escape_table = array(
"\\" => md5("\\"), "\\" => md5("\\"),
"`" => md5("`"), "`" => md5("`"),
"*" => md5("*"), "*" => md5("*"),
@ -36,38 +75,47 @@ $g_escape_table = array(
"." => md5("."), "." => md5("."),
"!" => md5("!") "!" => md5("!")
); );
$g_backslash_escape_table; # Create an identical table but for escaped characters.
foreach ($g_escape_table as $key => $char) $md_backslash_escape_table;
$g_backslash_escape_table["\\$key"] = $char; foreach ($md_escape_table as $key => $char)
$md_backslash_escape_table["\\$key"] = $char;
$g_urls;
$g_titles;
$g_html_blocks;
function Markdown($text) { function Markdown($text) {
global $g_urls, $g_titles, $g_html_blocks; global $md_urls, $md_titles, $md_html_blocks;
$g_urls = array(); $md_urls = array();
$g_titles = array(); $md_titles = array();
$g_html_blocks = array(); $md_html_blocks = array();
$text = str_replace(array("\r\n", "\r"), "\n", $text); $text = str_replace(array("\r\n", "\r"), "\n", $text);
$text .= "\n\n"; $text .= "\n\n";
$text = _Detab($text); $text = _Detab($text);
$text = preg_replace('/^[ \t]+$/m', '', $text); $text = preg_replace('/^[ \t]+$/m', '', $text);
$text = _HashHTMLBlocks($text); $text = _HashHTMLBlocks($text);
$text = _StripLinkDefinitions($text); $text = _StripLinkDefinitions($text);
$text = _EscapeSpecialChars($text); $text = _EscapeSpecialChars($text);
$text = _RunBlockGamut($text); $text = _RunBlockGamut($text);
$text = _UnescapeSpecialChars($text); $text = _UnescapeSpecialChars($text);
return $text . "\n"; return $text . "\n";
} }
function _StripLinkDefinitions($text) { function _StripLinkDefinitions($text) {
$text = preg_replace_callback('{ $text = preg_replace_callback('{
^[ \t]*\[(.+)\]: # id = $1 ^[ \t]*\[(.+)\]: # id = $1
[ \t]* [ \t]*
\n? # maybe *one* newline \n? # maybe *one* newline
[ \t]* [ \t]*
(\S+) # url = $2 <?(\S+?)>? # url = $2
[ \t]* [ \t]*
\n? # maybe one newline \n? # maybe one newline
[ \t]* [ \t]*
@ -85,20 +133,25 @@ function _StripLinkDefinitions($text) {
return $text; return $text;
} }
function _StripLinkDefinitions_callback($matches) { function _StripLinkDefinitions_callback($matches) {
global $g_urls, $g_titles; global $md_urls, $md_titles;
$link_id = strtolower($matches[1]); $link_id = strtolower($matches[1]);
$g_urls[$link_id] = _EncodeAmpsAndAngles($matches[2]); $md_urls[$link_id] = _EncodeAmpsAndAngles($matches[2]);
if (isset($matches[3])) if (isset($matches[3]))
$g_titles[$link_id] = htmlentities($matches[3]); $md_titles[$link_id] = htmlentities($matches[3]);
return ''; # String that will replace the block return ''; # String that will replace the block
} }
function _HashHTMLBlocks($text) { function _HashHTMLBlocks($text) {
$block_tag_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|script'; $block_tags_a = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|'.
'script|noscript|form|fieldset|iframe|math|ins|del';
$block_tags_b = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|'.
'script|noscript|form|fieldset|iframe|math';
$text = preg_replace_callback("{ $text = preg_replace_callback("{
( # save in $1 ( # save in $1
^ # start of line (with /m) ^ # start of line (with /m)
<($block_tag_re) # start tag = $2 <($block_tags_a) # start tag = $2
\\b # word break \\b # word break
(.*\\n)*? # any number of lines, minimally matching (.*\\n)*? # any number of lines, minimally matching
</\\2> # the matching end tag </\\2> # the matching end tag
@ -112,7 +165,7 @@ function _HashHTMLBlocks($text) {
$text = preg_replace_callback("{ $text = preg_replace_callback("{
( # save in $1 ( # save in $1
^ # start of line (with /m) ^ # start of line (with /m)
<($block_tag_re) # start tag = $2 <($block_tags_b) # start tag = $2
\\b # word break \\b # word break
(.*\\n)*? # any number of lines, minimally matching (.*\\n)*? # any number of lines, minimally matching
.*</\\2> # the matching end tag .*</\\2> # the matching end tag
@ -144,23 +197,25 @@ function _HashHTMLBlocks($text) {
return $text; return $text;
} }
function _HashHTMLBlocks_callback($matches) { function _HashHTMLBlocks_callback($matches) {
global $g_html_blocks; global $md_html_blocks;
$text = $matches[1]; $text = $matches[1];
$key = md5($text); $key = md5($text);
$g_html_blocks[$key] = $text; $md_html_blocks[$key] = $text;
return "\n\n$key\n\n"; # String that will replace the block return "\n\n$key\n\n"; # String that will replace the block
} }
function _RunBlockGamut($text) { function _RunBlockGamut($text) {
global $g_empty_element_suffix; global $md_empty_element_suffix;
$text = _DoHeaders($text); $text = _DoHeaders($text);
# Do Horizontal Rules:
$text = preg_replace( $text = preg_replace(
array('/^( ?\* ?){3,}$/m', array('/^( ?\* ?){3,}$/m',
'/^( ?- ?){3,}$/m'), '/^( ?- ?){3,}$/m',
array("\n<hr$g_empty_element_suffix\n", '/^( ?_ ?){3,}$/m'),
"\n<hr$g_empty_element_suffix\n"), "\n<hr$md_empty_element_suffix\n",
$text); $text);
$text = _DoLists($text); $text = _DoLists($text);
@ -169,6 +224,7 @@ function _RunBlockGamut($text) {
$text = _DoBlockQuotes($text); $text = _DoBlockQuotes($text);
# Make links out of things like `<http://example.com/>`
$text = _DoAutoLinks($text); $text = _DoAutoLinks($text);
$text = _HashHTMLBlocks($text); $text = _HashHTMLBlocks($text);
@ -180,45 +236,40 @@ function _RunBlockGamut($text) {
function _RunSpanGamut($text) { function _RunSpanGamut($text) {
global $g_empty_element_suffix; global $md_empty_element_suffix;
$text = _DoCodeSpans($text); $text = _DoCodeSpans($text);
# Fix unencoded ampersands and <'s:
$text = _EncodeAmpsAndAngles($text); $text = _EncodeAmpsAndAngles($text);
# Process anchor and image tags. Images must come first,
# because ![foo][f] looks like an anchor.
$text = _DoImages($text); $text = _DoImages($text);
$text = _DoAnchors($text); $text = _DoAnchors($text);
$text = _DoItalicsAndBold($text); $text = _DoItalicsAndBold($text);
# Do hard breaks: $text = preg_replace('/ {2,}\n/', "<br$md_empty_element_suffix\n", $text);
$text = preg_replace('/ {2,}\n/', "<br$g_empty_element_suffix\n", $text);
return $text; return $text;
} }
function _EscapeSpecialChars($text) { function _EscapeSpecialChars($text) {
global $g_escape_table; global $md_escape_table;
$tokens = _TokenizeHTML($text); $tokens = _TokenizeHTML($text);
$text = ''; # rebuild $text from the tokens $text = ''; # rebuild $text from the tokens
$in_pre = 0; # Keep track of when we're inside <pre> or <code> tags.
$tags_to_skip = "!<(/?)(?:pre|code|kbd|script)[\s>]!";
foreach ($tokens as $cur_token) { foreach ($tokens as $cur_token) {
if ($cur_token[0] == 'tag') { if ($cur_token[0] == 'tag') {
$cur_token[1] = str_replace(array('*', '_'), $cur_token[1] = str_replace(array('*', '_'),
array($g_escape_table['*'], $g_escape_table['_']), array($md_escape_table['*'], $md_escape_table['_']),
$cur_token[1]); $cur_token[1]);
$text .= $cur_token[1]; $text .= $cur_token[1];
} else { } else {
$t = $cur_token[1]; $t = $cur_token[1];
if (! $in_pre) { $t = _EncodeBackslashEscapes($t);
$t = _EncodeBackslashEscapes($t);
# $t =~ s{([a-z])/([a-z])}{$1&thinsp;/&thinsp;$2}ig;
}
$text .= $t; $text .= $t;
} }
} }
@ -227,48 +278,49 @@ function _EscapeSpecialChars($text) {
function _DoAnchors($text) { function _DoAnchors($text) {
global $g_nested_brackets; global $md_nested_brackets;
#
# First, handle reference-style links: [link text] [id]
#
$text = preg_replace_callback("{ $text = preg_replace_callback("{
( # wrap whole match in $1 ( # wrap whole match in $1
\\[ \\[
($g_nested_brackets) # link text = $2 ($md_nested_brackets) # link text = $2
\\] \\]
[ ]? # one optional space [ ]? # one optional space
(?:\\n[ ]*)? # one optional newline followed by spaces (?:\\n[ ]*)? # one optional newline followed by spaces
\\[ \\[
(.*?) # id = $3 (.*?) # id = $3
\\] \\]
) )
}xs", }xs",
'_DoAnchors_reference_callback', $text); '_DoAnchors_reference_callback', $text);
$text = preg_replace_callback("{ $text = preg_replace_callback("{
( # wrap whole match in $1 ( # wrap whole match in $1
\\[ \\[
($g_nested_brackets) # link text = $2 ($md_nested_brackets) # link text = $2
\\] \\]
\\( # literal paren \\( # literal paren
[ \\t]* [ \\t]*
(.+?) # href = $3 <?(.+?)>? # href = $3
[ \\t]* [ \\t]*
( # title = $4 ( # $4
(['\"]) # quote char = $5 (['\"]) # quote char = $5
.*? (.*?) # Title = $6
\\5 # matching quote \\5 # matching quote
)? # title is optional )? # title is optional
\\) \\)
) )
}xs", }xs",
'_DoAnchors_inline_callback', $text); '_DoAnchors_inline_callback', $text);
return $text; return $text;
} }
function _DoAnchors_reference_callback($matches) { function _DoAnchors_reference_callback($matches) {
global $g_urls, $g_titles; global $md_urls, $md_titles, $md_escape_table;
$result;
$whole_match = $matches[1]; $whole_match = $matches[1];
$link_text = $matches[2]; $link_text = $matches[2];
$link_id = strtolower($matches[3]); $link_id = strtolower($matches[3]);
@ -277,15 +329,18 @@ function _DoAnchors_reference_callback($matches) {
$link_id = strtolower($link_text); # for shortcut links like [this][]. $link_id = strtolower($link_text); # for shortcut links like [this][].
} }
if (isset($g_urls[$link_id])) { if (isset($md_urls[$link_id])) {
$url = $g_urls[$link_id]; $url = $md_urls[$link_id];
$url = str_replace(array('*', '_'), # We've got to encode these to avoid conflicting with italics/bold.
array('&#42;', '&#95;'), $url); $url = str_replace(array('*', '_'),
$result = "<a href='$url'"; array($md_escape_table['*'], $md_escape_table['_']),
if ( isset( $g_title[$link_id] ) ) { $url);
$title = $g_titles[$link_id]; $result = "<a href=\"$url\"";
if ( isset( $md_titles[$link_id] ) ) {
$title = $md_titles[$link_id];
$title = str_replace(array('*', '_'), $title = str_replace(array('*', '_'),
array('&#42;', '&#95;'), $title); array($md_escape_table['*'],
$md_escape_table['_']), $title);
$result .= " title=\"$title\""; $result .= " title=\"$title\"";
} }
$result .= ">$link_text</a>"; $result .= ">$link_text</a>";
@ -296,21 +351,25 @@ function _DoAnchors_reference_callback($matches) {
return $result; return $result;
} }
function _DoAnchors_inline_callback($matches) { function _DoAnchors_inline_callback($matches) {
$result; global $md_escape_table;
$whole_match = $matches[1]; $whole_match = $matches[1];
$link_text = $matches[2]; $link_text = $matches[2];
$url = $matches[3]; $url = $matches[3];
$title = $matches[4]; $title = $matches[6];
# We've got to encode these to avoid conflicting with italics/bold. # We've got to encode these to avoid conflicting with italics/bold.
$url = str_replace(array('*', '_'), $url = str_replace(array('*', '_'),
array('&#42;', '&#95;'), $url); array($md_escape_table['*'], $md_escape_table['_']),
$url);
$result = "<a href=\"$url\""; $result = "<a href=\"$url\"";
if ($title) { if (isset($title)) {
$title = str_replace(array('*', '_'), $title = str_replace('"', '&quot', $title);
array('&#42;', '&#95;'), $title); $title = str_replace(array('*', '_'),
$result .= " title=$title"; array($md_escape_table['*'], $md_escape_table['_']),
$title);
$result .= " title=\"$title\"";
} }
$result .= ">$link_text</a>"; $result .= ">$link_text</a>";
return $result; return $result;
@ -321,14 +380,14 @@ function _DoImages($text) {
$text = preg_replace_callback('{ $text = preg_replace_callback('{
( # wrap whole match in $1 ( # wrap whole match in $1
!\[ !\[
(.*?) # alt text = $2 (.*?) # alt text = $2
\] \]
[ ]? # one optional space [ ]? # one optional space
(?:\n[ ]*)? # one optional newline followed by spaces (?:\n[ ]*)? # one optional newline followed by spaces
\[ \[
(.*?) # id = $3 (.*?) # id = $3
\] \]
) )
@ -346,11 +405,11 @@ function _DoImages($text) {
\\] \\]
\\( # literal paren \\( # literal paren
[ \\t]* [ \\t]*
(\\S+) # src url = $3 <?(\S+?)>? # src url = $3
[ \\t]* [ \\t]*
( # title = $4 ( # $4
(['\"]) # quote char = $5 (['\"]) # quote char = $5
.*? (.*?) # title = $6
\\5 # matching quote \\5 # matching quote
[ \\t]* [ \\t]*
)? # title is optional )? # title is optional
@ -362,8 +421,7 @@ function _DoImages($text) {
return $text; return $text;
} }
function _DoImages_reference_callback($matches) { function _DoImages_reference_callback($matches) {
global $g_urls, $g_titles, $g_empty_element_suffix; global $md_urls, $md_titles, $md_empty_element_suffix, $md_escape_table;
$result;
$whole_match = $matches[1]; $whole_match = $matches[1];
$alt_text = $matches[2]; $alt_text = $matches[2];
$link_id = strtolower($matches[3]); $link_id = strtolower($matches[3]);
@ -371,43 +429,55 @@ function _DoImages_reference_callback($matches) {
if ($link_id == "") { if ($link_id == "") {
$link_id = strtolower($alt_text); # for shortcut links like ![this][]. $link_id = strtolower($alt_text); # for shortcut links like ![this][].
} }
if (isset($g_urls[$link_id])) { $alt_text = str_replace('"', '&quot;', $alt_text);
$url = $g_urls[$link_id]; if (isset($md_urls[$link_id])) {
$url = str_replace(array('*', '_'), $url = $md_urls[$link_id];
array('&#42;', '&#95;'), $url); # We've got to encode these to avoid conflicting with italics/bold.
$url = str_replace(array('*', '_'),
array($md_escape_table['*'], $md_escape_table['_']),
$url);
$result = "<img src=\"$url\" alt=\"$alt_text\""; $result = "<img src=\"$url\" alt=\"$alt_text\"";
if (isset($g_titles[$link_id])) { if (isset($md_titles[$link_id])) {
$title = $g_titles[$link_id]; $title = $md_titles[$link_id];
$title = str_replace(array('*', '_'), $title = str_replace(array('*', '_'),
array('&#42;', '&#95;'), $title); array($md_escape_table['*'],
$md_escape_table['_']), $title);
$result .= " title=\"$title\""; $result .= " title=\"$title\"";
} }
$result .= $g_empty_element_suffix; $result .= $md_empty_element_suffix;
} }
else { else {
# If there's no such link ID, leave intact:
$result = $whole_match; $result = $whole_match;
} }
return $result; return $result;
} }
function _DoImages_inline_callback($matches) { function _DoImages_inline_callback($matches) {
global $g_empty_element_suffix; global $md_empty_element_suffix, $md_escape_table;
$result;
$whole_match = $matches[1]; $whole_match = $matches[1];
$alt_text = $matches[2]; $alt_text = $matches[2];
$url = $matches[3]; $url = $matches[3];
$title = $matches[4]; $title = '';
if (isset($matches[6])) {
$title = $matches[6];
}
$url = str_replace(array('*', '_'), $alt_text = str_replace('"', '&quot;', $alt_text);
array('&#42;', '&#95;'), $url); $title = str_replace('"', '&quot;', $title);
# We've got to encode these to avoid conflicting with italics/bold.
$url = str_replace(array('*', '_'),
array($md_escape_table['*'], $md_escape_table['_']),
$url);
$result = "<img src=\"$url\" alt=\"$alt_text\""; $result = "<img src=\"$url\" alt=\"$alt_text\"";
if (isset($title)) { if (isset($title)) {
$title = str_replace(array('*', '_'), $title = str_replace(array('*', '_'),
array('&#42;', '&#95;'), $title); array($md_escape_table['*'], $md_escape_table['_']),
$result .= " title=$title"; # $title already quoted $title);
$result .= " title=\"$title\""; # $title already quoted
} }
$result .= $g_empty_element_suffix; $result .= $md_empty_element_suffix;
return $result; return $result;
} }
@ -437,23 +507,34 @@ function _DoHeaders($text) {
function _DoLists($text) { function _DoLists($text) {
global $g_tab_width; #
$less_than_tab = $g_tab_width - 1; # Form HTML ordered (numbered) and unordered (bulleted) lists.
#
global $md_tab_width;
$less_than_tab = $md_tab_width - 1;
# Re-usable patterns to match list item bullets and number markers:
$marker_ul = '[*+-]';
$marker_ol = '\d+[.]';
$marker_any = "(?:$marker_ul|$marker_ol)";
$text = preg_replace_callback("{ $text = preg_replace_callback("{
( ( # $1
( ( # $2
^[ ]{0,$less_than_tab} ^[ ]{0,$less_than_tab}
(\\*|\\d+[.]) ($marker_any) # $3 - first list item marker
[ \\t]+ [ \\t]+
) )
(?s:.+?) (?s:.+?)
( ( # $4
\\z \\z
| |
\\n{2,} \\n{2,}
(?=\\S) (?=\\S)
(?![ \\t]* (\\*|\\d+[.]) [ \\t]+) (?! # Negative lookahead for another list item marker
[ \\t]*
{$marker_any}[ \\t]+
)
) )
) )
}xm", }xm",
@ -462,26 +543,33 @@ function _DoLists($text) {
return $text; return $text;
} }
function _DoLists_callback($matches) { function _DoLists_callback($matches) {
$list_type = ($matches[3] == "*") ? "ul" : "ol"; # Re-usable patterns to match list item bullets and number markers:
$marker_ul = '[*+-]';
$marker_ol = '\d+[.]';
$marker_any = "(?:$marker_ul|$marker_ol)";
$list = $matches[1]; $list = $matches[1];
$list_type = preg_match('/[*+-]/', $matches[3]) ? "ul" : "ol";
# Turn double returns into triple returns, so that we can make a
# paragraph for the last item in a list, if necessary:
$list = preg_replace("/\n{2,}/", "\n\n\n", $list); $list = preg_replace("/\n{2,}/", "\n\n\n", $list);
$result = _ProcessListItems($list); $result = _ProcessListItems($list, $marker_any);
$result = "<$list_type>\n" . $result . "</$list_type>\n"; $result = "<$list_type>\n" . $result . "</$list_type>\n\n";
return $result; return $result;
} }
function _ProcessListItems($list_str) { function _ProcessListItems($list_str, $marker_any) {
# trim trailing blank lines: # trim trailing blank lines:
$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str); $list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str);
$list_str = preg_replace_callback('{ $list_str = preg_replace_callback('{
(\n)? # leading line = $1 (\n)? # leading line = $1
(^[ \t]*) # leading whitespace = $2 (^[ \t]*) # leading whitespace = $2
(\*|\d+[.]) [ \t]+ # list marker = $3 ('.$marker_any.') [ \t]+ # list marker = $3
((?s:.+?) # list item text = $4 ((?s:.+?) # list item text = $4
(\n{1,2})) (\n{1,2}))
(?= \n* (\z | \2 (\*|\d+[.]) [ \t]+)) (?= \n* (\z | \2 ('.$marker_any.') [ \t]+))
}xm', }xm',
'_ProcessListItems_callback', $list_str); '_ProcessListItems_callback', $list_str);
@ -508,41 +596,30 @@ function _ProcessListItems_callback($matches) {
function _DoCodeBlocks($text) { function _DoCodeBlocks($text) {
global $g_tab_width; global $md_tab_width;
$text = preg_replace_callback("{ $text = preg_replace_callback("{
(.?) # $1 = preceding character (?:\\n\\n|\\A)
(:) # $2 = colon delimiter ( # $1 = the code block -- one or more lines, starting with a space/tab
(\\n+) # $3 = newlines after colon
( # $4 = the code block -- one or more lines, starting with a space/tab
(?: (?:
(?:[ ]\{$g_tab_width} | \\t) # Lines must start with a tab or a tab-width of spaces (?:[ ]\{$md_tab_width} | \\t) # Lines must start with a tab or a tab-width of spaces
.*\\n+ .*\\n+
)+ )+
) )
((?=^[ ]{0,$g_tab_width}\\S)|\\Z) # Lookahead for non-space at line-start, or end of doc ((?=^[ ]{0,$md_tab_width}\\S)|\\Z) # Lookahead for non-space at line-start, or end of doc
}xm", }xm",
'_DoCodeBlocks_callback', $text); '_DoCodeBlocks_callback', $text);
return $text; return $text;
} }
function _DoCodeBlocks_callback($matches) { function _DoCodeBlocks_callback($matches) {
$prevchar = $matches[1]; $codeblock = $matches[1];
$newlines = $matches[2];
$codeblock = $matches[4];
$result; # return value
$prefix = "";
if (!(preg_match('/\s/', $prevchar) || ($prevchar == ""))) {
$prefix = "$prevchar:";
}
$codeblock = _EncodeCode(_Outdent($codeblock)); $codeblock = _EncodeCode(_Outdent($codeblock));
$codeblock = _Detab($codeblock); $codeblock = _Detab($codeblock);
# trim leading newlines and trailing whitespace # trim leading newlines and trailing whitespace
$codeblock = preg_replace(array('/\A\n+/', '/\s+\z/'), '', $codeblock); $codeblock = preg_replace(array('/\A\n+/', '/\s+\z/'), '', $codeblock);
$result = $prefix . "\n\n<pre><code>" . $codeblock . "\n</code></pre>\n\n"; $result = "\n\n<pre><code>" . $codeblock . "\n</code></pre>\n\n";
return $result; return $result;
} }
@ -550,8 +627,8 @@ function _DoCodeBlocks_callback($matches) {
function _DoCodeSpans($text) { function _DoCodeSpans($text) {
$text = preg_replace_callback("@ $text = preg_replace_callback("@
(`+) # Opening run of ` (`+) # $1 = Opening run of `
(.+?) # the code block (.+?) # $2 = The code block
(?<!`) (?<!`)
\\1 \\1
(?!`) (?!`)
@ -570,20 +647,15 @@ function _DoCodeSpans_callback($matches) {
function _EncodeCode($_) { function _EncodeCode($_) {
global $md_escape_table;
global $g_escape_table;
# Encode all ampersands; HTML entities are not
# entities within a Markdown code span.
$_ = str_replace('&', '&amp;', $_); $_ = str_replace('&', '&amp;', $_);
# Do the angle bracket song and dance:
$_ = str_replace(array('<', '>'), $_ = str_replace(array('<', '>'),
array('&lt;', '&gt;'), $_); array('&lt;', '&gt;'), $_);
# Now, escape characters that are magic in Markdown: $_ = str_replace(array_keys($md_escape_table),
$_ = str_replace(array_keys($g_escape_table), array_values($md_escape_table), $_);
array_values($g_escape_table), $_);
return $_; return $_;
} }
@ -618,26 +690,35 @@ function _DoBlockQuotes($text) {
} }
function _DoBlockQuotes_callback($matches) { function _DoBlockQuotes_callback($matches) {
$bq = $matches[1]; $bq = $matches[1];
$bq = preg_replace('/^[ \t]*>[ \t]?/m', '', $bq); # trim one level of quoting - trim whitespace-only lines
$bq = preg_replace(array('/^[ \t]*>[ \t]?/m', '/^[ \t]+$/m'), '', $bq);
$bq = _RunBlockGamut($bq); # recurse $bq = _RunBlockGamut($bq); # recurse
$bq = preg_replace('/^/m', "\t", $bq);
$bq = preg_replace('/^/m', " ", $bq);
# These leading spaces screw with <pre> content, so we need to fix that:
$bq = preg_replace_callback('{(\s*<pre>.+?</pre>)}sx',
'_DoBlockQuotes_callback2', $bq);
return "<blockquote>\n$bq\n</blockquote>\n\n"; return "<blockquote>\n$bq\n</blockquote>\n\n";
} }
function _DoBlockQuotes_callback2($matches) {
$pre = $matches[1];
$pre = preg_replace('/^ /m', '', $pre);
return $pre;
}
function _FormParagraphs($text) { function _FormParagraphs($text) {
global $g_html_blocks; global $md_html_blocks;
# Strip leading and trailing lines: # Strip leading and trailing lines:
$text = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $text); $text = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $text);
$grafs = preg_split('/\n{2,}/', $text); $grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
$count = count($graph); $count = count($grafs);
foreach ($grafs as $key => $value) { foreach ($grafs as $key => $value) {
if (!isset( $g_html_blocks[$value] )) { if (!isset( $md_html_blocks[$value] )) {
$value = _RunSpanGamut($value); $value = _RunSpanGamut($value);
$value = preg_replace('/^([ \t]*)/', '<p>', $value); $value = preg_replace('/^([ \t]*)/', '<p>', $value);
$value .= "</p>"; $value .= "</p>";
@ -645,10 +726,9 @@ function _FormParagraphs($text) {
} }
} }
foreach ($grafs as $key => $value) { foreach ($grafs as $key => $value) {
if (isset( $g_html_blocks[$value] )) { if (isset( $md_html_blocks[$value] )) {
$grafs[$key] = $g_html_blocks[$value]; $grafs[$key] = $md_html_blocks[$value];
} }
} }
@ -657,7 +737,7 @@ function _FormParagraphs($text) {
function _EncodeAmpsAndAngles($text) { function _EncodeAmpsAndAngles($text) {
$text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w{1,8});)/', $text = preg_replace('/&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)/',
'&amp;', $text);; '&amp;', $text);;
# Encode naked <'s # Encode naked <'s
@ -668,17 +748,17 @@ function _EncodeAmpsAndAngles($text) {
function _EncodeBackslashEscapes($text) { function _EncodeBackslashEscapes($text) {
global $g_escape_table, $g_backslash_escape_table; global $md_escape_table, $md_backslash_escape_table;
# Must process escaped backslashes first. # Must process escaped backslashes first.
return str_replace(array_keys($g_backslash_escape_table), return str_replace(array_keys($md_backslash_escape_table),
array_values($g_backslash_escape_table), $text); array_values($md_backslash_escape_table), $text);
} }
function _DoAutoLinks($text) { function _DoAutoLinks($text) {
$text = preg_replace("!<((https?|ftp):[^'\">\\s]+)>!", $text = preg_replace("!<((https?|ftp):[^'\">\\s]+)>!",
'<a href="\1">\1</a>', $text); '<a href="\1">\1</a>', $text);
# Email addresses: <address@domain.foo> # Email addresses: <address@domain.foo>
$text = preg_replace('{ $text = preg_replace('{
< <
@ -691,7 +771,7 @@ function _DoAutoLinks($text) {
}exi', }exi',
"_EncodeEmailAddress(_UnescapeSpecialChars(_UnslashQuotes('\\1')))", "_EncodeEmailAddress(_UnescapeSpecialChars(_UnslashQuotes('\\1')))",
$text); $text);
return $text; return $text;
} }
@ -700,10 +780,12 @@ function _EncodeEmailAddress($addr) {
$addr = "mailto:" . $addr; $addr = "mailto:" . $addr;
$length = strlen($addr); $length = strlen($addr);
# leave ':' alone (to spot mailto: later)
$addr = preg_replace_callback('/([^\:])/', $addr = preg_replace_callback('/([^\:])/',
'_EncodeEmailAddress_callback', $addr); '_EncodeEmailAddress_callback', $addr);
$addr = "<a href=\"$addr\">$addr</a>"; $addr = "<a href=\"$addr\">$addr</a>";
# strip the mailto: from the visible part
$addr = preg_replace('/">.+?:/', '">', $addr); $addr = preg_replace('/">.+?:/', '">', $addr);
return $addr; return $addr;
@ -711,6 +793,8 @@ function _EncodeEmailAddress($addr) {
function _EncodeEmailAddress_callback($matches) { function _EncodeEmailAddress_callback($matches) {
$char = $matches[1]; $char = $matches[1];
$r = rand(0, 100); $r = rand(0, 100);
# roughly 10% raw, 45% hex, 45% dec
# '@' *must* be encoded. I insist.
if ($r > 90 && $char != '@') return $char; if ($r > 90 && $char != '@') return $char;
if ($r < 45) return '&#x'.dechex(ord($char)).';'; if ($r < 45) return '&#x'.dechex(ord($char)).';';
return '&#'.ord($char).';'; return '&#'.ord($char).';';
@ -718,57 +802,49 @@ function _EncodeEmailAddress_callback($matches) {
function _UnescapeSpecialChars($text) { function _UnescapeSpecialChars($text) {
global $g_escape_table; global $md_escape_table;
return str_replace(array_values($g_escape_table), return str_replace(array_values($md_escape_table),
array_keys($g_escape_table), $text); array_keys($md_escape_table), $text);
} }
function _TokenizeHTML($str) { if (!function_exists('_TokenizeHTML')) {
$pos = 0; function _TokenizeHTML($str) {
$len = strlen($str); $index = 0;
$tokens = array(); $tokens = array();
$depth = 6; $depth = 6;
$nested_tags = str_repeat('(?:<[a-z\/!$](?:[^<>]|',$depth) $nested_tags = str_repeat('(?:<[a-z\/!$](?:[^<>]|',$depth)
.str_repeat(')*>)', $depth); .str_repeat(')*>)', $depth);
$match = "(?s:<!(--.*?--\s*)+>)|". # comment $match = "(?s:<!(?:--.*?--\s*)+>)|". # comment
"(?s:<\?.*?\?>)|". # processing instruction "(?s:<\?.*?\?>)|". # processing instruction
"$nested_tags"; # nested tags "$nested_tags"; # nested tags
preg_match_all("/($match)/", $str, $matches,
PREG_SET_ORDER | PREG_OFFSET_CAPTURE);
foreach ($matches as $element) { $parts = preg_split("/($match)/", $str, -1, PREG_SPLIT_DELIM_CAPTURE);
$whole_tag = $element[0][0];
$tag_start = $element[0][1]; foreach ($parts as $part) {
$sec_start = $tag_start + strlen($whole_tag); if (++$index % 2 && $part != '')
if ($pos < $tag_start) { array_push($tokens, array('text', $part));
array_push($tokens, array('text', else
substr($str, $pos, $tag_start - $pos))); array_push($tokens, array('tag', $part));
} }
array_push($tokens, array('tag', $whole_tag));
$pos = $sec_start; return $tokens;
} }
if ($pos < $len)
array_push($tokens, array('text',
substr($str, $pos, $len - $pos)));
return $tokens;
} }
function _Outdent($text) { function _Outdent($text) {
global $g_tab_width; global $md_tab_width;
return preg_replace("/^(\\t|[ ]{1,$g_tab_width})/m", "", $text); return preg_replace("/^(\\t|[ ]{1,$md_tab_width})/m", "", $text);
} }
function _Detab($text) { function _Detab($text) {
global $g_tab_width; global $md_tab_width;
$text = preg_replace( $text = preg_replace(
"/(.*?)\t/e", "/(.*?)\t/e",
"'\\1'.str_repeat(' ', $g_tab_width - strlen('\\1') % $g_tab_width)", "'\\1'.str_repeat(' ', $md_tab_width - strlen('\\1') % $md_tab_width)",
$text); $text);
return $text; return $text;
} }
@ -778,13 +854,4 @@ function _UnslashQuotes($text) {
return str_replace('\"', '"', $text); return str_replace('\"', '"', $text);
} }
// And now for the filters
remove_filter('the_content', 'wpautop');
remove_filter('the_excerpt', 'wpautop');
remove_filter('comment_text', 'wpautop');
add_filter('the_content', 'Markdown');
add_filter('the_excerpt', 'Markdown');
remove_filter('comment_text', 'Markdown');
?> ?>