-
- #
- List item two - #
- List item three - #
-
+ *
- List item two + *
- List item three + *
s around - # "paragraphs" that are wrapped in non-block-level tags, such as anchors, - # phrase emphasis, and spans. The list of tags we're looking for is - # hard-coded: - # - # * List "a" is made of tags which can be both inline or block-level. - # These will be treated block-level when the start tag is alone on - # its line, otherwise they're not matched here and will be taken as - # inline later. - # * List "b" is made of tags which are always block-level; - # + /** + * Hashify HTML blocks: + * + * We only want to do this for block-level HTML tags, such as headers, + * lists, and tables. That's because we still want to wrap
s around + * "paragraphs" that are wrapped in non-block-level tags, such as + * anchors, phrase emphasis, and spans. The list of tags we're looking + * for is hard-coded: + * + * * List "a" is made of tags which can be both inline or block-level. + * These will be treated block-level when the start tag is alone on + * its line, otherwise they're not matched here and will be taken as + * inline later. + * * List "b" is made of tags which are always block-level; + */ $block_tags_a_re = 'ins|del'; $block_tags_b_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|'. 'script|noscript|style|form|fieldset|iframe|math|svg|'. 'article|section|nav|aside|hgroup|header|footer|'. 'figure'; - # Regular expression for the content of a block tag. + // Regular expression for the content of a block tag. $nested_tags_level = 4; $attr = ' (?> # optional tag attributes @@ -281,7 +370,7 @@ class Markdown implements MarkdownInterface { | \'[^\']*\' # text inside single quotes (tolerate ">") )* - )? + )? '; $content = str_repeat(' @@ -293,29 +382,32 @@ class Markdown implements MarkdownInterface { (?> /> | - >', $nested_tags_level). # end of opening tag - '.*?'. # last level nested tag content + >', $nested_tags_level). // end of opening tag + '.*?'. // last level nested tag content str_repeat(' # closing nested tag ) - | + | <(?!/\2\s*> # other tags with a different name ) )*', $nested_tags_level); $content2 = str_replace('\2', '\3', $content); - # First, look for nested blocks, e.g.: - #
- #
- #
- # The outermost tags must start at the left margin for this to match, and
- # the inner nested divs must be indented.
- # We need to do this before the next, more liberal match, because the next
- # match will start at the first `
- # tags for inner block must be indented.
- #
- # ` and stop at the first `
`.
+ /**
+ * First, look for nested blocks, e.g.:
+ *
+ *
+ *
+ * The outermost tags must start at the left margin for this to match,
+ * and the inner nested divs must be indented.
+ * We need to do this before the next, more liberal match, because the
+ * next match will start at the first `
+ * tags for inner block must be indented.
+ *
+ * ` and stop at the
+ * first `
`.
+ */
$text = preg_replace_callback('{(?>
(?>
(?<=\n) # Starting on its own line
@@ -324,9 +416,9 @@ class Markdown implements MarkdownInterface {
)
( # save in $1
- # Match from `\n. It was easier to make a special + + | # Special case just for
. It was easier to make a special # case than to make the other regex more complicated. - + [ ]{0,'.$less_than_tab.'} <(hr) # start tag = $2 '.$attr.' # attributes /?> # the matching end tag [ ]* (?=\n{2,}|\Z) # followed by a blank line or end of document - + | # Special case for standalone HTML comments: - + [ ]{0,'.$less_than_tab.'} (?s: ) [ ]* (?=\n{2,}|\Z) # followed by a blank line or end of document - + | # PHP and ASP-style processor instructions (hashBlock($text); return "\n\n$key\n\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. + * + * @param string $text + * @param string $boundary + * @return string + */ protected function hashPart($text, $boundary = 'X') { - # - # 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. - # - # Swap back any tag hash found in $text so we do not have to `unhash` - # multiple times at the end. + // 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. + + // 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. + return $key; // String that will replace the tag. } - + /** + * Shortcut function for hashPart with block-level boundaries. + * @param string $text + * @return string + */ protected function hashBlock($text) { - # - # Shortcut function for hashPart with block-level boundaries. - # return $this->hashPart($text, 'B'); } - + /** + * Define the block gamut - these are all the transformations that form + * block-level tags like paragraphs, headers, and list items. + * @var array + */ protected $block_gamut = array( - # - # These are all the transformations that form block-level - # tags like paragraphs, headers, and list items. - # "doHeaders" => 10, "doHorizontalRules" => 20, - "doLists" => 40, "doCodeBlocks" => 50, "doBlockQuotes" => 60, - ); + ); + /** + * Run block gamut tranformations. + * + * We need to escape raw HTML in Markdown source before doing anything + * else. This need to be done for each block, and not only at the + * begining in the Markdown function since hashed blocks can be part of + * list items and could have been indented. Indented blocks would have + * been seen as a code block in a previous pass of hashHTMLBlocks. + * + * @param string $text + * @return string + */ protected function runBlockGamut($text) { - # - # Run block gamut tranformations. - # - # We need to escape raw HTML in Markdown source before doing anything - # else. This need to be done for each block, and not only at the - # begining in the Markdown function since hashed blocks can be part of - # list items and could have been indented. Indented blocks would have - # been seen as a code block in a previous pass of hashHTMLBlocks. $text = $this->hashHTMLBlocks($text); - return $this->runBasicBlockGamut($text); } - + + /** + * Run block gamut tranformations, without hashing HTML blocks. This is + * useful when HTML blocks are known to be already hashed, like in the first + * whole-document pass. + * + * @param string $text + * @return string + */ protected function runBasicBlockGamut($text) { - # - # Run block gamut tranformations, without hashing HTML blocks. This is - # useful when HTML blocks are known to be already hashed, like in the first - # whole-document pass. - # + foreach ($this->block_gamut as $method => $priority) { $text = $this->$method($text); } - - # Finally form paragraph and restore hashed blocks. + + // Finally form paragraph and restore hashed blocks. $text = $this->formParagraphs($text); return $text; } - - + + /** + * Convert horizontal rules + * @param string $text + * @return string + */ protected function doHorizontalRules($text) { - # Do Horizontal Rules: return preg_replace( '{ ^[ ]{0,3} # Leading space @@ -477,67 +589,82 @@ class Markdown implements MarkdownInterface { [ ]* # Tailing spaces $ # End of line. }mx', - "\n".$this->hashBlock("
` blocks.
+ * @param string $text
+ * @return string
+ */
protected function doCodeBlocks($text) {
- #
- # Process Markdown `` blocks.
- #
$text = preg_replace_callback('{
(?:\n\n|\A\n?)
( # $1 = the code block -- one or more lines, starting with a space/tab
@@ -1023,11 +1207,17 @@ class Markdown implements MarkdownInterface {
return $text;
}
+
+ /**
+ * Code block parsing callback
+ * @param array $matches
+ * @return string
+ */
protected function _doCodeBlocks_callback($matches) {
$codeblock = $matches[1];
$codeblock = $this->outdent($codeblock);
- if ($this->code_block_content_func) {
+ if (is_callable($this->code_block_content_func)) {
$codeblock = call_user_func($this->code_block_content_func, $codeblock, "");
} else {
$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES);
@@ -1037,96 +1227,126 @@ class Markdown implements MarkdownInterface {
$codeblock = preg_replace('/\A\n+|\n+\z/', '', $codeblock);
$codeblock = "$codeblock\n
";
- return "\n\n".$this->hashBlock($codeblock)."\n\n";
+ return "\n\n" . $this->hashBlock($codeblock) . "\n\n";
}
-
+ /**
+ * Create a code span markup for $code. Called from handleSpanToken.
+ * @param string $code
+ * @return string
+ */
protected function makeCodeSpan($code) {
- #
- # Create a code span markup for $code. Called from handleSpanToken.
- #
- $code = htmlspecialchars(trim($code), ENT_NOQUOTES);
+ if (is_callable($this->code_span_content_func)) {
+ $code = call_user_func($this->code_span_content_func, $code);
+ } else {
+ $code = htmlspecialchars(trim($code), ENT_NOQUOTES);
+ }
return $this->hashPart("$code");
}
-
+ /**
+ * Define the emphasis operators with their regex matches
+ * @var array
+ */
protected $em_relist = array(
'' => '(?:(? '(? '(? '(?:(? '(? '(? '(?:(? '(? '(?em_relist as $em => $em_re) {
foreach ($this->strong_relist as $strong => $strong_re) {
- # Construct list of allowed token expressions.
+ // 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) .')}';
+
+ // Construct master expression from list.
+ $token_re = '{(' . implode('|', $token_relist) . ')}';
$this->em_strong_prepared_relist["$em$strong"] = $token_re;
}
}
}
-
+
+ /**
+ * Convert Markdown italics (emphasis) and bold (strong) to HTML
+ * @param string $text
+ * @return string
+ */
protected function doItalicsAndBold($text) {
+ if ($this->in_emphasis_processing) {
+ return $text; // avoid reentrency
+ }
+ $this->in_emphasis_processing = true;
+
$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.
- #
+ // 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.
- #
+
+ // 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.
+ // 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 ($tree_char_em) {
- # Reached closing marker while inside a three-char emphasis.
+ // Reached closing marker while inside a three-char emphasis.
if ($token_len == 3) {
- # Three-char closing marker, close em and strong.
+ // Three-char closing marker, close em and strong.
array_shift($token_stack);
$span = array_shift($text_stack);
$span = $this->runSpanGamut($span);
@@ -1135,21 +1355,21 @@ class Markdown implements MarkdownInterface {
$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);
+ // 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";
$text_stack[0] = $this->hashPart($span);
- $$tag = ''; # $$tag stands for $em or $strong
+ $$tag = ''; // $$tag stands for $em or $strong
}
$tree_char_em = false;
} else if ($token_len == 3) {
if ($em) {
- # Reached closing marker for both em and strong.
- # Closing strong marker:
+ // Reached closing marker for both em and strong.
+ // Closing strong marker:
for ($i = 0; $i < 2; ++$i) {
$shifted_token = array_shift($token_stack);
$tag = strlen($shifted_token) == 2 ? "strong" : "em";
@@ -1157,12 +1377,12 @@ class Markdown implements MarkdownInterface {
$span = $this->runSpanGamut($span);
$span = "<$tag>$span";
$text_stack[0] .= $this->hashPart($span);
- $$tag = ''; # $$tag stands for $em or $strong
+ $$tag = ''; // $$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};
+ // 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, '');
@@ -1170,12 +1390,13 @@ class Markdown implements MarkdownInterface {
}
} else if ($token_len == 2) {
if ($strong) {
- # Unwind any dangling emphasis marker:
+ // 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);
+ $em = '';
}
- # Closing strong marker:
+ // Closing strong marker:
array_shift($token_stack);
$span = array_shift($text_stack);
$span = $this->runSpanGamut($span);
@@ -1188,10 +1409,10 @@ class Markdown implements MarkdownInterface {
$strong = $token;
}
} else {
- # Here $token_len == 1
+ // Here $token_len == 1
if ($em) {
if (strlen($token_stack[0]) == 1) {
- # Closing emphasis marker:
+ // Closing emphasis marker:
array_shift($token_stack);
$span = array_shift($text_stack);
$span = $this->runSpanGamut($span);
@@ -1208,10 +1429,15 @@ class Markdown implements MarkdownInterface {
}
}
}
+ $this->in_emphasis_processing = false;
return $text_stack[0];
}
-
+ /**
+ * Parse Markdown blockquotes to HTML
+ * @param string $text
+ * @return string
+ */
protected function doBlockQuotes($text) {
$text = preg_replace_callback('/
( # Wrap whole match in $1
@@ -1227,51 +1453,64 @@ class Markdown implements MarkdownInterface {
return $text;
}
+
+ /**
+ * Blockquote parsing callback
+ * @param array $matches
+ * @return string
+ */
protected function _doBlockQuotes_callback($matches) {
$bq = $matches[1];
- # trim one level of quoting - trim whitespace-only lines
+ // trim one level of quoting - trim whitespace-only lines
$bq = preg_replace('/^[ ]*>[ ]?|^[ ]+$/m', '', $bq);
- $bq = $this->runBlockGamut($bq); # recurse
+ $bq = $this->runBlockGamut($bq); // recurse
$bq = preg_replace('/^/m', " ", $bq);
- # These leading spaces cause problem with content,
- # so we need to fix that:
- $bq = preg_replace_callback('{(\s*.+?
)}sx',
+ // These leading spaces cause problem with content,
+ // so we need to fix that:
+ $bq = preg_replace_callback('{(\s*.+?
)}sx',
array($this, '_doBlockQuotes_callback2'), $bq);
- return "\n". $this->hashBlock("\n$bq\n
")."\n\n";
+ return "\n" . $this->hashBlock("\n$bq\n
") . "\n\n";
}
+
+ /**
+ * Blockquote parsing callback
+ * @param array $matches
+ * @return string
+ */
protected function _doBlockQuotes_callback2($matches) {
$pre = $matches[1];
$pre = preg_replace('/^ /m', '', $pre);
return $pre;
}
-
- protected function formParagraphs($text) {
- #
- # Params:
- # $text - string to process with html tags
- #
- # Strip leading and trailing lines:
+ /**
+ * Parse paragraphs
+ *
+ * @param string $text String to process in paragraphs
+ * @param boolean $wrap_in_p Whether paragraphs should be wrapped in
tags
+ * @return string
+ */
+ protected function formParagraphs($text, $wrap_in_p = true) {
+ // Strip leading and trailing lines:
$text = preg_replace('/\A\n+|\n+\z/', '', $text);
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
- #
- # Wrap
tags and unhashify HTML blocks
- #
+ // Wrap
tags and unhashify HTML blocks
foreach ($grafs as $key => $value) {
if (!preg_match('/^B\x1A[0-9]+B$/', $value)) {
- # Is a paragraph.
+ // Is a paragraph.
$value = $this->runSpanGamut($value);
- $value = preg_replace('/^([ ]*)/', "
", $value);
- $value .= "
";
+ if ($wrap_in_p) {
+ $value = preg_replace('/^([ ]*)/', "", $value);
+ $value .= "
";
+ }
$grafs[$key] = $this->unhash($value);
- }
- else {
- # Is a block.
- # Modify elements of @grafs in-place...
+ } else {
+ // Is a block.
+ // Modify elements of @grafs in-place...
$graf = $value;
$block = $this->html_hashes[$graf];
$graf = $block;
@@ -1296,11 +1535,11 @@ class Markdown implements MarkdownInterface {
// {
// list(, $div_open, , $div_content, $div_close) = $matches;
//
-// # We can't call Markdown(), because that resets the hash;
-// # that initialization code should be pulled into its own sub, though.
+// // We can't call Markdown(), because that resets the hash;
+// // that initialization code should be pulled into its own sub, though.
// $div_content = $this->hashHTMLBlocks($div_content);
-//
-// # Run document gamut methods on the content.
+//
+// // Run document gamut methods on the content.
// foreach ($this->document_gamut as $method => $priority) {
// $div_content = $this->$method($div_content);
// }
@@ -1317,71 +1556,78 @@ class Markdown implements MarkdownInterface {
return implode("\n\n", $grafs);
}
-
+ /**
+ * Encode text for a double-quoted HTML attribute. This function
+ * is *not* suitable for attributes enclosed in single quotes.
+ * @param string $text
+ * @return string
+ */
protected function encodeAttribute($text) {
- #
- # Encode text for a double-quoted HTML attribute. This function
- # is *not* suitable for attributes enclosed in single quotes.
- #
$text = $this->encodeAmpsAndAngles($text);
$text = str_replace('"', '"', $text);
return $text;
}
-
+ /**
+ * Encode text for a double-quoted HTML attribute containing a URL,
+ * applying the URL filter if set. Also generates the textual
+ * representation for the URL (removing mailto: or tel:) storing it in $text.
+ * This function is *not* suitable for attributes enclosed in single quotes.
+ *
+ * @param string $url
+ * @param string $text Passed by reference
+ * @return string URL
+ */
protected function encodeURLAttribute($url, &$text = null) {
- #
- # Encode text for a double-quoted HTML attribute containing a URL,
- # applying the URL filter if set. Also generates the textual
- # representation for the URL (removing mailto: or tel:) storing it in $text.
- # This function is *not* suitable for attributes enclosed in single quotes.
- #
- if ($this->url_filter_func)
+ if (is_callable($this->url_filter_func)) {
$url = call_user_func($this->url_filter_func, $url);
+ }
- if (preg_match('{^mailto:}i', $url))
+ if (preg_match('{^mailto:}i', $url)) {
$url = $this->encodeEntityObfuscatedAttribute($url, $text, 7);
- else if (preg_match('{^tel:}i', $url))
- {
+ } else if (preg_match('{^tel:}i', $url)) {
$url = $this->encodeAttribute($url);
$text = substr($url, 4);
- }
- else
- {
+ } else {
$url = $this->encodeAttribute($url);
$text = $url;
}
return $url;
}
-
-
+
+ /**
+ * 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.
+ * @param string $text
+ * @return string
+ */
protected function encodeAmpsAndAngles($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.
- #
if ($this->no_entities) {
$text = str_replace('&', '&', $text);
} else {
- # Ampersand-encoding based entirely on Nat Irons's Amputator
- # MT plugin:
+ // Email addresses:
$text = preg_replace_callback('{
<
(?:mailto:)?
@@ -1404,11 +1650,23 @@ class Markdown implements MarkdownInterface {
return $text;
}
+
+ /**
+ * Parse URL callback
+ * @param array $matches
+ * @return string
+ */
protected function _doAutoLinks_url_callback($matches) {
$url = $this->encodeURLAttribute($matches[1], $text);
$link = "$text";
return $this->hashPart($link);
}
+
+ /**
+ * Parse email address callback
+ * @param array $matches
+ * @return string
+ */
protected function _doAutoLinks_email_callback($matches) {
$addr = $matches[1];
$url = $this->encodeURLAttribute("mailto:$addr", $text);
@@ -1416,42 +1674,52 @@ class Markdown implements MarkdownInterface {
return $this->hashPart($link);
}
-
+ /**
+ * Input: some text to obfuscate, e.g. "mailto:foo@example.com"
+ *
+ * Output: the same text but with most characters encoded as either a
+ * decimal or hex entity, in the hopes of foiling most address
+ * harvesting spam bots. E.g.:
+ *
+ * mailto:foo
+ * @example.co
+ * m
+ *
+ * Note: the additional output $tail is assigned the same value as the
+ * ouput, minus the number of characters specified by $head_length.
+ *
+ * Based by a filter by Matthew Wickline, posted to BBEdit-Talk.
+ * With some optimizations by Milian Wolff. Forced encoding of HTML
+ * attribute special characters by Allan Odgaard.
+ *
+ * @param string $text
+ * @param string $tail Passed by reference
+ * @param integer $head_length
+ * @return string
+ */
protected function encodeEntityObfuscatedAttribute($text, &$tail = null, $head_length = 0) {
- #
- # Input: some text to obfuscate, e.g. "mailto:foo@example.com"
- #
- # Output: the same text but with most characters encoded as either a
- # decimal or hex entity, in the hopes of foiling most address
- # harvesting spam bots. E.g.:
- #
- # mailto:foo
- # @example.co
- # m
- #
- # Note: the additional output $tail is assigned the same value as the
- # ouput, minus the number of characters specified by $head_length.
- #
- # Based by a filter by Matthew Wickline, posted to BBEdit-Talk.
- # With some optimizations by Milian Wolff. Forced encoding of HTML
- # attribute special characters by Allan Odgaard.
- #
- if ($text == "") return $tail = "";
+ if ($text == "") {
+ return $tail = "";
+ }
$chars = preg_split('/(? $char) {
$ord = ord($char);
- # Ignore non-ascii chars.
+ // Ignore non-ascii chars.
if ($ord < 128) {
- $r = ($seed * (1 + $key)) % 100; # Pseudo-random function.
- # roughly 10% raw, 45% hex, 45% dec
- # '@' *must* be encoded. I insist.
- # '"' and '>' have to be encoded inside the attribute
- if ($r > 90 && strpos('@"&>', $char) === false) /* do nothing */;
- else if ($r < 45) $chars[$key] = '&#x'.dechex($ord).';';
- else $chars[$key] = '&#'.$ord.';';
+ $r = ($seed * (1 + $key)) % 100; // Pseudo-random function.
+ // roughly 10% raw, 45% hex, 45% dec
+ // '@' *must* be encoded. I insist.
+ // '"' and '>' have to be encoded inside the attribute
+ if ($r > 90 && strpos('@"&>', $char) === false) {
+ /* do nothing */
+ } else if ($r < 45) {
+ $chars[$key] = '&#x'.dechex($ord).';';
+ } else {
+ $chars[$key] = '&#'.$ord.';';
+ }
}
}
@@ -1461,14 +1729,15 @@ class Markdown implements MarkdownInterface {
return $text;
}
-
+ /**
+ * Take the string $str and parse it into tokens, hashing embeded HTML,
+ * escaped characters and handling code spans.
+ * @param string $str
+ * @return string
+ */
protected function parseSpan($str) {
- #
- # Take the string $str and parse it into tokens, hashing embeded HTML,
- # escaped characters and handling code spans.
- #
$output = '';
-
+
$span_re = '{
(
\\\\'.$this->escape_chars_re.'
@@ -1496,121 +1765,145 @@ class Markdown implements MarkdownInterface {
}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.
- #
+ // 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, $str, 2, PREG_SPLIT_DELIM_CAPTURE);
-
- # Create token from text preceding tag.
+
+ // Create token from text preceding tag.
if ($parts[0] != "") {
$output .= $parts[0];
}
-
- # Check if we reach the end.
+
+ // Check if we reach the end.
if (isset($parts[1])) {
$output .= $this->handleSpanToken($parts[1], $parts[2]);
$str = $parts[2];
- }
- else {
+ } else {
break;
}
}
-
+
return $output;
}
-
-
+
+ /**
+ * Handle $token provided by parseSpan by determining its nature and
+ * returning the corresponding value that should replace it.
+ * @param string $token
+ * @param string $str Passed by reference
+ * @return string
+ */
protected function handleSpanToken($token, &$str) {
- #
- # Handle $token provided by parseSpan by determining its nature and
- # returning the corresponding value that should replace it.
- #
- switch ($token{0}) {
+ switch ($token[0]) {
case "\\":
- return $this->hashPart("&#". ord($token{1}). ";");
+ return $this->hashPart("&#". ord($token[1]). ";");
case "`":
- # Search for end marker in remaining text.
- if (preg_match('/^(.*?[^`])'.preg_quote($token).'(?!`)(.*)$/sm',
+ // 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.
+ return $token; // Return as text since no ending marker found.
default:
return $this->hashPart($token);
}
}
-
+ /**
+ * Remove one level of line-leading tabs or spaces
+ * @param string $text
+ * @return string
+ */
protected function outdent($text) {
- #
- # Remove one level of line-leading tabs or spaces
- #
- return preg_replace('/^(\t|[ ]{1,'.$this->tab_width.'})/m', '', $text);
+ return preg_replace('/^(\t|[ ]{1,' . $this->tab_width . '})/m', '', $text);
}
- # String length function for detab. `_initDetab` will create a function to
- # hanlde UTF-8 if the default function does not exist.
+ /**
+ * String length function for detab. `_initDetab` will create a function to
+ * handle UTF-8 if the default function does not exist.
+ * @var string
+ */
protected $utf8_strlen = 'mb_strlen';
-
+
+ /**
+ * Replace tabs with the appropriate amount of spaces.
+ *
+ * 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.
+ *
+ * @param string $text
+ * @return string
+ */
protected function detab($text) {
- #
- # Replace tabs with the appropriate amount of space.
- #
- # 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;
}
+
+ /**
+ * Replace tabs callback
+ * @param string $matches
+ * @return string
+ */
protected function _detab_callback($matches) {
$line = $matches[0];
- $strlen = $this->utf8_strlen; # strlen function for UTF-8.
-
- # Split in blocks.
+ $strlen = $this->utf8_strlen; // strlen function for UTF-8.
+
+ // Split in blocks.
$blocks = explode("\t", $line);
- # Add each blocks to the line.
+ // Add each blocks to the line.
$line = $blocks[0];
- unset($blocks[0]); # Do not add first block twice.
+ 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 -
+ // Calculate amount of space, insert spaces, insert block.
+ $amount = $this->tab_width -
$strlen($line, 'UTF-8') % $this->tab_width;
$line .= str_repeat(" ", $amount) . $block;
}
return $line;
}
+
+ /**
+ * Check for the availability of the function in the `utf8_strlen` property
+ * (initially `mb_strlen`). If the function is not available, create a
+ * function that will loosely count the number of UTF-8 characters with a
+ * regular expression.
+ * @return void
+ */
protected function _initDetab() {
- #
- # Check for the availability of the function in the `utf8_strlen` property
- # (initially `mb_strlen`). If the function is not available, create a
- # function that will loosely count the number of UTF-8 characters with a
- # regular expression.
- #
- if (function_exists($this->utf8_strlen)) return;
- $this->utf8_strlen = create_function('$text', 'return preg_match_all(
- "/[\\\\x00-\\\\xBF]|[\\\\xC0-\\\\xFF][\\\\x80-\\\\xBF]*/",
- $text, $m);');
+
+ if (function_exists($this->utf8_strlen)) {
+ return;
+ }
+
+ $this->utf8_strlen = function($text) {
+ return preg_match_all('/[\x00-\xBF]|[\xC0-\xFF][\x80-\xBF]*/', $text, $m);
+ };
}
-
+ /**
+ * Swap back in all the tags hashed by _HashHTMLBlocks.
+ * @param string $text
+ * @return string
+ */
protected function unhash($text) {
- #
- # Swap back in all the tags hashed by _HashHTMLBlocks.
- #
- return preg_replace_callback('/(.)\x1A[0-9]+\1/',
+ return preg_replace_callback('/(.)\x1A[0-9]+\1/',
array($this, '_unhash_callback'), $text);
}
+
+ /**
+ * Unhashing callback
+ * @param array $matches
+ * @return string
+ */
protected function _unhash_callback($matches) {
return $this->html_hashes[$matches[0]];
}
-
}
diff --git a/third/Md/MarkdownExtra.inc.php b/third/Md/MarkdownExtra.inc.php
index e11b1ef..d09bd7a 100644
--- a/third/Md/MarkdownExtra.inc.php
+++ b/third/Md/MarkdownExtra.inc.php
@@ -1,10 +1,10 @@
-#
-# Original Markdown
-# Copyright (c) 2004-2006 John Gruber
-#
+ * @copyright 2004-2019 Michel Fortin s around
+ * "paragraphs" that are wrapped in non-block-level tags, such as anchors,
+ * phrase emphasis, and spans. The list of tags we're looking for is
+ * hard-coded.
+ *
+ * This works by calling _HashHTMLBlocks_InMarkdown, which then calls
+ * _HashHTMLBlocks_InHTML when it encounter block tags. When the markdown="1"
+ * attribute is found within a tag, _HashHTMLBlocks_InHTML calls back
+ * _HashHTMLBlocks_InMarkdown to handle the Markdown syntax within the tag.
+ * These two functions are calling each other. It's recursive!
+ * @param string $text
+ * @return string
+ */
protected function hashHTMLBlocks($text) {
- #
- # Hashify HTML Blocks and "clean tags".
- #
- # We only want to do this for block-level HTML tags, such as headers,
- # lists, and tables. That's because we still want to wrap
s around
- # "paragraphs" that are wrapped in non-block-level tags, such as anchors,
- # phrase emphasis, and spans. The list of tags we're looking for is
- # hard-coded.
- #
- # This works by calling _HashHTMLBlocks_InMarkdown, which then calls
- # _HashHTMLBlocks_InHTML when it encounter block tags. When the markdown="1"
- # attribute is found within a tag, _HashHTMLBlocks_InHTML calls back
- # _HashHTMLBlocks_InMarkdown to handle the Markdown syntax within the tag.
- # These two functions are calling each other. It's recursive!
- #
- if ($this->no_markup) return $text;
+ if ($this->no_markup) {
+ return $text;
+ }
- #
- # Call the HTML-in-Markdown hasher.
- #
+ // Call the HTML-in-Markdown hasher.
list($text, ) = $this->_hashHTMLBlocks_inMarkdown($text);
-
+
return $text;
}
+
+ /**
+ * Parse markdown text, calling _HashHTMLBlocks_InHTML for block tags.
+ *
+ * * $indent is the number of space to be ignored when checking for code
+ * blocks. This is important because if we don't take the indent into
+ * account, something like this (which looks right) won't work as expected:
+ *
+ *
+ *
+ * Hello World. <-- Is this a Markdown code block or text?
+ * <-- Is this a Markdown code block or a real tag?
+ *
+ *
+ * If you don't like this, just don't indent the tag on which
+ * you apply the markdown="1" attribute.
+ *
+ * * If $enclosing_tag_re is not empty, stops at the first unmatched closing
+ * tag with that name. Nested tags supported.
+ *
+ * * If $span is true, text inside must treated as span. So any double
+ * newline will be replaced by a single newline so that it does not create
+ * paragraphs.
+ *
+ * Returns an array of that form: ( processed text , remaining text )
+ *
+ * @param string $text
+ * @param integer $indent
+ * @param string $enclosing_tag_re
+ * @param boolean $span
+ * @return array
+ */
protected function _hashHTMLBlocks_inMarkdown($text, $indent = 0,
$enclosing_tag_re = '', $span = false)
{
- #
- # Parse markdown text, calling _HashHTMLBlocks_InHTML for block tags.
- #
- # * $indent is the number of space to be ignored when checking for code
- # blocks. This is important because if we don't take the indent into
- # account, something like this (which looks right) won't work as expected:
- #
- #
- #
- # Hello World. <-- Is this a Markdown code block or text?
- # <-- Is this a Markdown code block or a real tag?
- #
- #
- # If you don't like this, just don't indent the tag on which
- # you apply the markdown="1" attribute.
- #
- # * If $enclosing_tag_re is not empty, stops at the first unmatched closing
- # tag with that name. Nested tags supported.
- #
- # * If $span is true, text inside must treated as span. So any double
- # newline will be replaced by a single newline so that it does not create
- # paragraphs.
- #
- # Returns an array of that form: ( processed text , remaining text )
- #
+
if ($text === '') return array('', '');
- # Regex to check for the presense of newlines around a block tag.
+ // Regex to check for the presense of newlines around a block tag.
$newline_before_re = '/(?:^\n?|\n\n)*$/';
- $newline_after_re =
+ $newline_after_re =
'{
^ # Start of text following the tag.
(?>[ ]*)? # Optional comment.
[ ]*\n # Must be followed by newline.
}xs';
-
- # Regex to match any tag.
+
+ // Regex to match any tag.
$block_tag_re =
'{
( # $2: Capture whole tag.
# Tag name.
- '.$this->block_tags_re.' |
- '.$this->context_block_tags_re.' |
- '.$this->clean_tags_re.' |
- (?!\s)'.$enclosing_tag_re.'
+ ' . $this->block_tags_re . ' |
+ ' . $this->context_block_tags_re . ' |
+ ' . $this->clean_tags_re . ' |
+ (?!\s)'.$enclosing_tag_re . '
)
(?:
(?=[\s"\'/a-zA-Z0-9]) # Allowed characters after tag name.
@@ -341,25 +472,25 @@ class MarkdownExtra extends \Michelf\Markdown {
<\?.*?\?> | <%.*?%> # Processing instruction
|
# CData Block
- '. ( !$span ? ' # If not in span.
+ ' . ( !$span ? ' # If not in span.
|
# Indented code block
(?: ^[ ]*\n | ^ | \n[ ]*\n )
- [ ]{'.($indent+4).'}[^\n]* \n
+ [ ]{' . ($indent + 4) . '}[^\n]* \n
(?>
- (?: [ ]{'.($indent+4).'}[^\n]* | [ ]* ) \n
+ (?: [ ]{' . ($indent + 4) . '}[^\n]* | [ ]* ) \n
)*
|
# Fenced code block marker
(?<= ^ | \n )
- [ ]{0,'.($indent+3).'}(?:~{3,}|`{3,})
+ [ ]{0,' . ($indent + 3) . '}(?:~{3,}|`{3,})
[ ]*
(?: \.?[-_:a-zA-Z0-9]+ )? # standalone class name
[ ]*
- (?: '.$this->id_class_attr_nocatch_re.' )? # extra attributes
+ (?: ' . $this->id_class_attr_nocatch_re . ' )? # extra attributes
[ ]*
(?= \n )
- ' : '' ). ' # End (if not is span).
+ ' : '' ) . ' # End (if not is span).
|
# Code span marker
# Note, this regex needs to go after backtick fenced
@@ -369,185 +500,172 @@ class MarkdownExtra extends \Michelf\Markdown {
)
}xs';
-
- $depth = 0; # Current depth inside the tag tree.
- $parsed = ""; # Parsed text that will be returned.
- #
- # Loop through every tag until we find the closing tag of the parent
- # or loop until reaching the end of text if no parent tag specified.
- #
+ $depth = 0; // Current depth inside the tag tree.
+ $parsed = ""; // Parsed text that will be returned.
+
+ // Loop through every tag until we find the closing tag of the parent
+ // or loop until reaching the end of text if no parent tag specified.
do {
- #
- # Split the text using the first $tag_match pattern found.
- # Text before pattern will be first in the array, text after
- # pattern will be at the end, and between will be any catches made
- # by the pattern.
- #
- $parts = preg_split($block_tag_re, $text, 2,
+ // Split the text using the first $tag_match pattern found.
+ // Text before pattern will be first in the array, text after
+ // pattern will be at the end, and between will be any catches made
+ // by the pattern.
+ $parts = preg_split($block_tag_re, $text, 2,
PREG_SPLIT_DELIM_CAPTURE);
-
- # If in Markdown span mode, add a empty-string span-level hash
- # after each newline to prevent triggering any block element.
+
+ // If in Markdown span mode, add a empty-string span-level hash
+ // after each newline to prevent triggering any block element.
if ($span) {
$void = $this->hashPart("", ':');
- $newline = "$void\n";
+ $newline = "\n$void";
$parts[0] = $void . str_replace("\n", $newline, $parts[0]) . $void;
}
-
- $parsed .= $parts[0]; # Text before current tag.
-
- # If end of $text has been reached. Stop loop.
+
+ $parsed .= $parts[0]; // Text before current tag.
+
+ // If end of $text has been reached. Stop loop.
if (count($parts) < 3) {
$text = "";
break;
}
-
- $tag = $parts[1]; # Tag to handle.
- $text = $parts[2]; # Remaining text after current tag.
- $tag_re = preg_quote($tag); # For use in a regular expression.
-
- #
- # Check for: Fenced code block marker.
- # Note: need to recheck the whole tag to disambiguate backtick
- # fences from code spans
- #
- if (preg_match('{^\n?([ ]{0,'.($indent+3).'})(~{3,}|`{3,})[ ]*(?:\.?[-_:a-zA-Z0-9]+)?[ ]*(?:'.$this->id_class_attr_nocatch_re.')?[ ]*\n?$}', $tag, $capture)) {
- # Fenced code block marker: find matching end marker.
- $fence_indent = strlen($capture[1]); # use captured indent in re
- $fence_re = $capture[2]; # use captured fence in re
- if (preg_match('{^(?>.*\n)*?[ ]{'.($fence_indent).'}'.$fence_re.'[ ]*(?:\n|$)}', $text,
- $matches))
+
+ $tag = $parts[1]; // Tag to handle.
+ $text = $parts[2]; // Remaining text after current tag.
+
+ // Check for: Fenced code block marker.
+ // Note: need to recheck the whole tag to disambiguate backtick
+ // fences from code spans
+ if (preg_match('{^\n?([ ]{0,' . ($indent + 3) . '})(~{3,}|`{3,})[ ]*(?:\.?[-_:a-zA-Z0-9]+)?[ ]*(?:' . $this->id_class_attr_nocatch_re . ')?[ ]*\n?$}', $tag, $capture)) {
+ // Fenced code block marker: find matching end marker.
+ $fence_indent = strlen($capture[1]); // use captured indent in re
+ $fence_re = $capture[2]; // use captured fence in re
+ if (preg_match('{^(?>.*\n)*?[ ]{' . ($fence_indent) . '}' . $fence_re . '[ ]*(?:\n|$)}', $text,
+ $matches))
{
- # End marker found: pass text unchanged until marker.
+ // End marker found: pass text unchanged until marker.
$parsed .= $tag . $matches[0];
$text = substr($text, strlen($matches[0]));
}
else {
- # No end marker: just skip it.
+ // No end marker: just skip it.
$parsed .= $tag;
}
}
- #
- # Check for: Indented code block.
- #
- else if ($tag{0} == "\n" || $tag{0} == " ") {
- # Indented code block: pass it unchanged, will be handled
- # later.
+ // Check for: Indented code block.
+ else if ($tag[0] === "\n" || $tag[0] === " ") {
+ // Indented code block: pass it unchanged, will be handled
+ // later.
$parsed .= $tag;
}
- #
- # Check for: Code span marker
- # Note: need to check this after backtick fenced code blocks
- #
- else if ($tag{0} == "`") {
- # Find corresponding end marker.
+ // Check for: Code span marker
+ // Note: need to check this after backtick fenced code blocks
+ else if ($tag[0] === "`") {
+ // Find corresponding end marker.
$tag_re = preg_quote($tag);
- if (preg_match('{^(?>.+?|\n(?!\n))*?(?.+?|\n(?!\n))*?(?block_tags_re.')\b}', $tag) ||
- ( preg_match('{^<(?:'.$this->context_block_tags_re.')\b}', $tag) &&
+ // Check for: Opening Block level tag or
+ // Opening Context Block tag (like ins and del)
+ // used as a block tag (tag is alone on it's line).
+ else if (preg_match('{^<(?:' . $this->block_tags_re . ')\b}', $tag) ||
+ ( preg_match('{^<(?:' . $this->context_block_tags_re . ')\b}', $tag) &&
preg_match($newline_before_re, $parsed) &&
preg_match($newline_after_re, $text) )
)
{
- # Need to parse tag and following text using the HTML parser.
- list($block_text, $text) =
+ // Need to parse tag and following text using the HTML parser.
+ list($block_text, $text) =
$this->_hashHTMLBlocks_inHTML($tag . $text, "hashBlock", true);
-
- # Make sure it stays outside of any paragraph by adding newlines.
+
+ // Make sure it stays outside of any paragraph by adding newlines.
$parsed .= "\n\n$block_text\n\n";
}
- #
- # Check for: Clean tag (like script, math)
- # HTML Comments, processing instructions.
- #
- else if (preg_match('{^<(?:'.$this->clean_tags_re.')\b}', $tag) ||
- $tag{1} == '!' || $tag{1} == '?')
+ // Check for: Clean tag (like script, math)
+ // HTML Comments, processing instructions.
+ else if (preg_match('{^<(?:' . $this->clean_tags_re . ')\b}', $tag) ||
+ $tag[1] === '!' || $tag[1] === '?')
{
- # Need to parse tag and following text using the HTML parser.
- # (don't check for markdown attribute)
- list($block_text, $text) =
+ // Need to parse tag and following text using the HTML parser.
+ // (don't check for markdown attribute)
+ list($block_text, $text) =
$this->_hashHTMLBlocks_inHTML($tag . $text, "hashClean", false);
-
+
$parsed .= $block_text;
}
- #
- # Check for: Tag with same name as enclosing tag.
- #
+ // Check for: Tag with same name as enclosing tag.
else if ($enclosing_tag_re !== '' &&
- # Same name as enclosing tag.
- preg_match('{^= 0);
-
+
return array($parsed, $text);
}
+
+ /**
+ * Parse HTML, calling _HashHTMLBlocks_InMarkdown for block tags.
+ *
+ * * Calls $hash_method to convert any blocks.
+ * * Stops when the first opening tag closes.
+ * * $md_attr indicate if the use of the `markdown="1"` attribute is allowed.
+ * (it is not inside clean tags)
+ *
+ * Returns an array of that form: ( processed text , remaining text )
+ * @param string $text
+ * @param string $hash_method
+ * @param bool $md_attr Handle `markdown="1"` attribute
+ * @return array
+ */
protected function _hashHTMLBlocks_inHTML($text, $hash_method, $md_attr) {
- #
- # Parse HTML, calling _HashHTMLBlocks_InMarkdown for block tags.
- #
- # * Calls $hash_method to convert any blocks.
- # * Stops when the first opening tag closes.
- # * $md_attr indicate if the use of the `markdown="1"` attribute is allowed.
- # (it is not inside clean tags)
- #
- # Returns an array of that form: ( processed text , remaining text )
- #
if ($text === '') return array('', '');
-
- # Regex to match `markdown` attribute inside of a tag.
+
+ // Regex to match `markdown` attribute inside of a tag.
$markdown_attr_re = '
{
\s* # Eat whitespace before the `markdown` attribute
markdown
\s*=\s*
(?>
- (["\']) # $1: quote delimiter
+ (["\']) # $1: quote delimiter
(.*?) # $2: attribute value
- \1 # matching delimiter
+ \1 # matching delimiter
|
([^\s>]*) # $3: unquoted attribute value
)
() # $4: make $3 always defined (avoid warnings)
}xs';
-
- # Regex to match any tag.
+
+ // Regex to match any tag.
$tag_re = '{
( # $2: Capture whole tag.
# CData Block
)
}xs';
-
- $original_text = $text; # Save original text in case of faliure.
-
- $depth = 0; # Current depth inside the tag tree.
- $block_text = ""; # Temporary text holder for current text.
- $parsed = ""; # Parsed text that will be returned.
- #
- # Get the name of the starting tag.
- # (This pattern makes $base_tag_name_re safe without quoting.)
- #
+ $original_text = $text; // Save original text in case of faliure.
+
+ $depth = 0; // Current depth inside the tag tree.
+ $block_text = ""; // Temporary text holder for current text.
+ $parsed = ""; // Parsed text that will be returned.
+ $base_tag_name_re = '';
+
+ // Get the name of the starting tag.
+ // (This pattern makes $base_tag_name_re safe without quoting.)
if (preg_match('/^<([\w:$]*)\b/', $text, $matches))
$base_tag_name_re = $matches[1];
- #
- # Loop through every tag until we find the corresponding closing tag.
- #
+ // Loop through every tag until we find the corresponding closing tag.
do {
- #
- # Split the text using the first $tag_match pattern found.
- # Text before pattern will be first in the array, text after
- # pattern will be at the end, and between will be any catches made
- # by the pattern.
- #
+ // Split the text using the first $tag_match pattern found.
+ // Text before pattern will be first in the array, text after
+ // pattern will be at the end, and between will be any catches made
+ // by the pattern.
$parts = preg_split($tag_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE);
-
+
if (count($parts) < 3) {
- #
- # End of $text reached with unbalenced tag(s).
- # In that case, we return original text unchanged and pass the
- # first character as filtered to prevent an infinite loop in the
- # parent function.
- #
- return array($original_text{0}, substr($original_text, 1));
+ // End of $text reached with unbalenced tag(s).
+ // In that case, we return original text unchanged and pass the
+ // first character as filtered to prevent an infinite loop in the
+ // parent function.
+ return array($original_text[0], substr($original_text, 1));
}
-
- $block_text .= $parts[0]; # Text before current tag.
- $tag = $parts[1]; # Tag to handle.
- $text = $parts[2]; # Remaining text after current tag.
-
- #
- # Check for: Auto-close tag (like
)
- # Comments and Processing Instructions.
- #
- if (preg_match('{^auto_close_tags_re.')\b}', $tag) ||
- $tag{1} == '!' || $tag{1} == '?')
+
+ $block_text .= $parts[0]; // Text before current tag.
+ $tag = $parts[1]; // Tag to handle.
+ $text = $parts[2]; // Remaining text after current tag.
+
+ // Check for: Auto-close tag (like
)
+ // Comments and Processing Instructions.
+ if (preg_match('{^auto_close_tags_re . ')\b}', $tag) ||
+ $tag[1] === '!' || $tag[1] === '?')
{
- # Just add the tag to the block as if it was text.
+ // Just add the tag to the block as if it was text.
$block_text .= $tag;
}
else {
- #
- # Increase/decrease nested tag count. Only do so if
- # the tag's name match base tag's.
- #
- if (preg_match('{^mode = $attr_m[2] . $attr_m[3];
- $span_mode = $this->mode == 'span' || $this->mode != 'block' &&
- preg_match('{^<(?:'.$this->contain_span_tags_re.')\b}', $tag);
-
- # Calculate indent before tag.
+
+ // Check if text inside this tag must be parsed in span mode.
+ $mode = $attr_m[2] . $attr_m[3];
+ $span_mode = $mode === 'span' || ($mode !== 'block' &&
+ preg_match('{^<(?:' . $this->contain_span_tags_re . ')\b}', $tag));
+
+ // Calculate indent before tag.
if (preg_match('/(?:^|\n)( *?)(?! ).*?$/', $block_text, $matches)) {
$strlen = $this->utf8_strlen;
$indent = $strlen($matches[1], 'UTF-8');
} else {
$indent = 0;
}
-
- # End preceding block with this tag.
+
+ // End preceding block with this tag.
$block_text .= $tag;
$parsed .= $this->$hash_method($block_text);
-
- # Get enclosing tag name for the ParseMarkdown function.
- # (This pattern makes $tag_name_re safe without quoting.)
+
+ // Get enclosing tag name for the ParseMarkdown function.
+ // (This pattern makes $tag_name_re safe without quoting.)
preg_match('/^<([\w:$]*)\b/', $tag, $matches);
$tag_name_re = $matches[1];
-
- # Parse the content using the HTML-in-Markdown parser.
+
+ // Parse the content using the HTML-in-Markdown parser.
list ($block_text, $text)
- = $this->_hashHTMLBlocks_inMarkdown($text, $indent,
+ = $this->_hashHTMLBlocks_inMarkdown($text, $indent,
$tag_name_re, $span_mode);
-
- # Outdent markdown text.
+
+ // Outdent markdown text.
if ($indent > 0) {
- $block_text = preg_replace("/^[ ]{1,$indent}/m", "",
+ $block_text = preg_replace("/^[ ]{1,$indent}/m", "",
$block_text);
}
-
- # Append tag content to parsed text.
- if (!$span_mode) $parsed .= "\n\n$block_text\n\n";
- else $parsed .= "$block_text";
-
- # Start over with a new block.
+
+ // Append tag content to parsed text.
+ if (!$span_mode) {
+ $parsed .= "\n\n$block_text\n\n";
+ } else {
+ $parsed .= (string) $block_text;
+ }
+
+ // Start over with a new block.
$block_text = "";
}
else $block_text .= $tag;
}
-
+
} while ($depth > 0);
-
- #
- # Hash last block text that wasn't processed inside the loop.
- #
+
+ // Hash last block text that wasn't processed inside the loop.
$parsed .= $this->$hash_method($block_text);
-
+
return array($parsed, $text);
}
-
+ /**
+ * Called whenever a tag must be hashed when a function inserts a "clean" tag
+ * in $text, it passes through this function and is automaticaly escaped,
+ * blocking invalid nested overlap.
+ * @param string $text
+ * @return string
+ */
protected function hashClean($text) {
- #
- # Called whenever a tag must be hashed when a function inserts a "clean" tag
- # in $text, it passes through this function and is automaticaly escaped,
- # blocking invalid nested overlap.
- #
return $this->hashPart($text, 'C');
}
-
+ /**
+ * Turn Markdown link shortcuts into XHTML tags.
+ * @param string $text
+ * @return string
+ */
protected function doAnchors($text) {
- #
- # Turn Markdown link shortcuts into XHTML tags.
- #
- if ($this->in_anchor) return $text;
+ if ($this->in_anchor) {
+ return $text;
+ }
$this->in_anchor = true;
-
- #
- # First, handle reference-style links: [link text] [id]
- #
+
+ // First, handle reference-style links: [link text] [id]
$text = preg_replace_callback('{
( # wrap whole match in $1
\[
- ('.$this->nested_brackets_re.') # link text = $2
+ (' . $this->nested_brackets_re . ') # link text = $2
\]
[ ]? # one optional space
@@ -729,20 +840,18 @@ class MarkdownExtra extends \Michelf\Markdown {
}xs',
array($this, '_doAnchors_reference_callback'), $text);
- #
- # Next, inline-style links: [link text](url "optional title")
- #
+ // Next, inline-style links: [link text](url "optional title")
$text = preg_replace_callback('{
( # wrap whole match in $1
\[
- ('.$this->nested_brackets_re.') # link text = $2
+ (' . $this->nested_brackets_re . ') # link text = $2
\]
\( # literal paren
[ \n]*
(?:
<(.+?)> # href = $3
|
- ('.$this->nested_url_parenthesis_re.') # href = $4
+ (' . $this->nested_url_parenthesis_re . ') # href = $4
)
[ \n]*
( # $5
@@ -752,16 +861,14 @@ class MarkdownExtra extends \Michelf\Markdown {
[ \n]* # ignore any spaces/tabs between closing quote and )
)? # title is optional
\)
- (?:[ ]? '.$this->id_class_attr_catch_re.' )? # $8 = id/class attributes
+ (?:[ ]? ' . $this->id_class_attr_catch_re . ' )? # $8 = id/class attributes
)
}xs',
array($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)
- #
+ // 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
\[
@@ -774,24 +881,30 @@ class MarkdownExtra extends \Michelf\Markdown {
$this->in_anchor = false;
return $text;
}
+
+ /**
+ * Callback for reference anchors
+ * @param array $matches
+ * @return string
+ */
protected function _doAnchors_reference_callback($matches) {
$whole_match = $matches[1];
$link_text = $matches[2];
$link_id =& $matches[3];
if ($link_id == "") {
- # for shortcut links like [this][] or [this].
+ // for shortcut links like [this][] or [this].
$link_id = $link_text;
}
-
- # lower-case and turn embedded newlines into spaces
+
+ // 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 = $this->encodeURLAttribute($url);
-
+
$result = "titles[$link_id] ) ) {
$title = $this->titles[$link_id];
@@ -800,7 +913,7 @@ class MarkdownExtra extends \Michelf\Markdown {
}
if (isset($this->ref_attr[$link_id]))
$result .= $this->ref_attr[$link_id];
-
+
$link_text = $this->runSpanGamut($link_text);
$result .= ">$link_text";
$result = $this->hashPart($result);
@@ -810,17 +923,22 @@ class MarkdownExtra extends \Michelf\Markdown {
}
return $result;
}
+
+ /**
+ * Callback for inline anchors
+ * @param array $matches
+ * @return string
+ */
protected function _doAnchors_inline_callback($matches) {
- $whole_match = $matches[1];
$link_text = $this->runSpanGamut($matches[2]);
- $url = $matches[3] == '' ? $matches[4] : $matches[3];
+ $url = $matches[3] === '' ? $matches[4] : $matches[3];
$title =& $matches[7];
$attr = $this->doExtraAttributes("a", $dummy =& $matches[8]);
// if the URL was of the form it got caught by the HTML
// tag parser and hashed. Need to reverse the process before using the URL.
$unhashed = $this->unhash($url);
- if ($unhashed != $url)
+ if ($unhashed !== $url)
$url = preg_replace('/^<(.*)>$/', '\1', $unhashed);
$url = $this->encodeURLAttribute($url);
@@ -831,25 +949,24 @@ class MarkdownExtra extends \Michelf\Markdown {
$result .= " title=\"$title\"";
}
$result .= $attr;
-
+
$link_text = $this->runSpanGamut($link_text);
$result .= ">$link_text";
return $this->hashPart($result);
}
-
+ /**
+ * Turn Markdown image shortcuts into ![]()
tags.
+ * @param string $text
+ * @return string
+ */
protected function doImages($text) {
- #
- # Turn Markdown image shortcuts into ![]()
tags.
- #
- #
- # First, handle reference-style labeled images: ![alt text][id]
- #
+ // First, handle reference-style labeled images: ![alt text][id]
$text = preg_replace_callback('{
( # wrap whole match in $1
!\[
- ('.$this->nested_brackets_re.') # alt text = $2
+ (' . $this->nested_brackets_re . ') # alt text = $2
\]
[ ]? # one optional space
@@ -860,17 +977,15 @@ class MarkdownExtra extends \Michelf\Markdown {
\]
)
- }xs',
+ }xs',
array($this, '_doImages_reference_callback'), $text);
- #
- # Next, handle inline images: 
- # Don't forget: encode * and _
- #
+ // Next, handle inline images: 
+ // Don't forget: encode * and _
$text = preg_replace_callback('{
( # wrap whole match in $1
!\[
- ('.$this->nested_brackets_re.') # alt text = $2
+ (' . $this->nested_brackets_re . ') # alt text = $2
\]
\s? # One optional whitespace character
\( # literal paren
@@ -878,7 +993,7 @@ class MarkdownExtra extends \Michelf\Markdown {
(?:
<(\S*)> # src url = $3
|
- ('.$this->nested_url_parenthesis_re.') # src url = $4
+ (' . $this->nested_url_parenthesis_re . ') # src url = $4
)
[ \n]*
( # $5
@@ -888,20 +1003,26 @@ class MarkdownExtra extends \Michelf\Markdown {
[ \n]*
)? # title is optional
\)
- (?:[ ]? '.$this->id_class_attr_catch_re.' )? # $8 = id/class attributes
+ (?:[ ]? ' . $this->id_class_attr_catch_re . ' )? # $8 = id/class attributes
)
}xs',
array($this, '_doImages_inline_callback'), $text);
return $text;
}
+
+ /**
+ * Callback for referenced images
+ * @param array $matches
+ * @return string
+ */
protected function _doImages_reference_callback($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][].
+ if ($link_id === "") {
+ $link_id = strtolower($alt_text); // for shortcut links like ![this][].
}
$alt_text = $this->encodeAttribute($alt_text);
@@ -913,22 +1034,28 @@ class MarkdownExtra extends \Michelf\Markdown {
$title = $this->encodeAttribute($title);
$result .= " title=\"$title\"";
}
- if (isset($this->ref_attr[$link_id]))
+ if (isset($this->ref_attr[$link_id])) {
$result .= $this->ref_attr[$link_id];
+ }
$result .= $this->empty_element_suffix;
$result = $this->hashPart($result);
}
else {
- # If there's no such link ID, leave intact:
+ // If there's no such link ID, leave intact:
$result = $whole_match;
}
return $result;
}
+
+ /**
+ * Callback for inline images
+ * @param array $matches
+ * @return string
+ */
protected function _doImages_inline_callback($matches) {
- $whole_match = $matches[1];
$alt_text = $matches[2];
- $url = $matches[3] == '' ? $matches[4] : $matches[3];
+ $url = $matches[3] === '' ? $matches[4] : $matches[3];
$title =& $matches[7];
$attr = $this->doExtraAttributes("img", $dummy =& $matches[8]);
@@ -937,7 +1064,7 @@ class MarkdownExtra extends \Michelf\Markdown {
$result = "![\"$alt_text\"";](\"$url\")
encodeAttribute($title);
- $result .= " title=\"$title\""; # $title already quoted
+ $result .= " title=\"$title\""; // $title already quoted
}
$result .= $attr;
$result .= $this->empty_element_suffix;
@@ -945,40 +1072,41 @@ class MarkdownExtra extends \Michelf\Markdown {
return $this->hashPart($result);
}
-
+ /**
+ * Process markdown headers. Redefined to add ID and class attribute support.
+ * @param string $text
+ * @return string
+ */
protected function doHeaders($text) {
- #
- # Redefined to add id and class attribute support.
- #
- # Setext-style headers:
- # Header 1 {#header1}
- # ========
- #
- # Header 2 {#header2 .class1 .class2}
- # --------
- #
+ // Setext-style headers:
+ // Header 1 {#header1}
+ // ========
+ //
+ // Header 2 {#header2 .class1 .class2}
+ // --------
+ //
$text = preg_replace_callback(
'{
(^.+?) # $1: Header text
- (?:[ ]+ '.$this->id_class_attr_catch_re.' )? # $3 = id/class attributes
+ (?:[ ]+ ' . $this->id_class_attr_catch_re . ' )? # $3 = id/class attributes
[ ]*\n(=+|-+)[ ]*\n+ # $3: Header footer
}mx',
array($this, '_doHeaders_callback_setext'), $text);
- # atx-style headers:
- # # Header 1 {#header1}
- # ## Header 2 {#header2}
- # ## Header 2 with closing hashes ## {#header3.class1.class2}
- # ...
- # ###### Header 6 {.class2}
- #
+ // atx-style headers:
+ // # Header 1 {#header1}
+ // ## Header 2 {#header2}
+ // ## Header 2 with closing hashes ## {#header3.class1.class2}
+ // ...
+ // ###### Header 6 {.class2}
+ //
$text = preg_replace_callback('{
^(\#{1,6}) # $1 = string of #\'s
- [ ]*
+ [ ]'.($this->hashtag_protection ? '+' : '*').'
(.+?) # $2 = Header text
[ ]*
\#* # optional closing #\'s (not counted)
- (?:[ ]+ '.$this->id_class_attr_catch_re.' )? # $3 = id/class attributes
+ (?:[ ]+ ' . $this->id_class_attr_catch_re . ' )? # $3 = id/class attributes
[ ]*
\n+
}xm',
@@ -986,51 +1114,63 @@ class MarkdownExtra extends \Michelf\Markdown {
return $text;
}
- protected function _doHeaders_callback_setext($matches) {
- if ($matches[3] == '-' && preg_match('{^- }', $matches[1]))
- return $matches[0];
- $level = $matches[3]{0} == '=' ? 1 : 2;
+ /**
+ * Callback for setext headers
+ * @param array $matches
+ * @return string
+ */
+ protected function _doHeaders_callback_setext($matches) {
+ if ($matches[3] === '-' && preg_match('{^- }', $matches[1])) {
+ return $matches[0];
+ }
+
+ $level = $matches[3][0] === '=' ? 1 : 2;
$defaultId = is_callable($this->header_id_func) ? call_user_func($this->header_id_func, $matches[1]) : null;
$attr = $this->doExtraAttributes("h$level", $dummy =& $matches[2], $defaultId);
- $block = "".$this->runSpanGamut($matches[1])."";
+ $block = "" . $this->runSpanGamut($matches[1]) . "";
return "\n" . $this->hashBlock($block) . "\n\n";
}
+
+ /**
+ * Callback for atx headers
+ * @param array $matches
+ * @return string
+ */
protected function _doHeaders_callback_atx($matches) {
$level = strlen($matches[1]);
$defaultId = is_callable($this->header_id_func) ? call_user_func($this->header_id_func, $matches[2]) : null;
$attr = $this->doExtraAttributes("h$level", $dummy =& $matches[3], $defaultId);
- $block = "".$this->runSpanGamut($matches[2])."";
+ $block = "" . $this->runSpanGamut($matches[2]) . "";
return "\n" . $this->hashBlock($block) . "\n\n";
}
-
+ /**
+ * Form HTML tables.
+ * @param string $text
+ * @return string
+ */
protected function doTables($text) {
- #
- # Form HTML tables.
- #
$less_than_tab = $this->tab_width - 1;
- #
- # Find tables with leading pipe.
- #
- # | Header 1 | Header 2
- # | -------- | --------
- # | Cell 1 | Cell 2
- # | Cell 3 | Cell 4
- #
+ // Find tables with leading pipe.
+ //
+ // | Header 1 | Header 2
+ // | -------- | --------
+ // | Cell 1 | Cell 2
+ // | Cell 3 | Cell 4
$text = preg_replace_callback('
{
^ # Start of a line
- [ ]{0,'.$less_than_tab.'} # Allowed whitespace.
+ [ ]{0,' . $less_than_tab . '} # Allowed whitespace.
[|] # Optional leading pipe (present)
(.+) \n # $1: Header row (at least one pipe)
-
- [ ]{0,'.$less_than_tab.'} # Allowed whitespace.
+
+ [ ]{0,' . $less_than_tab . '} # Allowed whitespace.
[|] ([ ]*[-:]+[-| :]*) \n # $2: Header underline
-
+
( # $3: Cells
(?>
[ ]* # Allowed whitespace.
@@ -1040,24 +1180,22 @@ class MarkdownExtra extends \Michelf\Markdown {
(?=\n|\Z) # Stop at final double newline.
}xm',
array($this, '_doTable_leadingPipe_callback'), $text);
-
- #
- # Find tables without leading pipe.
- #
- # Header 1 | Header 2
- # -------- | --------
- # Cell 1 | Cell 2
- # Cell 3 | Cell 4
- #
+
+ // Find tables without leading pipe.
+ //
+ // Header 1 | Header 2
+ // -------- | --------
+ // Cell 1 | Cell 2
+ // Cell 3 | Cell 4
$text = preg_replace_callback('
{
^ # Start of a line
- [ ]{0,'.$less_than_tab.'} # Allowed whitespace.
+ [ ]{0,' . $less_than_tab . '} # Allowed whitespace.
(\S.*[|].*) \n # $1: Header row (at least one pipe)
-
- [ ]{0,'.$less_than_tab.'} # Allowed whitespace.
+
+ [ ]{0,' . $less_than_tab . '} # Allowed whitespace.
([-:]+[ ]*[|][-| :]*) \n # $2: Header underline
-
+
( # $3: Cells
(?>
.* [|] .* \n # Row content
@@ -1069,35 +1207,52 @@ class MarkdownExtra extends \Michelf\Markdown {
return $text;
}
+
+ /**
+ * Callback for removing the leading pipe for each row
+ * @param array $matches
+ * @return string
+ */
protected function _doTable_leadingPipe_callback($matches) {
$head = $matches[1];
$underline = $matches[2];
$content = $matches[3];
-
- # Remove leading pipe for each row.
+
$content = preg_replace('/^ *[|]/m', '', $content);
-
+
return $this->_doTable_callback(array($matches[0], $head, $underline, $content));
}
- protected function _doTable_makeAlignAttr($alignname)
- {
- if (empty($this->table_align_class_tmpl))
+
+ /**
+ * Make the align attribute in a table
+ * @param string $alignname
+ * @return string
+ */
+ protected function _doTable_makeAlignAttr($alignname) {
+ if (empty($this->table_align_class_tmpl)) {
return " align=\"$alignname\"";
+ }
$classname = str_replace('%%', $alignname, $this->table_align_class_tmpl);
return " class=\"$classname\"";
}
+
+ /**
+ * Calback for processing tables
+ * @param array $matches
+ * @return string
+ */
protected function _doTable_callback($matches) {
$head = $matches[1];
$underline = $matches[2];
$content = $matches[3];
- # Remove any tailing pipes for each line.
+ // Remove any tailing pipes for each line.
$head = preg_replace('/[|] *$/m', '', $head);
$underline = preg_replace('/[|] *$/m', '', $underline);
$content = preg_replace('/[|] *$/m', '', $content);
-
- # Reading alignement from header underline.
+
+ // Reading alignement from header underline.
$separators = preg_split('/ *[|] */', $underline);
foreach ($separators as $n => $s) {
if (preg_match('/^ *-+: *$/', $s))
@@ -1109,62 +1264,65 @@ class MarkdownExtra extends \Michelf\Markdown {
else
$attr[$n] = '';
}
-
- # Parsing span elements, including code spans, character escapes,
- # and inline HTML tags, so that pipes inside those gets ignored.
+
+ // Parsing span elements, including code spans, character escapes,
+ // and inline HTML tags, so that pipes inside those gets ignored.
$head = $this->parseSpan($head);
$headers = preg_split('/ *[|] */', $head);
$col_count = count($headers);
$attr = array_pad($attr, $col_count, '');
-
- # Write column headers.
+
+ // Write column headers.
$text = "\n";
$text .= "\n";
$text .= "\n";
- foreach ($headers as $n => $header)
- $text .= " ".$this->runSpanGamut(trim($header))."\n";
+ foreach ($headers as $n => $header) {
+ $text .= " " . $this->runSpanGamut(trim($header)) . "\n";
+ }
$text .= " \n";
$text .= "\n";
-
- # Split content by row.
+
+ // Split content by row.
$rows = explode("\n", trim($content, "\n"));
-
+
$text .= "\n";
foreach ($rows as $row) {
- # Parsing span elements, including code spans, character escapes,
- # and inline HTML tags, so that pipes inside those gets ignored.
+ // Parsing span elements, including code spans, character escapes,
+ // and inline HTML tags, so that pipes inside those gets ignored.
$row = $this->parseSpan($row);
-
- # Split row by cell.
+
+ // Split row by cell.
$row_cells = preg_split('/ *[|] */', $row, $col_count);
$row_cells = array_pad($row_cells, $col_count, '');
-
+
$text .= "\n";
- foreach ($row_cells as $n => $cell)
- $text .= " ".$this->runSpanGamut(trim($cell))."\n";
+ foreach ($row_cells as $n => $cell) {
+ $text .= " " . $this->runSpanGamut(trim($cell)) . "\n";
+ }
$text .= " \n";
}
$text .= "\n";
$text .= "
";
-
+
return $this->hashBlock($text) . "\n";
}
-
+ /**
+ * Form HTML definition lists.
+ * @param string $text
+ * @return string
+ */
protected function doDefLists($text) {
- #
- # Form HTML definition lists.
- #
$less_than_tab = $this->tab_width - 1;
- # Re-usable pattern to match any entire dl list:
+ // Re-usable pattern to match any entire dl list:
$whole_list_re = '(?>
( # $1 = whole list
( # $2
- [ ]{0,'.$less_than_tab.'}
+ [ ]{0,' . $less_than_tab . '}
((?>.*\S.*\n)+) # $3 = defined term
\n?
- [ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition
+ [ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition
)
(?s:.+?)
( # $4
@@ -1173,13 +1331,13 @@ class MarkdownExtra extends \Michelf\Markdown {
\n{2,}
(?=\S)
(?! # Negative lookahead for another term
- [ ]{0,'.$less_than_tab.'}
+ [ ]{0,' . $less_than_tab . '}
(?: \S.*\n )+? # defined term
\n?
- [ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition
+ [ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition
)
(?! # Negative lookahead for another definition
- [ ]{0,'.$less_than_tab.'}:[ ]+ # colon starting definition
+ [ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition
)
)
)
@@ -1187,67 +1345,81 @@ class MarkdownExtra extends \Michelf\Markdown {
$text = preg_replace_callback('{
(?>\A\n?|(?<=\n\n))
- '.$whole_list_re.'
+ ' . $whole_list_re . '
}mx',
array($this, '_doDefLists_callback'), $text);
return $text;
}
+
+ /**
+ * Callback for processing definition lists
+ * @param array $matches
+ * @return string
+ */
protected function _doDefLists_callback($matches) {
- # Re-usable patterns to match list item bullets and number markers:
+ // Re-usable patterns to match list item bullets and number markers:
$list = $matches[1];
-
- # Turn double returns into triple returns, so that we can make a
- # paragraph for the last item in a list, if necessary:
+
+ // Turn double returns into triple returns, so that we can make a
+ // paragraph for the last item in a list, if necessary:
$result = trim($this->processDefListItems($list));
$result = "\n" . $result . "\n
";
return $this->hashBlock($result) . "\n\n";
}
-
+ /**
+ * Process the contents of a single definition list, splitting it
+ * into individual term and definition list items.
+ * @param string $list_str
+ * @return string
+ */
protected function processDefListItems($list_str) {
- #
- # Process the contents of a single definition list, splitting it
- # into individual term and definition list items.
- #
+
$less_than_tab = $this->tab_width - 1;
-
- # trim trailing blank lines:
+
+ // Trim trailing blank lines:
$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str);
- # Process definition terms.
+ // Process definition terms.
$list_str = preg_replace_callback('{
- (?>\A\n?|\n\n+) # leading line
- ( # definition terms = $1
- [ ]{0,'.$less_than_tab.'} # leading whitespace
- (?!\:[ ]|[ ]) # negative lookahead for a definition
- # mark (colon) or more whitespace.
- (?> \S.* \n)+? # actual term (not whitespace).
- )
- (?=\n?[ ]{0,3}:[ ]) # lookahead for following line feed
- # with a definition mark.
+ (?>\A\n?|\n\n+) # leading line
+ ( # definition terms = $1
+ [ ]{0,' . $less_than_tab . '} # leading whitespace
+ (?!\:[ ]|[ ]) # negative lookahead for a definition
+ # mark (colon) or more whitespace.
+ (?> \S.* \n)+? # actual term (not whitespace).
+ )
+ (?=\n?[ ]{0,3}:[ ]) # lookahead for following line feed
+ # with a definition mark.
}xm',
array($this, '_processDefListItems_callback_dt'), $list_str);
- # Process actual definitions.
+ // Process actual definitions.
$list_str = preg_replace_callback('{
- \n(\n+)? # leading line = $1
- ( # marker space = $2
- [ ]{0,'.$less_than_tab.'} # whitespace before colon
- \:[ ]+ # definition mark (colon)
+ \n(\n+)? # leading line = $1
+ ( # marker space = $2
+ [ ]{0,' . $less_than_tab . '} # whitespace before colon
+ \:[ ]+ # definition mark (colon)
)
- ((?s:.+?)) # definition text = $3
- (?= \n+ # stop at next definition mark,
- (?: # next term or end of text
- [ ]{0,'.$less_than_tab.'} \:[ ] |
+ ((?s:.+?)) # definition text = $3
+ (?= \n+ # stop at next definition mark,
+ (?: # next term or end of text
+ [ ]{0,' . $less_than_tab . '} \:[ ] |
| \z
- )
- )
+ )
+ )
}xm',
array($this, '_processDefListItems_callback_dd'), $list_str);
return $list_str;
}
+
+ /**
+ * Callback for elements in definition lists
+ * @param array $matches
+ * @return string
+ */
protected function _processDefListItems_callback_dt($matches) {
$terms = explode("\n", trim($matches[1]));
$text = '';
@@ -1257,13 +1429,19 @@ class MarkdownExtra extends \Michelf\Markdown {
}
return $text . "\n";
}
+
+ /**
+ * Callback for elements in definition lists
+ * @param array $matches
+ * @return string
+ */
protected function _processDefListItems_callback_dd($matches) {
$leading_line = $matches[1];
$marker_space = $matches[2];
$def = $matches[3];
if ($leading_line || preg_match('/\n{2,}/', $def)) {
- # Replace marker with the appropriate whitespace indentation
+ // Replace marker with the appropriate whitespace indentation
$def = str_repeat(' ', strlen($marker_space)) . $def;
$def = $this->runBlockGamut($this->outdent($def . "\n\n"));
$def = "\n". $def ."\n";
@@ -1276,17 +1454,18 @@ class MarkdownExtra extends \Michelf\Markdown {
return "\n" . $def . " \n";
}
-
+ /**
+ * Adding the fenced code block syntax to regular Markdown:
+ *
+ * ~~~
+ * Code block
+ * ~~~
+ *
+ * @param string $text
+ * @return string
+ */
protected function doFencedCodeBlocks($text) {
- #
- # Adding the fenced code block syntax to regular Markdown:
- #
- # ~~~
- # Code block
- # ~~~
- #
- $less_than_tab = $this->tab_width;
-
+
$text = preg_replace_callback('{
(?:\n|\A)
# 1: Opening marker
@@ -1299,10 +1478,10 @@ class MarkdownExtra extends \Michelf\Markdown {
)?
[ ]*
(?:
- '.$this->id_class_attr_catch_re.' # 3: Extra attributes
+ ' . $this->id_class_attr_catch_re . ' # 3: Extra attributes
)?
[ ]* \n # Whitespace and newline following marker.
-
+
# 4: Content
(
(?>
@@ -1310,7 +1489,7 @@ class MarkdownExtra extends \Michelf\Markdown {
.*\n+
)+
)
-
+
# Closing marker.
\1 [ ]* (?= \n )
}xm',
@@ -1318,6 +1497,12 @@ class MarkdownExtra extends \Michelf\Markdown {
return $text;
}
+
+ /**
+ * Callback to process fenced code blocks
+ * @param array $matches
+ * @return string
+ */
protected function _doFencedCodeBlocks_callback($matches) {
$classname =& $matches[2];
$attrs =& $matches[3];
@@ -1333,204 +1518,256 @@ class MarkdownExtra extends \Michelf\Markdown {
array($this, '_doFencedCodeBlocks_newlines'), $codeblock);
$classes = array();
- if ($classname != "") {
- if ($classname{0} == '.')
+ if ($classname !== "") {
+ if ($classname[0] === '.') {
$classname = substr($classname, 1);
- $classes[] = $this->code_class_prefix.$classname;
+ }
+ $classes[] = $this->code_class_prefix . $classname;
}
$attr_str = $this->doExtraAttributes($this->code_attr_on_pre ? "pre" : "code", $attrs, null, $classes);
$pre_attr_str = $this->code_attr_on_pre ? $attr_str : '';
$code_attr_str = $this->code_attr_on_pre ? '' : $attr_str;
$codeblock = "$codeblock";
-
+
return "\n\n".$this->hashBlock($codeblock)."\n\n";
}
+
+ /**
+ * Replace new lines in fenced code blocks
+ * @param array $matches
+ * @return string
+ */
protected function _doFencedCodeBlocks_newlines($matches) {
- return str_repeat("empty_element_suffix",
+ return str_repeat("empty_element_suffix",
strlen($matches[0]));
}
-
- #
- # Redefining emphasis markers so that emphasis by underscore does not
- # work in the middle of a word.
- #
+ /**
+ * Redefining emphasis markers so that emphasis by underscore does not
+ * work in the middle of a word.
+ * @var array
+ */
protected $em_relist = array(
'' => '(?:(? '(? '(? '(?:(? '(? '(? '(?:(? '(? '(? tags
- #
- # Strip leading and trailing lines:
+ /**
+ * Parse text into paragraphs
+ * @param string $text String to process in paragraphs
+ * @param boolean $wrap_in_p Whether paragraphs should be wrapped in tags
+ * @return string HTML output
+ */
+ protected function formParagraphs($text, $wrap_in_p = true) {
+ // Strip leading and trailing lines:
$text = preg_replace('/\A\n+|\n+\z/', '', $text);
-
+
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
- #
- # Wrap
tags and unhashify HTML blocks
- #
+ // Wrap
tags and unhashify HTML blocks
foreach ($grafs as $key => $value) {
$value = trim($this->runSpanGamut($value));
-
- # Check if this should be enclosed in a paragraph.
- # Clean tag hashes & block tag hashes are left alone.
- $is_p = !preg_match('/^B\x1A[0-9]+B|^C\x1A[0-9]+C$/', $value);
-
+
+ // Check if this should be enclosed in a paragraph.
+ // Clean tag hashes & block tag hashes are left alone.
+ $is_p = $wrap_in_p && !preg_match('/^B\x1A[0-9]+B|^C\x1A[0-9]+C$/', $value);
+
if ($is_p) {
$value = "
$value
";
}
$grafs[$key] = $value;
}
-
- # Join grafs in one text, then unhash HTML tags.
+
+ // Join grafs in one text, then unhash HTML tags.
$text = implode("\n\n", $grafs);
-
- # Finish by removing any tag hashes still present in $text.
+
+ // Finish by removing any tag hashes still present in $text.
$text = $this->unhash($text);
-
+
return $text;
}
-
-
- ### Footnotes
-
+
+
+ /**
+ * Footnotes - Strips link definitions from text, stores the URLs and
+ * titles in hash references.
+ * @param string $text
+ * @return string
+ */
protected function stripFootnotes($text) {
- #
- # Strips link definitions from text, stores the URLs and titles in
- # hash references.
- #
$less_than_tab = $this->tab_width - 1;
- # Link defs are in the form: [^id]: url "optional title"
+ // Link defs are in the form: [^id]: url "optional title"
$text = preg_replace_callback('{
- ^[ ]{0,'.$less_than_tab.'}\[\^(.+?)\][ ]?: # note_id = $1
+ ^[ ]{0,' . $less_than_tab . '}\[\^(.+?)\][ ]?: # note_id = $1
[ ]*
\n? # maybe *one* newline
( # text = $2 (no blank lines allowed)
- (?:
+ (?:
.+ # actual text
|
- \n # newlines but
+ \n # newlines but
(?!\[.+?\][ ]?:\s)# negative lookahead for footnote or link definition marker.
- (?!\n+[ ]{0,3}\S)# ensure line is not blank and followed
+ (?!\n+[ ]{0,3}\S)# ensure line is not blank and followed
# by non-indented content
)*
- )
+ )
}xm',
array($this, '_stripFootnotes_callback'),
$text);
return $text;
}
+
+ /**
+ * Callback for stripping footnotes
+ * @param array $matches
+ * @return string
+ */
protected function _stripFootnotes_callback($matches) {
$note_id = $this->fn_id_prefix . $matches[1];
$this->footnotes[$note_id] = $this->outdent($matches[2]);
- return ''; # String that will replace the block
+ return ''; // String that will replace the block
}
-
+ /**
+ * Replace footnote references in $text [^id] with a special text-token
+ * which will be replaced by the actual footnote marker in appendFootnotes.
+ * @param string $text
+ * @return string
+ */
protected function doFootnotes($text) {
- #
- # Replace footnote references in $text [^id] with a special text-token
- # which will be replaced by the actual footnote marker in appendFootnotes.
- #
if (!$this->in_anchor) {
$text = preg_replace('{\[\^(.+?)\]}', "F\x1Afn:\\1\x1A:", $text);
}
return $text;
}
-
+ /**
+ * Append footnote list to text
+ * @param string $text
+ * @return string
+ */
protected function appendFootnotes($text) {
- #
- # Append footnote list to text.
- #
- $text = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}',
+ $text = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}',
array($this, '_appendFootnotes_callback'), $text);
-
- if (!empty($this->footnotes_ordered)) {
- $text .= "\n\n";
- $text .= "\n";
- $text .= "empty_element_suffix ."\n";
- $text .= "\n\n";
- $attr = "";
- if ($this->fn_backlink_class != "") {
- $class = $this->fn_backlink_class;
- $class = $this->encodeAttribute($class);
- $attr .= " class=\"$class\"";
+ if ( ! empty( $this->footnotes_ordered ) ) {
+ $this->_doFootnotes();
+ if ( ! $this->omit_footnotes ) {
+ $text .= "\n\n";
+ $text .= "\n";
+ $text .= "empty_element_suffix . "\n";
+ $text .= $this->footnotes_assembled;
+ $text .= " ";
}
- if ($this->fn_backlink_title != "") {
- $title = $this->fn_backlink_title;
- $title = $this->encodeAttribute($title);
- $attr .= " title=\"$title\"";
- }
- $backlink_text = $this->fn_backlink_html;
- $num = 0;
-
- while (!empty($this->footnotes_ordered)) {
- $footnote = reset($this->footnotes_ordered);
- $note_id = key($this->footnotes_ordered);
- unset($this->footnotes_ordered[$note_id]);
- $ref_count = $this->footnotes_ref_count[$note_id];
- unset($this->footnotes_ref_count[$note_id]);
- unset($this->footnotes[$note_id]);
-
- $footnote .= "\n"; # Need to append newline before parsing.
- $footnote = $this->runBlockGamut("$footnote\n");
- $footnote = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}',
- array($this, '_appendFootnotes_callback'), $footnote);
-
- $attr = str_replace("%%", ++$num, $attr);
- $note_id = $this->encodeAttribute($note_id);
+ }
+ return $text;
+ }
- # Prepare backlink, multiple backlinks if multiple references
- $backlink = "$backlink_text";
- for ($ref_num = 2; $ref_num <= $ref_count; ++$ref_num) {
- $backlink .= " $backlink_text";
+
+ /**
+ * Generates the HTML for footnotes. Called by appendFootnotes, even if
+ * footnotes are not being appended.
+ * @return void
+ */
+ protected function _doFootnotes() {
+ $attr = array();
+ if ($this->fn_backlink_class !== "") {
+ $class = $this->fn_backlink_class;
+ $class = $this->encodeAttribute($class);
+ $attr['class'] = " class=\"$class\"";
+ }
+ $attr['role'] = " role=\"doc-backlink\"";
+ $num = 0;
+
+ $text = "\n\n";
+ while (!empty($this->footnotes_ordered)) {
+ $footnote = reset($this->footnotes_ordered);
+ $note_id = key($this->footnotes_ordered);
+ unset($this->footnotes_ordered[$note_id]);
+ $ref_count = $this->footnotes_ref_count[$note_id];
+ unset($this->footnotes_ref_count[$note_id]);
+ unset($this->footnotes[$note_id]);
+
+ $footnote .= "\n"; // Need to append newline before parsing.
+ $footnote = $this->runBlockGamut("$footnote\n");
+ $footnote = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}',
+ array($this, '_appendFootnotes_callback'), $footnote);
+
+ $num++;
+ $note_id = $this->encodeAttribute($note_id);
+
+ // Prepare backlink, multiple backlinks if multiple references
+ // Do not create empty backlinks if the html is blank
+ $backlink = "";
+ if (!empty($this->fn_backlink_html)) {
+ for ($ref_num = 1; $ref_num <= $ref_count; ++$ref_num) {
+ if (!empty($this->fn_backlink_title)) {
+ $attr['title'] = ' title="' . $this->encodeAttribute($this->fn_backlink_title) . '"';
+ }
+ if (!empty($this->fn_backlink_label)) {
+ $attr['label'] = ' aria-label="' . $this->encodeAttribute($this->fn_backlink_label) . '"';
+ }
+ $parsed_attr = $this->parseFootnotePlaceholders(
+ implode('', $attr),
+ $num,
+ $ref_num
+ );
+ $backlink_text = $this->parseFootnotePlaceholders(
+ $this->fn_backlink_html,
+ $num,
+ $ref_num
+ );
+ $ref_count_mark = $ref_num > 1 ? $ref_num : '';
+ $backlink .= " $backlink_text";
}
- # Add backlink to last paragraph; create new paragraph if needed.
+ $backlink = trim($backlink);
+ }
+
+ // Add backlink to last paragraph; create new paragraph if needed.
+ if (!empty($backlink)) {
if (preg_match('{$}', $footnote)) {
$footnote = substr($footnote, 0, -4) . " $backlink";
} else {
$footnote .= "\n\n$backlink
";
}
-
- $text .= "- \n";
- $text .= $footnote . "\n";
- $text .= "
\n\n";
}
-
- $text .= "
\n";
- $text .= "
";
+
+ $text .= "\n";
+ $text .= $footnote . "\n";
+ $text .= " \n\n";
}
- return $text;
+ $text .= "\n";
+
+ $this->footnotes_assembled = $text;
}
+
+ /**
+ * Callback for appending footnotes
+ * @param array $matches
+ * @return string
+ */
protected function _appendFootnotes_callback($matches) {
$node_id = $this->fn_id_prefix . $matches[1];
-
- # Create footnote marker only if it has a corresponding footnote *and*
- # the footnote hasn't been used by another marker.
+
+ // Create footnote marker only if it has a corresponding footnote *and*
+ // the footnote hasn't been used by another marker.
if (isset($this->footnotes[$node_id])) {
$num =& $this->footnotes_numbers[$node_id];
if (!isset($num)) {
- # Transfer footnote content to the ordered list and give it its
- # number
+ // Transfer footnote content to the ordered list and give it its
+ // number
$this->footnotes_ordered[$node_id] = $this->footnotes[$node_id];
$this->footnotes_ref_count[$node_id] = 1;
$num = $this->footnote_counter++;
@@ -1540,86 +1777,117 @@ class MarkdownExtra extends \Michelf\Markdown {
}
$attr = "";
- if ($this->fn_link_class != "") {
+ if ($this->fn_link_class !== "") {
$class = $this->fn_link_class;
$class = $this->encodeAttribute($class);
$attr .= " class=\"$class\"";
}
- if ($this->fn_link_title != "") {
+ if ($this->fn_link_title !== "") {
$title = $this->fn_link_title;
$title = $this->encodeAttribute($title);
$attr .= " title=\"$title\"";
}
-
+ $attr .= " role=\"doc-noteref\"";
+
$attr = str_replace("%%", $num, $attr);
$node_id = $this->encodeAttribute($node_id);
-
+
return
"".
"$num".
"";
}
-
- return "[^".$matches[1]."]";
+
+ return "[^" . $matches[1] . "]";
}
-
-
- ### Abbreviations ###
-
+
+ /**
+ * Build footnote label by evaluating any placeholders.
+ * - ^^ footnote number
+ * - %% footnote reference number (Nth reference to footnote number)
+ * @param string $label
+ * @param int $footnote_number
+ * @param int $reference_number
+ * @return string
+ */
+ protected function parseFootnotePlaceholders($label, $footnote_number, $reference_number) {
+ return str_replace(
+ array('^^', '%%'),
+ array($footnote_number, $reference_number),
+ $label
+ );
+ }
+
+
+ /**
+ * Abbreviations - strips abbreviations from text, stores titles in hash
+ * references.
+ * @param string $text
+ * @return string
+ */
protected function stripAbbreviations($text) {
- #
- # Strips abbreviations from text, stores titles in hash references.
- #
$less_than_tab = $this->tab_width - 1;
- # Link defs are in the form: [id]*: url "optional title"
+ // Link defs are in the form: [id]*: url "optional title"
$text = preg_replace_callback('{
- ^[ ]{0,'.$less_than_tab.'}\*\[(.+?)\][ ]?: # abbr_id = $1
- (.*) # text = $2 (no blank lines allowed)
+ ^[ ]{0,' . $less_than_tab . '}\*\[(.+?)\][ ]?: # abbr_id = $1
+ (.*) # text = $2 (no blank lines allowed)
}xm',
array($this, '_stripAbbreviations_callback'),
$text);
return $text;
}
+
+ /**
+ * Callback for stripping abbreviations
+ * @param array $matches
+ * @return string
+ */
protected function _stripAbbreviations_callback($matches) {
$abbr_word = $matches[1];
$abbr_desc = $matches[2];
- if ($this->abbr_word_re)
+ if ($this->abbr_word_re) {
$this->abbr_word_re .= '|';
+ }
$this->abbr_word_re .= preg_quote($abbr_word);
$this->abbr_desciptions[$abbr_word] = trim($abbr_desc);
- return ''; # String that will replace the block
+ return ''; // String that will replace the block
}
-
-
+
+ /**
+ * Find defined abbreviations in text and wrap them in elements.
+ * @param string $text
+ * @return string
+ */
protected function doAbbreviations($text) {
- #
- # Find defined abbreviations in text and wrap them in elements.
- #
if ($this->abbr_word_re) {
- // cannot use the /x modifier because abbr_word_re may
+ // cannot use the /x modifier because abbr_word_re may
// contain significant spaces:
- $text = preg_replace_callback('{'.
- '(?abbr_word_re.')'.
- '(?![\w\x1A])'.
- '}',
+ $text = preg_replace_callback('{' .
+ '(?abbr_word_re . ')' .
+ '(?![\w\x1A])' .
+ '}',
array($this, '_doAbbreviations_callback'), $text);
}
return $text;
}
+
+ /**
+ * Callback for processing abbreviations
+ * @param array $matches
+ * @return string
+ */
protected function _doAbbreviations_callback($matches) {
$abbr = $matches[0];
if (isset($this->abbr_desciptions[$abbr])) {
$desc = $this->abbr_desciptions[$abbr];
if (empty($desc)) {
return $this->hashPart("$abbr");
- } else {
- $desc = $this->encodeAttribute($desc);
- return $this->hashPart("$abbr");
}
- } else {
- return $matches[0];
+ $desc = $this->encodeAttribute($desc);
+ return $this->hashPart("$abbr");
}
+ return $matches[0];
}
}
diff --git a/third/Md/MarkdownInterface.inc.php b/third/Md/MarkdownInterface.inc.php
index a023ed4..c4e9ac7 100644
--- a/third/Md/MarkdownInterface.inc.php
+++ b/third/Md/MarkdownInterface.inc.php
@@ -1,9 +1,9 @@
-#
-# Original Markdown
-# Copyright (c) 2004-2006 John Gruber
-#
+ * @copyright 2004-2019 Michel Fortin