Header...
+ + Paragraph: p. (also applied by default) + Example: p. Text ->Text
+ + Blockquote: bq. + Example: bq. Block quotation... ->Block quotation...+ + Blockquote with citation: bq.:http://citation.url + Example: bq.:http://textism.com/ Text... + ->
Text...+ + Footnote: fn(1-100). + Example: fn1. Footnote... ->
Footnote...
+ + Numeric list: #, ## + Consecutive paragraphs beginning with # are wrapped in ordered list tags. + Example:- ordered list
- unordered list
- term
- definition
computer code
+ %(bob)span% -> span
+
+ ==notextile== -> leave text alone (do not format)
+
+ "linktext":url -> linktext
+"linktext(title)":url -> linktext
+ "$":url -> url
+ "$(title)":url -> url
+
+ !imageurl! -> Scientists say1 the moon is small.
+ + + + The 'a b c' backlink characters can be altered too. + For example if you wanted the notes to have numeric backlinks starting from 1: + + notelist:1. + +Table syntax: + + Simple tables: + + |a|simple|table|row| + |And|Another|table|row| + |With an||empty|cell| + + |=. My table caption goes here + |_. A|_. table|_. header|_.row| + |A|simple|table|row| + + Note: Table captions *must* be the first line of the table else treated as a center-aligned cell. + + Tables with attributes: + + table{border:1px solid black}. My table summary here + {background:#ddd;color:red}. |{}| | | | + + To specify thead / tfoot / tbody groups, add one of these on its own line + above the row(s) you wish to wrap (you may specify attributes before the dot): + + |^. # thead + |-. # tbody + |~. # tfoot + + Column groups: + + |:\3. 100| + + Becomes: +paragraph
+ + p(#fluid). paragraph ->paragraph
+ + (classes and ids can be combined) + p(hector#fluid). paragraph ->paragraph
+ + Curly {brackets} insert arbitrary css style + + p{line-height:18px}. paragraph ->paragraph
+ + h3{color:red}. header 3 ->header 3
+ + Square [brackets] insert language attributes + + p[no]. paragraph ->paragraph
+ + %[fr]phrase% -> phrase + + Usually Textile block element syntax requires a dot and space before the block + begins, but since lists don't, they can be styled just using braces + + #{color:blue} one ->-
+ # big
- one + # list
- big +
- list +
+ * $parser = new \Netcarver\Textile\Parser();
+ * echo $parser->textileThis('h1. Hello World!');
+ *
+ *
+ * @see Parser::__construct()
+ * @see Parser::textileThis()
+ * @see Parser::textileRestricted()
+ */
+
+class Parser
+{
+ /**
+ * Version number.
+ *
+ * @var string
+ */
+
+ protected $ver = '3.5.5';
+
+ /**
+ * Regular expression snippets.
+ *
+ * @var array
+ */
+
+ protected $regex_snippets;
+
+ /**
+ * Pattern for horizontal align.
+ *
+ * @var string
+ */
+
+ protected $hlgn = "(?:\<(?!>)|<>|>|<|(?|\<\>|\=|[()]+(?! ))";
+
+ /**
+ * Pattern for vertical align.
+ *
+ * @var string
+ */
+
+ protected $vlgn = "[\-^~]";
+
+ /**
+ * Pattern for HTML classes and IDs.
+ *
+ * Does not allow classes/ids/languages/styles to span across
+ * newlines if used in a dotall regular expression.
+ *
+ * @var string
+ */
+
+ protected $clas = "(?:\([^)\n]+\))";
+
+ /**
+ * Pattern for language attribute.
+ *
+ * @var string
+ */
+
+ protected $lnge = "(?:\[[^]\n]+\])";
+
+ /**
+ * Pattern for style attribute.
+ *
+ * @var string
+ */
+
+ protected $styl = "(?:\{[^}\n]+\})";
+
+ /**
+ * Regular expression pattern for column spans in tables.
+ *
+ * @var string
+ */
+
+ protected $cspn = "(?:\\\\[0-9]+)";
+
+ /**
+ * Regular expression for row spans in tables.
+ *
+ * @var string
+ */
+
+ protected $rspn = "(?:\/[0-9]+)";
+
+ /**
+ * Regular expression for horizontal or vertical alignment.
+ *
+ * @var string
+ */
+
+ protected $a;
+
+ /**
+ * Regular expression for column or row spans in tables.
+ *
+ * @var string
+ */
+
+ protected $s;
+
+ /**
+ * Pattern that matches a class, style, language and horizontal alignment attributes.
+ *
+ * @var string
+ */
+
+ protected $c;
+
+ /**
+ * Pattern that matches class, style and language attributes.
+ *
+ * Will allows all 16 possible permutations of class, style and language attributes.
+ *
+ * $parser = new \Netcarver\Textile\Parser('html5');
+ * echo $parser->textileThis('HTML(HyperText Markup Language)");
+ *
+ *
+ * @param string $doctype The output document type, either 'xhtml' or 'html5'
+ * @throws \InvalidArgumentException
+ * @api
+ */
+
+ public function __construct($doctype = 'xhtml')
+ {
+ $doctypes = array(
+ 'xhtml',
+ 'html5',
+ );
+
+ if (!in_array($doctype, $doctypes, true)) {
+ throw new \InvalidArgumentException('Invalid doctype given.');
+ } else {
+ $this->doctype = $doctype;
+ }
+
+ $uid = uniqid(rand());
+ $this->uid = 'textileRef:'.$uid.':';
+ $this->linkPrefix = $uid.'-';
+ $this->a = "(?:$this->hlgn|$this->vlgn)*";
+ $this->s = "(?:$this->cspn|$this->rspn)*";
+ $this->c = "(?:$this->clas|$this->styl|$this->lnge|$this->hlgn)*";
+
+ $this->cls = '(?:'.
+ "$this->clas(?:".
+ "$this->lnge(?:$this->styl)?|$this->styl(?:$this->lnge)?".
+ ')?|'.
+ "$this->lnge(?:".
+ "$this->clas(?:$this->styl)?|$this->styl(?:$this->clas)?".
+ ')?|'.
+ "$this->styl(?:".
+ "$this->clas(?:$this->lnge)?|$this->lnge(?:$this->clas)?".
+ ')?'.
+ ')?';
+
+ if ($this->isUnicodePcreSupported()) {
+ $this->regex_snippets = array(
+ 'acr' => '\p{Lu}\p{Nd}',
+ 'abr' => '\p{Lu}',
+ 'nab' => '\p{Ll}',
+ 'wrd' => '(?:\p{L}|\p{M}|\p{N}|\p{Pc})',
+ 'mod' => 'u', // Make sure to mark the unicode patterns as such, Some servers seem to need this.
+ 'cur' => '\p{Sc}',
+ 'digit' => '\p{N}',
+ 'space' => '(?:\p{Zs}|\h|\v)',
+ 'char' => '(?:[^\p{Zs}\h\v])',
+ );
+ } else {
+ $this->regex_snippets = array(
+ 'acr' => 'A-Z0-9',
+ 'abr' => 'A-Z',
+ 'nab' => 'a-z',
+ 'wrd' => '\w',
+ 'mod' => '',
+ 'cur' => '',
+ 'digit' => '\d',
+ 'space' => '(?:\s|\h|\v)',
+ 'char' => '\S',
+ );
+ }
+ extract($this->regex_snippets);
+ $this->urlch = '['.$wrd.'"$\-_.+!*\'(),";\/?:@=&%#{}|\\^~\[\]`]';
+ $this->quote_starts = implode('|', array_map('preg_quote', array_keys($this->quotes)));
+
+ if (defined('DIRECTORY_SEPARATOR')) {
+ $this->ds = constant('DIRECTORY_SEPARATOR');
+ }
+
+ if (php_sapi_name() === 'cli') {
+ $this->doc_root = getcwd();
+ } elseif (!empty($_SERVER['DOCUMENT_ROOT'])) {
+ $this->doc_root = $_SERVER['DOCUMENT_ROOT'];
+ } elseif (!empty($_SERVER['PATH_TRANSLATED'])) {
+ $this->doc_root = $_SERVER['PATH_TRANSLATED'];
+ }
+
+ $this->doc_root = rtrim($this->doc_root, $this->ds).$this->ds;
+ }
+
+ /**
+ * Defines a substitution symbol.
+ *
+ * Call this you need to redefine a substitution symbol to
+ * be used when parsing a Textile document.
+ *
+ * @param string $name Name of the symbol to assign a new value to.
+ * @param string $value New value for the symbol.
+ * @return Parser
+ * @api
+ */
+
+ public function setSymbol($name, $value)
+ {
+ $this->symbols[$name] = $value;
+ $this->rebuild_glyphs = true;
+ return $this;
+ }
+
+ /**
+ * Gets a symbol definitions.
+ *
+ * This method can be used to get a symbol definition, or an
+ * array containing the full symbol table.
+ *
+ * @param string|null $name The name of the symbol, or NULL if requesting the symbol table
+ * @return array|string The symbol table or the requested symbol
+ * @throws \InvalidArgumentException
+ * @api
+ */
+
+ public function getSymbol($name = null)
+ {
+ if ($name !== null) {
+ if (isset($this->symbols[$name])) {
+ return $this->symbols[$name];
+ }
+
+ throw new \InvalidArgumentException('The specified name does not match any symbols.');
+ }
+
+ return $this->symbols;
+ }
+
+ /**
+ * Sets base image directory path.
+ *
+ * This is used when Textile is supplied with a relative image path.
+ * Allows client systems to have PHP-Textile convert relative image paths to
+ * absolute or prefixed paths. This method is used to set that base path,
+ * usually a absolute HTTP address pointing to a directory.
+ *
+ *
+ * $parser = new \Netcarver\Textile\Parser();
+ * $parser->setRelativeImagePrefix('http://static.example.com/');
+ *
+ *
+ * @param string $prefix The string to prefix all relative image paths with
+ * @return Parser
+ * @api
+ */
+
+ public function setRelativeImagePrefix($prefix = '')
+ {
+ $this->relativeImagePrefix = $prefix;
+ return $this;
+ }
+
+ /**
+ * Toggles image dimension attributes.
+ *
+ * If $dimensionless is set to TRUE, image width and height attributes
+ * will not be included in rendered image tags. Normally, Textile will add
+ * dimensions height images that specify a relative path, as long
+ * as the image file can be accessed.
+ *
+ *
+ * $parser = new \Netcarver\Textile\Parser();
+ * echo $parser->setDimensionlessImages(false)->textileThis('Hello World!');
+ *
+ *
+ * @param bool $dimensionless TRUE to disable image dimensions, FALSE to enable
+ * @return Parser
+ * @api
+ */
+
+ public function setDimensionlessImages($dimensionless = true)
+ {
+ $this->dimensionless_images = (bool) $dimensionless;
+ return $this;
+ }
+
+ /**
+ * Whether images will get dimensions or not.
+ *
+ * This method will return the state of
+ * the state of the $dimensionless_images property.
+ *
+ *
+ * $parser = new \Netcarver\Textile\Parser();
+ * if ($parser->getDimensionlessImages() === true)
+ * {
+ * echo 'Images do not get dimensions.';
+ * }
+ *
+ *
+ * @return bool TRUE if images will not get dimensions, FALSE otherwise
+ * @api
+ */
+
+ public function getDimensionlessImages()
+ {
+ return (bool) $this->dimensionless_images;
+ }
+
+ /**
+ * Gets Textile version number.
+ *
+ *
+ * $parser = new \Netcarver\Textile\Parser();
+ * echo $parser->getVersion();
+ *
+ *
+ * @return string Version
+ * @api
+ */
+
+ public function getVersion()
+ {
+ return $this->ver;
+ }
+
+ /**
+ * Encodes the given text.
+ *
+ *
+ * $parser = new \Netcarver\Textile\Parser();
+ * $parser->textileEncode('Some content to encode.');
+ *
+ *
+ * @param string $text The text to be encoded
+ * @return string The encoded text
+ * @api
+ */
+
+ public function textileEncode($text)
+ {
+ $text = preg_replace("/&(?![#a-z0-9]+;)/i", "x%x%", $text);
+ $text = str_replace("x%x%", "&", $text);
+ return $text;
+ }
+
+ /**
+ * Parses the given Textile input in un-restricted mode.
+ *
+ * This method should be used to parse any trusted Textile
+ * input, such as articles created by well-known
+ * authorised users.
+ *
+ * This method allows users to mix raw HTML and Textile.
+ * If you want to parse untrusted input, see the
+ * textileRestricted method instead. Using this less
+ * restrictive method on untrusted input, like comments
+ * and forum posts, will lead to XSS issues, as users
+ * will be able to use any HTML code, JavaScript links
+ * and Textile attributes in their input.
+ *
+ *
+ * $parser = new \Netcarver\Textile\Parser();
+ * echo $parser->textileThis('h1. Hello World!');
+ *
+ *
+ * @param string $text The Textile input to parse
+ * @param bool $lite Switch to lite mode
+ * @param bool $encode Encode input and return
+ * @param bool $noimage Disables images
+ * @param bool $strict This argument is ignored
+ * @param string $rel Relationship attribute applied to generated links
+ * @return string Parsed $text
+ * @see Parser::textileRestricted()
+ * @api
+ */
+
+ public function textileThis($text, $lite = false, $encode = false, $noimage = false, $strict = false, $rel = '')
+ {
+ $this->prepare($lite, $noimage, $rel);
+ $this->url_schemes = $this->unrestricted_url_schemes;
+
+ if ($encode) {
+ trigger_error(
+ 'Use of the $encode argument is discouraged. Use Parser::textileEncode() instead.',
+ E_USER_DEPRECATED
+ );
+ return $this->textileEncode($text);
+ }
+
+ return $this->textileCommon($text, $lite);
+ }
+
+ /**
+ * Parses the given Textile input in restricted mode.
+ *
+ * This method should be used for any untrusted user input,
+ * including comments or forum posts.
+ *
+ * This method escapes any raw HTML input, ignores unsafe
+ * attributes, links only whitelisted URL schemes
+ * and by default also prevents the use of images and
+ * extra Textile formatting, accepting only paragraphs
+ * and blockquotes as valid block tags.
+ *
+ *
+ * $parser = new \Netcarver\Textile\Parser();
+ * echo $parser->textileRestricted('h1. Hello World!');
+ *
+ *
+ * @param string $text The Textile input to parse
+ * @param bool $lite Controls lite mode, allowing extra formatting
+ * @param bool $noimage Allow images
+ * @param string $rel Relationship attribute applied to generated links
+ * @return string Parsed $text
+ * @see Parser::textileThis()
+ * @api
+ */
+
+ public function textileRestricted($text, $lite = true, $noimage = true, $rel = 'nofollow')
+ {
+ $this->prepare($lite, $noimage, $rel);
+ $this->url_schemes = $this->restricted_url_schemes;
+ $this->restricted = true;
+
+ // Escape any raw html
+ $text = $this->encodeHTML($text, 0);
+
+ return $this->textileCommon($text, $lite);
+ }
+
+ /**
+ * Parses Textile syntax.
+ *
+ * This method performs common parse actions.
+ *
+ * @param string $text The input to parses
+ * @param bool $lite Controls lite mode
+ * @return string Parsed input
+ */
+
+ protected function textileCommon($text, $lite)
+ {
+ $text = $this->cleanWhiteSpace($text);
+ $text = $this->cleanUniqueTokens($text);
+
+ if ($lite) {
+ $this->blocktag_whitelist = array('bq', 'p');
+ $text = $this->blocks($text."\n\n");
+ } else {
+ $this->blocktag_whitelist = array(
+ 'bq',
+ 'p',
+ 'bc',
+ 'notextile',
+ 'pre',
+ 'h[1-6]',
+ 'fn'.$this->regex_snippets['digit'].'+',
+ '###',
+ );
+ $text = $this->blocks($text);
+ $text = $this->placeNoteLists($text);
+ }
+
+ $text = $this->retrieve($text);
+ $text = $this->replaceGlyphs($text);
+ $text = $this->retrieveTags($text);
+ $text = $this->retrieveURLs($text);
+
+ $text = str_replace("", "
\n", $text); + + return $text; + } + + /** + * Prepares the glyph patterns from the symbol table. + * + * @see Parser::setSymbol() + * @see Parser::getSymbol() + */ + + protected function prepGlyphs() + { + if ($this->rebuild_glyphs === false) { + return; + } + + extract($this->symbols, EXTR_PREFIX_ALL, 'txt'); + extract($this->regex_snippets); + $pnc = '[[:punct:]]'; + + if ($cur) { + $cur = '(?:['.$cur.']'.$space.'*)?'; + } + + $this->glyph_search = array(); + $this->glyph_replace = array(); + + // Dimension sign + $this->glyph_search[] = '/([0-9]+[\])]?[\'"]? ?)[xX]( ?[\[(]?)(?=[+-]?'.$cur.'[0-9]*\.?[0-9]+)/'.$mod; + $this->glyph_replace[] = '$1'.$txt_dimension.'$2'; + + // Apostrophe + $this->glyph_search[] = '/('.$wrd.'|\))\'('.$wrd.')/'.$mod; + $this->glyph_replace[] = '$1'.$txt_apostrophe.'$2'; + + // Back in '88/the '90s but not in his '90s', '1', '1.' '10m' or '5.png' + $this->glyph_search[] = '/('.$space.')\'(\d+'.$wrd.'?)\b(?![.]?['.$wrd.']*?\')/'.$mod; + $this->glyph_replace[] = '$1'.$txt_apostrophe.'$2'; + + // Single open following open bracket + $this->glyph_search[] = "/([([{])'(?=\S)/".$mod; + $this->glyph_replace[] = '$1'.$txt_quote_single_open; + + // Single closing + $this->glyph_search[] = '/(\S)\'(?='.$space.'|'.$pnc.'|<|$)/'.$mod; + $this->glyph_replace[] = '$1'.$txt_quote_single_close; + + // Default single opening + $this->glyph_search[] = "/'/"; + $this->glyph_replace[] = $txt_quote_single_open; + + // Double open following an open bracket. Allows things like Hello ["(Mum) & dad"] + $this->glyph_search[] = '/([([{])"(?=\S)/'.$mod; + $this->glyph_replace[] = '$1'.$txt_quote_double_open; + + // Double closing + $this->glyph_search[] = '/(\S)"(?='.$space.'|'.$pnc.'|<|$)/'.$mod; + $this->glyph_replace[] = '$1'.$txt_quote_double_close; + + // Default double opening + $this->glyph_search[] = '/"/'; + $this->glyph_replace[] = $txt_quote_double_open; + + // 3+ uppercase acronym + $this->glyph_search[] = '/\b(['.$abr.']['.$acr.']{2,})\b(?:[(]([^)]*)[)])/'.$mod; + + if ($this->doctype === 'html5') { + $this->glyph_replace[] = '$1'; + } else { + $this->glyph_replace[] = '$1'; + } + + // 3+ uppercase + $this->glyph_search[] = '/('.$space.'|^|[>(;-])(['.$abr.']{3,})'. + '(['.$nab.']*)(?='.$space.'|'.$pnc.'|<|$)(?=[^">]*?(<|$))/'.$mod; + $this->glyph_replace[] = '$1'.$this->uid.':glyph:$2$3'; + + // Ellipsis + $this->glyph_search[] = '/([^.]?)\.{3}/'; + $this->glyph_replace[] = '$1'.$txt_ellipsis; + + // em dash + $this->glyph_search[] = '/--/'; + $this->glyph_replace[] = $txt_emdash; + + // en dash + $this->glyph_search[] = '/ - /'; + $this->glyph_replace[] = ' '.$txt_endash.' '; + + // Trademark + $this->glyph_search[] = '/(\b ?|'.$space.'|^)[([]TM[])]/i'.$mod; + $this->glyph_replace[] = '$1'.$txt_trademark; + + // Registered + $this->glyph_search[] = '/(\b ?|'.$space.'|^)[([]R[])]/i'.$mod; + $this->glyph_replace[] = '$1'.$txt_registered; + + // Copyright + $this->glyph_search[] = '/(\b ?|'.$space.'|^)[([]C[])]/i'.$mod; + $this->glyph_replace[] = '$1'.$txt_copyright; + + // 1/4 + $this->glyph_search[] = '/[([]1\/4[])]/'; + $this->glyph_replace[] = $txt_quarter; + + // 1/2 + $this->glyph_search[] = '/[([]1\/2[])]/'; + $this->glyph_replace[] = $txt_half; + + // 3/4 + $this->glyph_search[] = '/[([]3\/4[])]/'; + $this->glyph_replace[] = $txt_threequarters; + + // Degrees -- that's a small 'oh' + $this->glyph_search[] = '/[([]o[])]/'; + $this->glyph_replace[] = $txt_degrees; + + // Plus minus + $this->glyph_search[] = '/[([]\+\/-[])]/'; + $this->glyph_replace[] = $txt_plusminus; + + // No need to rebuild next run unless a symbol is redefined + $this->rebuild_glyphs = false; + } + + /** + * Sets the maximum allowd link index. + * + * @return int + * @since 3.5.5 + */ + + protected function getMaxLinkIndex() + { + return 1000000; + } + + /** + * Prepares the parser for parsing. + * + * This method prepares the transient internal state of + * Textile parser in preparation for parsing a new document. + * + * @param bool $lite Controls lite mode + * @param bool $noimage Disallow images + * @param string $rel A relationship attribute applied to links + */ + + protected function prepare($lite, $noimage, $rel) + { + if ($this->linkIndex >= $this->getMaxLinkIndex()) { + $this->linkPrefix .= '-'; + $this->linkIndex = 1; + } + + $this->unreferencedNotes = array(); + $this->notelist_cache = array(); + $this->notes = array(); + $this->urlrefs = array(); + $this->shelf = array(); + $this->fn = array(); + $this->span_depth = 0; + $this->refIndex = 1; + $this->refCache = array(); + $this->note_index = 1; + $this->rel = $rel; + $this->lite = $lite; + $this->noimage = $noimage; + $this->prepGlyphs(); + } + + /** + * Cleans a HTML attribute value. + * + * This method checks for presence of URL encoding in the value. + * If the number encoded characters exceeds the thereshold, + * the input is discarded. Otherwise the encoded + * instances are decoded. + * + * This method also strips any ", ' and = characters + * from the given value. This method does not guarantee + * valid HTML or full sanitization. + * + * @param string $in The input string + * @return string Cleaned string + */ + + protected function cleanAttribs($in) + { + $tmp = $in; + $before = -1; + $after = 0; + $max = 3; + $i = 0; + + while (($after != $before) && ($i < $max)) { + $before = strlen($tmp); + $tmp = rawurldecode($tmp); + $after = strlen($tmp); + $i++; + } + + if ($i === $max) { + // If we hit the max allowed decodes, assume the input is tainted and consume it. + $out = ''; + } else { + $out = str_replace(array('"', "'", '='), '', $tmp); + } + + return $out; + } + + /** + * Constructs a HTML tag from an object. + * + * This is a helper method that creates a new + * instance of \Netcarver\Textile\Tag. + * + * @param string $name The HTML element name + * @param array $atts HTML attributes applied to the tag + * @param bool $selfclosing Determines if the tag should be selfclosing + * @return Tag + */ + + protected function newTag($name, $atts, $selfclosing = true) + { + return new Tag($name, $atts, $selfclosing); + } + + /** + * Parses Textile attributes. + * + * @param string $in The Textile attribute string to be parsed + * @param string $element Focus the routine to interpret the attributes as applying to a specific HTML tag + * @param bool $include_id If FALSE, IDs are not included in the attribute list + * @param string $autoclass An additional classes applied to the output + * @return string HTML attribute list + * @see Parser::parseAttribsToArray() + */ + + protected function parseAttribs($in, $element = '', $include_id = true, $autoclass = '') + { + $o = $this->parseAttribsToArray($in, $element, $include_id, $autoclass); + + return $this->formatAttributeString($o); + } + + /** + * Converts an array of named attribute => value mappings to a string. + * + * @param array $attribute_array + * @return string + */ + + protected function formatAttributeString(array $attribute_array) + { + $out = ''; + + if (count($attribute_array)) { + foreach ($attribute_array as $k => $v) { + $out .= " $k=\"$v\""; + } + } + + return $out; + } + + /** + * Parses Textile attributes into an array. + * + * @param string $in The Textile attribute string to be parsed + * @param string $element Focus the routine to interpret the attributes as applying to a specific HTML tag + * @param bool $include_id If FALSE, IDs are not included in the attribute list + * @param string $autoclass An additional classes applied to the output + * @return array HTML attributes as key => value mappings + * @see Parser::parseAttribs() + */ + + protected function parseAttribsToArray($in, $element = '', $include_id = true, $autoclass = '') + { + $style = ''; + $class = ''; + $lang = ''; + $colspan = ''; + $rowspan = ''; + $span = ''; + $width = ''; + $id = ''; + $atts = ''; + $align = ''; + + $matched = $in; + if ($element == 'td') { + if (preg_match("/\\\\([0-9]+)/", $matched, $csp)) { + $colspan = $csp[1]; + } + + if (preg_match("/\/([0-9]+)/", $matched, $rsp)) { + $rowspan = $rsp[1]; + } + } + + if ($element == 'td' or $element == 'tr') { + if (preg_match("/($this->vlgn)/", $matched, $vert)) { + $style[] = "vertical-align:" . $this->vAlign($vert[1]); + } + } + + if (preg_match("/\{([^}]*)\}/", $matched, $sty)) { + if ($sty[1] = $this->cleanAttribs($sty[1])) { + $style[] = rtrim($sty[1], ';'); + } + $matched = str_replace($sty[0], '', $matched); + } + + if (preg_match("/\[([^]]+)\]/U", $matched, $lng)) { + // Consume entire lang block -- valid or invalid. + $matched = str_replace($lng[0], '', $matched); + if (preg_match("/\[([a-zA-Z]{2}(?:[\-\_][a-zA-Z]{2})?)\]/U", $lng[0], $lng)) { + $lang = $lng[1]; + } + } + + if (preg_match("/\(([^()]+)\)/U", $matched, $cls)) { + + $class_regex = "/^([-a-zA-Z 0-9_\.]*)$/"; + + // Consume entire class block -- valid or invalid. + $matched = str_replace($cls[0], '', $matched); + + // Only allow a restricted subset of the CSS standard characters for classes/ids. + // No encoding markers allowed. + if (preg_match("/\(([-a-zA-Z 0-9_\.\:\#]+)\)/U", $cls[0], $cls)) { + $hashpos = strpos($cls[1], '#'); + // If a textile class block attribute was found with a '#' in it + // split it into the css class and css id... + if (false !== $hashpos) { + if (preg_match("/#([-a-zA-Z0-9_\.\:]*)$/", substr($cls[1], $hashpos), $ids)) { + $id = $ids[1]; + } + + if (preg_match($class_regex, substr($cls[1], 0, $hashpos), $ids)) { + $class = $ids[1]; + } + } else { + if (preg_match($class_regex, $cls[1], $ids)) { + $class = $ids[1]; + } + } + } + } + + if (preg_match("/([(]+)/", $matched, $pl)) { + $style[] = "padding-left:" . strlen($pl[1]) . "em"; + $matched = str_replace($pl[0], '', $matched); + } + + if (preg_match("/([)]+)/", $matched, $pr)) { + $style[] = "padding-right:" . strlen($pr[1]) . "em"; + $matched = str_replace($pr[0], '', $matched); + } + + if (preg_match("/($this->hlgn)/", $matched, $horiz)) { + $style[] = "text-align:" . $this->hAlign($horiz[1]); + } + + if ($element == 'col') { + if (preg_match("/(?:\\\\([0-9]+))?{$this->regex_snippets['space']}*([0-9]+)?/", $matched, $csp)) { + $span = isset($csp[1]) ? $csp[1] : ''; + $width = isset($csp[2]) ? $csp[2] : ''; + } + } + + if ($this->restricted) { + $o = array(); + $class = trim($autoclass); + if ($class) { + $o['class'] = $this->cleanAttribs($class); + } + + if ($lang) { + $o['lang'] = $this->cleanAttribs($lang); + } + + ksort($o); + return $o; + } else { + $class = trim($class . ' ' . $autoclass); + } + + $o = array(); + if ($class) { + $o['class'] = $this->cleanAttribs($class); + } + + if ($colspan) { + $o['colspan'] = $this->cleanAttribs($colspan); + } + + if ($id && $include_id) { + $o['id'] = $this->cleanAttribs($id); + } + + if ($lang) { + $o['lang'] = $this->cleanAttribs($lang); + } + + if ($rowspan) { + $o['rowspan'] = $this->cleanAttribs($rowspan); + } + + if ($span) { + $o['span'] = $this->cleanAttribs($span); + } + + if ($style) { + $so = ''; + $tmps = array(); + foreach ($style as $s) { + $parts = explode(';', $s); + foreach ($parts as $p) { + if ($p = trim(trim($p), ":")) { + $tmps[] = $p; + } + } + } + + sort($tmps); + foreach ($tmps as $p) { + if (!empty($p)) { + $so .= $p.';'; + } + } + $style = trim(str_replace(array("\n", ';;'), array('', ';'), $so)); + + $o['style'] = $style; + } + + if ($width) { + $o['width'] = $this->cleanAttribs($width); + } + + ksort($o); + return $o; + } + + /** + * Checks whether the text is not enclosed by a block tag. + * + * @param string $text The input string + * @return bool TRUE if the text is not enclosed + */ + + protected function hasRawText($text) + { + $r = preg_replace( + '@<(p|hr|br|img|blockquote|div|form|table|ul|ol|dl|pre|h[1-6])[^>]*?'.chr(62).'.*]*?>@si', + '', + trim($text) + ); + $r = trim(preg_replace('@<(br|hr|img)[^>]*?/?>@i', '', trim($r))); + return '' != $r; + } + + /** + * Parses textile table structures into HTML. + * + * @param string $text The textile input + * @return string The parsed text + */ + + protected function tables($text) + { + $text = $text . "\n\n"; + return preg_replace_callback( + "/^(?:table(?P
- ';
+ }
+ }
+
+ if ('' != $term) {
+ $pos = strpos($def, "\n");
+ $def = str_replace("\n", "
- $def "; + } + } + } + } + $out[] = '
", trim($def)); + if (0 === $pos) { + $def = '
' . $def . '
'; + } + $term = str_replace("\n", "", $term); + + $term = $this->graf($term); + $def = $this->graf($def); + + $out[] = "\t
- so flag not to start another
+ $lists[$tl] = 2;
+ }
+
+ $tabs = str_repeat("\t", strlen($tl)-1);
+ $atts = $this->parseAttribs($atts);
+ if (!isset($lists[$tl])) {
+ $lists[$tl] = 1;
+ $line = "$tabs<" . $ltype . "l$atts$st>" . (($showitem) ? "\n$tabs\t<$litem>" . $content : '');
+ } else {
+ $line = ($showitem) ? "$tabs\t<$litem$atts>" . $content : '';
+ }
+
+ if ((strlen($nl) <= strlen($tl))) {
+ $line .= (($showitem) ? "" : '');
+ }
+
+ foreach (array_reverse($lists) as $k => $v) {
+ if (strlen($k) > strlen($nl)) {
+ $line .= ($v==2) ? '' : "\n$tabsliType($k) . "l>";
+ if ((strlen($k) > 1) && ($v != 2)) {
+ $line .= "";
+ }
+ unset($lists[$k]);
+ }
+ }
+ $pt = $tl; // Remember the current Textile tag
+ }
+
+ $out[] = $line;
+ }
+
+ $out = implode("\n", $out);
+ return $this->doTagBr($litem, $out);
+ }
+
+ /**
+ * Determines the list type from the Textile input symbol.
+ *
+ * @param string $in Textile input containing the possible list marker
+ * @return string Either 'd', 'o', 'u'
+ */
+
+ protected function liType($in)
+ {
+ $m = array();
+ $type = 'd';
+ if (preg_match('/^(?P
\n instances that are not followed by white-space, + // or at end, with single LF. + $content = preg_replace( + "~
. + $content = preg_replace("/\n(?![\s|])/", '
', $content); + return '<'.$m['tag'].$m['atts'].'>'.$content.$m['closetag']; + } + + /** + * Formats line breaks. + * + * @param string $m The input + * @return string + */ + + protected function fBr($m) + { + $content = preg_replace("@(.+)(?|
|||)\n(?![#*;:\s|])@", '$1
', $m['content']); + return '<'.$m['tag'].$m['atts'].'>'.$content.$m['closetag']; + } + + /** + * Splits the given input into blocks. + * + * Blocks are separated by double line-break boundaries, and processed + * the blocks one by one. + * + * @param string $text Textile source text + * @return string Input text with blocks processed + */ + + protected function blocks($text) + { + $regex = '/^(?P
', '
', $block); + + if ($ext && $anonymous_block) { + $out[count($out)-1] .= $block; + } elseif (!$eat) { + $out[] = $block; + } + + if ($eat) { + $eatWhitespace = true; + } else { + $whitespace = ''; + } + + } + + if ($ext) { + $out[count($out)-1] .= $c1; + } + + return join('', $out); + } + + /** + * Formats the given block. + * + * Adds block tags and formats the text content inside + * the block. + * + * @param string $m The block content to format + * @return array + */ + + protected function fBlock($m) + { + list(, $tag, $att, $ext, $cite, $content) = $m; + $atts = $this->parseAttribs($att); + $space = $this->regex_snippets['space']; + + $o1 = ''; + $o2 = ''; + $c2 = ''; + $c1 = ''; + $eat = false; + + if ($tag === 'p') { + // Is this an anonymous block with a note definition? + $notedef = preg_replace_callback( + "/ + ^note\# # start of note def marker + (?P