Skip to content
Permalink
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
5265 lines (4457 sloc) 152 KB
<?php
/**
* Textile - A Humane Web Text Generator.
*
* @link https://github.com/textile/php-textile
*/
/*
* Textile - A Humane Web Text Generator
*
* Copyright (c) 2003-2004, Dean Allen
* All rights reserved.
*
* Thanks to Carlo Zottmann <carlo@g-blog.net> for refactoring
* Textile's procedural code into a class framework
*
* Additions and fixes Copyright (c) 2006 Alex Shiels https://twitter.com/tellyworth
* Additions and fixes Copyright (c) 2010 Stef Dawson http://stefdawson.com/
* Additions and fixes Copyright (c) 2010-17 Netcarver https://github.com/netcarver
* Additions and fixes Copyright (c) 2011 Jeff Soo http://ipsedixit.net/
* Additions and fixes Copyright (c) 2012 Robert Wetzlmayr http://wetzlmayr.com/
* Additions and fixes Copyright (c) 2012-19 Jukka Svahn http://rahforum.biz/
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
*
* * Redistributions in binary form must reproduce the above copyright notice,
* this list of conditions and the following disclaimer in the documentation
* and/or other materials provided with the distribution.
*
* * Neither the name Textile nor the names of its contributors may be used to
* endorse or promote products derived from this software without specific
* prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
/*
Textile usage examples.
Block modifier syntax:
Header: h(1-6).
Paragraphs beginning with 'hn. ' (where n is 1-6) are wrapped in header tags.
Example: h1. Header... -> <h1>Header...</h1>
Paragraph: p. (also applied by default)
Example: p. Text -> <p>Text</p>
Blockquote: bq.
Example: bq. Block quotation... -> <blockquote>Block quotation...</blockquote>
Blockquote with citation: bq.:http://citation.url
Example: bq.:http://example.com/ Text...
-> <blockquote cite="http://example.com">Text...</blockquote>
Footnote: fn(1-100).
Example: fn1. Footnote... -> <p id="fn1">Footnote...</p>
Numeric list: #, ##
Consecutive paragraphs beginning with # are wrapped in ordered list tags.
Example: <ol><li>ordered list</li></ol>
Bulleted list: *, **
Consecutive paragraphs beginning with * are wrapped in unordered list tags.
Example: <ul><li>unordered list</li></ul>
Definition list:
Terms ;, ;;
Definitions :, ::
Consecutive paragraphs beginning with ; or : are wrapped in definition list tags.
Example: <dl><dt>term</dt><dd>definition</dd></dl>
Redcloth-style Definition list:
- Term1 := Definition1
- Term2 := Extended
definition =:
Phrase modifier syntax:
_emphasis_ -> <em>emphasis</em>
__italic__ -> <i>italic</i>
*strong* -> <strong>strong</strong>
**bold** -> <b>bold</b>
??citation?? -> <cite>citation</cite>
-deleted text- -> <del>deleted</del>
+inserted text+ -> <ins>inserted</ins>
^superscript^ -> <sup>superscript</sup>
~subscript~ -> <sub>subscript</sub>
@code@ -> <code>computer code</code>
%(bob)span% -> <span class="bob">span</span>
==notextile== -> leave text alone (do not format)
"linktext":url -> <a href="url">linktext</a>
"linktext(title)":url -> <a href="url" title="title">linktext</a>
"$":url -> <a href="url">url</a>
"$(title)":url -> <a href="url" title="title">url</a>
!imageurl! -> <img src="imageurl" />
!imageurl(alt text)! -> <img src="imageurl" alt="alt text" />
!imageurl!:linkurl -> <a href="linkurl"><img src="imageurl" /></a>
ABC(Always Be Closing) -> <acronym title="Always Be Closing">ABC</acronym>
Linked Notes:
Allows the generation of an automated list of notes with links.
Linked notes are composed of three parts, a set of named _definitions_, a set of
_references_ to those definitions and one or more _placeholders_ indicating where
the consolidated list of notes is to be placed in your document.
Definitions:
Each note definition must occur in its own paragraph and should look like this...
note#mynotelabel. Your definition text here.
You are free to use whatever label you wish after the # as long as it is made up
of letters, numbers, colon(:) or dash(-).
References:
Each note reference is marked in your text like this[#mynotelabel] and
it will be replaced with a superscript reference that links into the list of
note definitions.
List placeholder(s):
The note list can go anywhere in your document. You have to indicate where
like this:
notelist.
notelist can take attributes (class#id) like this: notelist(class#id).
By default, the note list will show each definition in the order that they
are referenced in the text by the _references_. It will show each definition with
a full list of backlinks to each reference. If you do not want this, you can choose
to override the backlinks like this...
notelist(class#id)!. Produces a list with no backlinks.
notelist(class#id)^. Produces a list with only the first backlink.
Should you wish to have a specific definition display backlinks differently to this
then you can override the backlink method by appending a link override to the
_definition_ you wish to customise.
note#label. Uses the citelist's setting for backlinks.
note#label!. Causes that definition to have no backlinks.
note#label^. Causes that definition to have one backlink (to the first ref.)
note#label*. Causes that definition to have all backlinks.
Any unreferenced notes will be left out of the list unless you explicitly state
you want them by adding a '+'. Like this...
notelist(class#id)!+. Giving a list of all notes without any backlinks.
You can mix and match the list backlink control and unreferenced links controls
but the backlink control (if any) must go first. Like so: notelist^+. , not
like this: notelist+^.
Example...
Scientists say[#lavader] the moon is small.
note#other. An unreferenced note.
note#lavader(myliclass). "Proof":http://example.com of a small moon.
notelist(myclass#myid)+.
Would output (the actual IDs used would be randomised)...
<p>Scientists say<sup><a href="#note1" id="noteref1">1</sup> the moon is small.</p>
<ol class="myclass" id="myid">
<li class="myliclass"><a href="#noteref1"><sup>a</sup></a>
<span id="note1"> </span><a href="http://example.com">Proof</a> of a small moon.</li>
<li>An unreferenced note.</li>
</ol>
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:
<colgroup span="3" width="100"></colgroup>
You can omit either or both of the \N or width values. You may also
add cells after the colgroup definition to specify col elements with
span, width, or standard Textile attributes:
|:. 50|(firstcol). |\2. 250||300|
Becomes:
<colgroup width="50">
<col class="firstcol" />
<col span="2" width="250" />
<col />
<col width="300" />
</colgroup>
(Note that, per the HTML specification, you should not add span
to the colgroup if specifying col elements.)
Applying Attributes:
Most anywhere Textile code is used, attributes such as arbitrary css style,
css classes, and ids can be applied. The syntax is fairly consistent.
The following characters quickly alter the alignment of block elements:
< -> left align ex. p<. left-aligned para
> -> right align h3>. right-aligned header 3
= -> centred h4=. centred header 4
<> -> justified p<>. justified paragraph
These will change vertical alignment in table cells:
^ -> top ex. |^. top-aligned table cell|
- -> middle |-. middle aligned|
~ -> bottom |~. bottom aligned cell|
Plain (parentheses) inserted between block syntax and the closing dot-space
indicate classes and ids:
p(hector). paragraph -> <p class="hector">paragraph</p>
p(#fluid). paragraph -> <p id="fluid">paragraph</p>
(classes and ids can be combined)
p(hector#fluid). paragraph -> <p class="hector" id="fluid">paragraph</p>
Curly {brackets} insert arbitrary css style
p{line-height:18px}. paragraph -> <p style="line-height:18px">paragraph</p>
h3{color:red}. header 3 -> <h3 style="color:red">header 3</h3>
Square [brackets] insert language attributes
p[no]. paragraph -> <p lang="no">paragraph</p>
%[fr]phrase% -> <span lang="fr">phrase</span>
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 -> <ol style="color:blue">
# big <li>one</li>
# list <li>big</li>
<li>list</li>
</ol>
Using the span tag to style a phrase
It goes like this, %{color:red}the fourth the fifth%
-> It goes like this, <span style="color:red">the fourth the fifth</span>
Ordered list start and continuation:
You can control the start attribute of an ordered list like so;
#5 Item 5
# Item 6
You can resume numbering list items after some intervening anonymous block like so...
#_ Item 7
# Item 8
*/
namespace Netcarver\Textile;
/**
* Textile parser.
*
* The Parser class takes Textile input and converts it to well formatted HTML.
* This is the library's main class, hosting the parsing functionality and
* exposing a simple public interface for you to use.
*
* The most basic use case would involve initialising a instance of the class
* and calling the Parser::parse() method:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->parse('h1. Hello World!');
*
* The above generates:
*
* bc. <h1>Hello World!</h1>
*
* The functionality of the parser can be customized with the setters:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* $parser->setImages(false)->parse('!no-image.jpg!');
*
* The Parser class can also be extended to create pre-configured classes:
*
* bc.. namespace MyApp;
*
* use \Netcarver\Textile\Parser;
*
* class CommentParser extends Parser
* {
* protected function configure()
* {
* $this->setImages(false)->setRestricted(true)->setLite(true);
* }
* }
*
* p. Keep in mind that the classes' protected methods and properties should be
* considered part of the private API and depending on them should be avoided.
* Instead try to only use the public methods marked as being part of the
* public API.
*
* @see Parser::__construct()
* @see Parser::parse()
*/
class Parser
{
/**
* Version number.
*
* @var string
*/
protected $ver = '3.7.2-dev';
/**
* Regular expression snippets.
*
* @var array
*/
protected $regex_snippets;
/**
* Pattern for horizontal align.
*
* @var string
*/
protected $hlgn = "(?:\<(?!>)|&lt;&gt;|&gt;|&lt;|(?<!<)\>|\<\>|\=|[()]+(?! ))";
/**
* 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.
*
* Allows all 16 possible permutations of class, style and language attributes.
* No attribute, c, cl, cs, cls, csl, l, lc, ls, lcs, lsc, s, sc, sl, scl or slc.
*
* @var string
*/
protected $cls;
/**
* Whitelisted block tags.
*
* @var array
*/
protected $blocktag_whitelist = array();
/**
* Whether raw blocks are enabled.
*
* @var bool
* @since 3.7.0
*/
protected $rawBlocksEnabled = false;
/**
* An array of patterns used for matching phrasing tags.
*
* Phrasing tags, unline others, are wrapped in a paragraph even if they
* already wrap the block.
*
* @var array
* @since 3.7.0
*/
protected $phrasingContent = array(
'a',
'abbr',
'acronym',
'area',
'audio',
'b',
'bdo',
'br',
'button',
'canvas',
'cite',
'code',
'command',
'data',
'datalist',
'del',
'dfn',
'em',
'embed',
'i',
'iframe',
'img',
'input',
'ins',
'kbd',
'keygen',
'label',
'link',
'map',
'mark',
'math',
'meta',
'meter',
'noscript',
'object',
'output',
'progress',
'q',
'ruby',
'samp',
'script',
'select',
'small',
'span',
'strong',
'sub',
'sup',
'svg',
'textarea',
'time',
'var',
'video',
'wbr',
);
/**
* An array of patterns used to match divider tags.
*
* Blocks containing only self-closing divider tags are not wrapped in
* paragraph tags.
*
* @var array
* @since 3.7.0
*/
protected $dividerContent = array(
'br',
'hr',
'img',
);
/**
* An array of patterns used to match unwrappable block tags.
*
* Blocks containing any of these unwrappable tags will not be wrapped in
* paragraphs.
*
* @var array
* @since 3.7.0
*/
protected $blockContent = array(
'address',
'article',
'aside',
'blockquote',
'details',
'div',
'dl',
'fieldset',
'figure',
'footer',
'form',
'h1',
'h2',
'h3',
'h4',
'h5',
'h6',
'header',
'hgroup',
'main',
'menu',
'nav',
'ol',
'p',
'pre',
's',
'section',
'table',
'template',
'ul',
);
/**
* An array of built patterns.
*
* @var array
* @since 3.7.0
*/
protected $patterns;
/**
* Whether block tags are enabled.
*
* @var bool
* @since 3.6.0
*/
protected $blockTagsEnabled = true;
/**
* Whether lines are wrapped.
*
* @var bool
* @since 3.6.0
*/
protected $lineWrapEnabled = true;
/**
* Pattern for punctation.
*
* @var string
*/
protected $pnct = '[\!"#\$%&\'()\*\+,\-\./:;<=>\?@\[\\\]\^_`{\|}\~]';
/**
* Pattern for URL.
*
* @var string
*/
protected $urlch;
/**
* Matched marker symbols.
*
* @var string
*/
protected $syms = '¤§µ¶†‡•∗∴◊♠♣♥♦';
/**
* HTML rel attribute used for links.
*
* @var string
*/
protected $rel = '';
/**
* Array of footnotes.
*
* @var array
*/
protected $fn;
/**
* Shelved content.
*
* Stores fragments of the source text that have been parsed
* and require no more processing.
*
* @var array
*/
protected $shelf = array();
/**
* Restricted mode.
*
* @var bool
*/
protected $restricted = false;
/**
* Disallow images.
*
* @var bool
*/
protected $noimage = false;
/**
* Lite mode.
*
* @var bool
*/
protected $lite = false;
/**
* Accepted link protocols.
*
* @var array
*/
protected $url_schemes = array();
/**
* Restricted link protocols.
*
* @var array
*/
protected $restricted_url_schemes = array(
'http',
'https',
'ftp',
'mailto',
);
/**
* Unrestricted link protocols.
*
* @var array
*/
protected $unrestricted_url_schemes = array(
'http',
'https',
'ftp',
'mailto',
'file',
'tel',
'callto',
'sftp',
);
/**
* Span tags.
*
* @var array
*/
protected $span_tags = array(
'*' => 'strong',
'**' => 'b',
'??' => 'cite',
'_' => 'em',
'__' => 'i',
'-' => 'del',
'%' => 'span',
'+' => 'ins',
'~' => 'sub',
'^' => 'sup',
);
/**
* Patterns for finding glyphs.
*
* An array of regex patterns used to find text features
* such as apostrophes, fractions and em-dashes. Each
* entry in this array must have a corresponding entry in
* the $glyph_replace array.
*
* @var array
* @see Parser::$glyph_replace
*/
protected $glyph_search = array();
/**
* Glyph replacements.
*
* An array of replacements used to insert typographic glyphs
* into the text. Each entry must have a corresponding entry in
* the $glyph_search array and may refer to values captured in
* the corresponding search regex.
*
* @var array
* @see Parser::$glyph_search
*/
protected $glyph_replace = array();
/**
* Indicates whether glyph substitution is required.
*
* Dirty flag, set by Parser::setSymbol(), indicating the parser needs to
* rebuild the glyph substitutions before the next parse.
*
* @var bool
* @see Parser::setSymbol()
*/
protected $rebuild_glyphs = true;
/**
* Relative image path.
*
* @var string
* @deprecated in 3.7.0
* @see Parser::$relImagePrefix
* @see Parser::$relLinkPrefix
*/
protected $relativeImagePrefix;
/**
* Relative link prefix.
*
* @var string
* @since 3.7.0
*/
protected $relLinkPrefix = '';
/**
* Prefix applied to relative images.
*
* @var string
* @since 3.7.0
*/
protected $relImagePrefix = '';
/**
* Maximum nesting level for inline elements.
*
* @var int
*/
protected $max_span_depth = 5;
/**
* Server document root.
*
* @var string
*/
protected $doc_root;
/**
* Target document type.
*
* @var string
*/
protected $doctype;
/**
* An array of supported doctypes.
*
* @var array
* @since 3.6.0
*/
protected $doctypes = array(
'xhtml',
'html5',
);
/**
* Substitution symbols.
*
* Basic symbols used in textile glyph replacements. To override these, call
* setSymbol method before calling Parser::parse().
*
* @var array
* @see Parser::setSymbol()
* @see Parser::parse()
*/
protected $symbols = array(
'quote_single_open' => '&#8216;',
'quote_single_close' => '&#8217;',
'quote_double_open' => '&#8220;',
'quote_double_close' => '&#8221;',
'apostrophe' => '&#8217;',
'prime' => '&#8242;',
'prime_double' => '&#8243;',
'ellipsis' => '&#8230;',
'emdash' => '&#8212;',
'endash' => '&#8211;',
'dimension' => '&#215;',
'trademark' => '&#8482;',
'registered' => '&#174;',
'copyright' => '&#169;',
'half' => '&#189;',
'quarter' => '&#188;',
'threequarters' => '&#190;',
'degrees' => '&#176;',
'plusminus' => '&#177;',
'fn_ref_pattern' => '<sup{atts}>{marker}</sup>',
'fn_foot_pattern' => '<sup{atts}>{marker}</sup>',
'nl_ref_pattern' => '<sup{atts}>{marker}</sup>',
'caps' => '<span class="caps">{content}</span>',
'acronym' => null,
);
/**
* Dimensionless images flag.
*
* @var bool
*/
protected $dimensionless_images = false;
/**
* Directory separator.
*
* @var string
*/
protected $ds = '/';
/**
* Whether mbstring extension is installed.
*
* @var bool
*/
protected $mb;
/**
* Multi-byte conversion map.
*
* @var array
*/
protected $cmap = array(0x0080, 0xffff, 0, 0xffff);
/**
* Stores note index.
*
* @var int
*/
protected $note_index = 1;
/**
* Stores unreferenced notes.
*
* @var array
*/
protected $unreferencedNotes = array();
/**
* Stores note lists.
*
* @var array
*/
protected $notelist_cache = array();
/**
* Stores notes.
*
* @var array
*/
protected $notes = array();
/**
* Stores URL references.
*
* @var array
*/
protected $urlrefs = array();
/**
* Stores span depth.
*
* @var int
*/
protected $span_depth = 0;
/**
* Unique ID used for reference tokens.
*
* @var string
*/
protected $uid;
/**
* Token reference index.
*
* @var int
*/
protected $refIndex = 1;
/**
* Stores references values.
*
* @var array
*/
protected $refCache = array();
/**
* Matched open and closed quotes.
*
* @var array
*/
protected $quotes = array(
'"' => '"',
"'" => "'",
'(' => ')',
'{' => '}',
'[' => ']',
'«' => '»',
'»' => '«',
'' => '',
'' => '',
'' => '',
'' => '',
'' => '',
'' => '',
);
/**
* Regular expression that matches starting quotes.
*
* @var string
*/
protected $quote_starts;
/**
* Ordered list starts.
*
* @var array
*/
protected $olstarts = array();
/**
* Link prefix.
*
* @var string
*/
protected $linkPrefix;
/**
* Link index.
*
* @var int
*/
protected $linkIndex = 1;
/**
* Constructor.
*
* The constructor allows setting options that affect the class instance as
* a whole, such as the output doctype. To instruct the parser to return
* HTML5 markup instead of XHTML, set $doctype argument to 'html5'.
*
* bc. $parser = new \Netcarver\Textile\Parser('html5');
* echo $parser->parse('HTML(HyperText Markup Language)");
*
* @param string $doctype The output document type, either 'xhtml' or 'html5'
* @throws \InvalidArgumentException
* @see Parser::configure()
* @see Parser::parse()
* @see Parser::setDocumentType()
* @api
*/
public function __construct($doctype = 'xhtml')
{
$this->setDocumentType($doctype)->setRestricted(false);
$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',
);
}
$this->urlch = '['.$this->regex_snippets['wrd'].'"$\-_.+!*\'(),";\/?:@=&%#{}|\\^~\[\]`]';
$this->quote_starts = implode('|', array_map('preg_quote', array_keys($this->quotes)));
if (defined('DIRECTORY_SEPARATOR')) {
$this->ds = DIRECTORY_SEPARATOR;
}
if (php_sapi_name() === 'cli') {
if (($cwd = getcwd()) !== false) {
$this->setDocumentRootDirectory($cwd);
}
} elseif (!empty($_SERVER['DOCUMENT_ROOT'])) {
$this->setDocumentRootDirectory($_SERVER['DOCUMENT_ROOT']);
} elseif (!empty($_SERVER['PATH_TRANSLATED'])) {
$this->setDocumentRootDirectory($_SERVER['PATH_TRANSLATED']);
}
$this->configure();
}
/**
* Configure the current parser.
*
* This method can be extended to create a pre-configured parser class.
*
* bc.. namespace MyApp;
*
* use Netcarver\Textile\Parser;
*
* class CommentParser extends Parser
* {
* protected function configure()
* {
* $this->setImages(false)->setRestricted(true)->setLite(true);
* }
* }
*
* @since 3.7.0
* @return void Return value is ignored
* @api
*/
protected function configure()
{
}
/**
* Sets the output document type.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setDocumentType('html5')
* ->parse('HTML(HyperText Markup Language)");
*
* @param string $doctype Either 'xhtml' or 'html5'
* @return Parser This instance
* @since 3.6.0
* @see Parser::getDocumentType()
* @api
*/
public function setDocumentType($doctype)
{
if (in_array($doctype, $this->doctypes, true)) {
if ($this->getDocumentType() !== $doctype) {
$this->doctype = $doctype;
$this->rebuild_glyphs = true;
}
return $this;
}
throw new \InvalidArgumentException('Invalid doctype given.');
}
/**
* Gets the current output document type.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->getDocumentType();
*
* @return string The document type
* @since 3.6.0
* @see Parser::setDocumentType()
* @api
*/
public function getDocumentType()
{
return $this->doctype;
}
/**
* Sets the document root directory path.
*
* This method sets the path that is used to resolve relative file paths
* within local filesystem. This is used to fetch image dimensions, for
* instance.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* $parser->setDocumentRootDirectory('/path/to/document/root/dir');
*
* If not set, document root defaults to the current working directory if
* PHP-Textile is used via CLI. On server environment, DOCUMENT_ROOT or
* PATH_TRANSLATED server variable is used based on which ever is available.
*
* @param string $path The root path
* @return Parser This instance
* @since 3.6.0
* @see Parser::getDocumentRootDirectory()
* @api
*/
public function setDocumentRootDirectory($path)
{
$this->doc_root = rtrim($path, '\\/').$this->ds;
return $this;
}
/**
* Gets the current document root directory path.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->getDocumentRootDirectory();
*
* @return string Path to the document root directory
* @since 3.6.0
* @see Parser::setDocumentRootDirectory()
* @api
*/
public function getDocumentRootDirectory()
{
return $this->doc_root;
}
/**
* Enables lite mode.
*
* If enabled, allowed tags are limited. Parser will prevent the use extra
* Textile formatting, accepting only paragraphs and blockquotes as valid
* block tags.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* $parser
* ->setLite(true)
* ->parse('h1. Headings are disabled too');
*
* Generates:
*
* bc. <p>h1. Headings are disabled too</p>
*
* This doesn't prevent unsafe input values. If you wish to parse untrusted
* user-given Textile input, also enable the restricted parser mode with
* Parser::setRestricted().
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setRestricted(true)
* ->setLite(true)
* ->parse('h1. Hello World!');
*
* @param bool $lite TRUE to enable
* @return Parser This instance
* @since 3.6.0
* @see Parser::isLiteModeEnabled()
* @see Parser::setRestricted()
* @api
*/
public function setLite($lite)
{
$this->lite = (bool) $lite;
return $this;
}
/**
* Gets the lite mode status.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* if ($parser->isLiteModeEnabled() === true) {
* echo 'Lite mode is enabled.';
* }
*
* @return bool TRUE if enabled, FALSE otherwise
* @since 3.6.0
* @see Parser::setLite()
* @api
*/
public function isLiteModeEnabled()
{
return (bool) $this->lite;
}
/**
* Disables and enables images.
*
* If disabled, image tags are not generated. This option is ideal for
* minimalist output such as text-only comments.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setImages(true)
* ->parse('!image.png!');
*
* Generates:
*
* bc. <p>!image.png!</p>
*
* @param bool $enabled TRUE to enable, FALSE to disable
* @return Parser This instance
* @since 3.6.0
* @see Parser::isImageTagEnabled()
* @api
*/
public function setImages($enabled)
{
$this->noimage = !$enabled;
return $this;
}
/**
* Whether images are enabled.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* if ($parser->isImageTagEnabled() === true) {
* echo 'Images are enabled.';
* }
*
* @return bool TRUE if enabled, FALSE otherwise
* @since 3.6.0
* @see Parser::setImages()
* @api
*/
public function isImageTagEnabled()
{
return !$this->noimage;
}
/**
* Sets link relationship status value.
*
* This method sets the HTML relationship tokens that are applied to links
* generated by PHP-Textile.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setLinkRelationShip('nofollow')
* ->parse('"Link":http://example.com/');
*
* Generates:
*
* bc. <p><a href="http://example.com/" rel="nofollow">Link</a></p>
*
* @param string|array $relationship The HTML rel attribute value
* @return Parser This instance
* @since 3.6.0
* @see Parser::getLinkRelationShip()
* @api
*/
public function setLinkRelationShip($relationship)
{
$this->rel = (string) implode(' ', (array) $relationship);
return $this;
}
/**
* Gets the link relationship status value.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parse
* ->setLinkRelationShip('nofollow')
* ->getLinkRelationShip();
*
* The above outputs "nofollow".
*
* @return string The value
* @since 3.6.0
* @see Parser::setLinkRelationShip()
* @api
*/
public function getLinkRelationShip()
{
return $this->rel;
}
/**
* Enables restricted parser mode.
*
* This option should be enabled when parsing untrusted user input,
* including comments or forum posts. When enabled, the parser escapes any
* raw HTML input, ignores unsafe attributes and links only whitelisted URL
* schemes.
*
* For instance the following malicious input:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setRestricted(true)
* ->parse('Innocent _looking_ "link":javacript:window.alert().');
*
* Returns safe, sanitized HTML with valid Textile input still parsed:
*
* bc. <p>Innocent <em>looking</em> &#8220;link&#8221;:javacript:window.alert().</p>
*
* If left disabled, the parser allows users to mix raw HTML and Textile.
* Using the parser in non-restricted 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.
*
* @param bool $enabled TRUE to enable, FALSE to disable
* @return Parser This instance
* @since 3.6.0
* @see Parser::isRestrictedModeEnabled()
* @api
*/
public function setRestricted($enabled)
{
if ($enabled) {
$this->url_schemes = $this->restricted_url_schemes;
$this->restricted = true;
} else {
$this->url_schemes = $this->unrestricted_url_schemes;
$this->restricted = false;
}
return $this;
}
/**
* Whether restricted parser mode is enabled.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* if ($parser->isRestrictedModeEnabled() === true) {
* echo 'PHP-Textile is in restricted mode.';
* }
*
* @return bool TRUE if enabled, FALSE otherwise
* @since 3.6.0
* @see Parser::setRestricted()
* @api
*/
public function isRestrictedModeEnabled()
{
return (bool) $this->restricted;
}
/**
* Enables and disables raw blocks.
*
* When raw blocks are enabled, any paragraph blocks wrapped in a tag
* not matching Parser::$blockContent or Parser::$phrasingContent will not
* be parsed, and instead is left as is.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setRawBlocks(true)
* ->parse('<div>A *raw* block.</div>');
*
* The above generates:
*
* bc. <div>A *raw* block.</div>
*
* @param bool $enabled TRUE to enable, FALSE to disable
* @return Parser This instance
* @since 3.7.0
* @see Parser::isRawBlocksEnabled()
* @see Parser::isRawBlock()
* @api
*/
public function setRawBlocks($enabled)
{
$this->rawBlocksEnabled = (bool) $enabled;
return $this;
}
/**
* Whether raw blocks are enabled.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* if ($parser->isRawBlocksEnabled() === true) {
* echo 'Raw blocks are enabled';
* }
*
* @return bool TRUE if enabled, FALSE otherwise
* @since 3.7.0
* @see Parser::setRawBlocks()
* @api
*/
public function isRawBlocksEnabled()
{
return (bool) $this->rawBlocksEnabled;
}
/**
* Enables and disables block-level tags and formatting features.
*
* When disabled, block-level tags aren't rendered. This allows PHP-Textile
* to operate on a single line of text, rather than blocks of text and does
* not wrap the output in paragraph tags.
*
* bc. $parser = new \Netcarving\Textile\Parser();
* echo $parser
* ->setBlockTags(false)
* ->parse('h1. Hello *strong* world!');
*
* The above generates:
*
* bc. h1. Hello <strong>strong</strong> world!
*
* @param bool $enabled TRUE to enable, FALSE to disable
* @return Parser This instance
* @since 3.6.0
* @see Parser::isBlockTagEnabled()
* @api
*/
public function setBlockTags($enabled)
{
$this->blockTagsEnabled = (bool) $enabled;
return $this;
}
/**
* Whether block-level tags are enabled and parsed.
*
* bc. $parser = new \Netcarving\Textile\Parser();
* if ($parser->isBlockTagAllowed() === true) {
* echo 'Block tags are enabled.';
* }
*
* @return bool TRUE if enabled, FALSE otherwise
* @since 3.6.0
* @see Parser::setBlockTags()
* @api
*/
public function isBlockTagEnabled()
{
return (bool) $this->blockTagsEnabled;
}
/**
* Enables and disables line-wrapping.
*
* If enabled, line-breaks are replaced by target document's break tag. If
* disabled, input document's line-breaks are ignored. This setting can be
* used if the the input document's lines are pre-wrapped. For instance,
* in case the input is from CLI content, or source code documentation.
*
* bc. $parser = new \Netcarving\Textile\Parser();
* echo $parser
* ->setLineWrap(false)
* ->parse("Hello\nworld!");
*
* The above generates:
*
* bc. <p>Hello world!</p>
*
* @param bool $enabled TRUE to enable, FALSE to disable
* @return Parser This instance
* @since 3.6.0
* @see Parser::isLineWrapEnabled()
* @api
*/
public function setLineWrap($enabled)
{
$this->lineWrapEnabled = (bool) $enabled;
return $this;
}
/**
* Whether line-wrapping is enabled.
*
* bc. $parser = new \Netcarving\Textile\Parser();
* if ($parser->isLineWrapEnabled() === true) {
* echo 'Line-wrapping is enabled.';
* }
*
* @return bool TRUE if enabled, FALSE otherwise
* @see Parser::setLineWrap()
* @since 3.6.0
* @api
*/
public function isLineWrapEnabled()
{
return (bool) $this->lineWrapEnabled;
}
/**
* Sets a substitution symbol.
*
* This method lets you to redefine a substitution symbol. The following
* sets the 'half' glyph:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setSymbol('half', '1&#8260;2')
* ->parse('Hello [1/2] World!');
*
* Generates:
*
* bc. <p>Hello 1&#⁄2 World!</p>
*
* Symbol can be set to FALSE to disable it:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* $parser->setSymbol('dimension', false);
*
* See Parser::getSymbol() to find out all available symbols.
*
* @param string $name Name of the symbol to assign a new value to
* @param string|bool $value New value for the symbol, or FALSE to disable
* @return Parser This instance
* @see Parser::getSymbol()
* @api
*/
public function setSymbol($name, $value)
{
if ($value !== false) {
$value = (string) $value;
}
$this->symbols[(string) $name] = $value;
$this->rebuild_glyphs = true;
return $this;
}
/**
* Gets a symbol definitions.
*
* This method gets a symbol definition by name, or the full symbol table
* as an array.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->getSymbol('dimension');
*
* To get all available symbol definitions:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* print_r($parser->getSymbol());
*
* @param string|null $name The name of the symbol, or NULL if requesting the symbol table
* @return array|string The symbol table or the requested symbol
* @throws \InvalidArgumentException
* @see Parser::setSymbol()
* @api
*/
public function getSymbol($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 relative image prefix.
*
* The given string is used to prefix relative image paths, usually an
* absolute HTTP address pointing a the site's image, or upload, directory.
* PHP-Textile to convert relative paths to absolute, or prefixed paths.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* $parser->setImagePrefix('https://static.example.com/images/');
*
* @param string $prefix The prefix
* @return Parser This instance
* @since 3.7.0
* @see Parser::getImagePrefix()
* @api
*/
public function setImagePrefix($prefix)
{
$this->relImagePrefix = (string) $prefix;
return $this;
}
/**
* Gets base relative image prefix.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->getImagePrefix();
*
* @return string The prefix
* @since 3.7.0
* @see Parser::setImagePrefix()
* @api
*/
public function getImagePrefix()
{
return (string) $this->relImagePrefix;
}
/**
* Sets base relative link prefix.
*
* The given string is used to prefix relative link paths. This allows
* PHP-Textile convert relative paths to absolute, or prefixed, links.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* $parser->setLinkPrefix('https://example.com/');
*
* @param string $prefix The prefix
* @return Parser This instance
* @since 3.7.0
* @see Parser::getLinkPrefix()
* @api
*/
public function setLinkPrefix($prefix)
{
$this->relLinkPrefix = (string) $prefix;
return $this;
}
/**
* Gets base relative link prefix.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->getLinkPrefix();
*
* @return string The prefix
* @since 3.7.0
* @see Parser::setLinkPrefix()
* @api
*/
public function getLinkPrefix()
{
return (string) $this->relLinkPrefix;
}
/**
* Sets base relative image and link directory path.
*
* This is used when Textile is supplied with a relative image or link path.
* Allows client systems to have PHP-Textile convert relative paths to
* absolute or prefixed paths. This method is used to set that base path,
* usually an absolute HTTP address pointing to a directory. Note that
* despite its name it applies to both links and images.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* $parser->setRelativeImagePrefix('https://example.com/');
*
* @param string $prefix The string to prefix all relative image paths with
* @return Parser This instance
* @deprecated in 3.7.0
* @see Parser::setImagePrefix
* @see Parser::setLinkPrefix
* @api
*/
public function setRelativeImagePrefix($prefix = '')
{
trigger_error(
'Parser::setRelativeImagePrefix() is deprecated.'.
'Use Parser::setImagePrefix() and Parser::setLinkPrefix() instead.',
E_USER_DEPRECATED
);
$this->relativeImagePrefix = $prefix;
return $this;
}
/**
* Enables dimensionless images.
*
* If enabled, image width and height attributes will not be included in
* rendered image tags. Normally, PHP-Textile will add width and height
* to images linked with a local relative path, as long as the image file
* can be accessed.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setDimensionlessImages(true)
* ->parse('!image.jpg!');
*
* @param bool $dimensionless TRUE to disable image dimensions, FALSE to enable
* @return Parser This instance
* @see Parser::getDimensionlessImages()
* @api
*/
public function setDimensionlessImages($dimensionless = true)
{
$this->dimensionless_images = (bool) $dimensionless;
return $this;
}
/**
* Whether dimensionless images are enabled.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* if ($parser->getDimensionlessImages() === true) {
* echo 'Images do not get dimensions.';
* }
*
* @return bool TRUE if images will not get dimensions, FALSE otherwise
* @see Parser::setDimensionlessImages()
* @api
*/
public function getDimensionlessImages()
{
return (bool) $this->dimensionless_images;
}
/**
* Gets PHP-Textile version number.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->getVersion();
*
* @return string Version number
* @api
*/
public function getVersion()
{
return $this->ver;
}
/**
* Encodes the given text.
*
* bc. $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)
{
return (string)preg_replace('/&(?!(?:[a-z][a-z\d]*|#(?:\d+|x[a-f\d]+));)/i', '&amp;', $text);
}
/**
* Parses the given Textile input according to the previously set options.
*
* The parser's features can be changed by using the various public setter
* methods this class has. The most basic use case is:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->parse('h1. Hello World!');
*
* The above parses trusted input in full-feature mode, generating:
*
* bc. <h1>Hello World!</h1>
*
* Additinally the parser can be run in safe, restricted mode using the
* Parser::setRestricted() method.
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setRestricted(true)
* ->parse('h1. Hello World!');
*
* This enables restricted mode and allows safe parsing of untrusted input.
* PHP-Textile will disable unsafe attributes, links and escapes any raw
* HTML input. This option should be enabled when parsing untrusted user
* input.
*
* If restricted mode is disabled, the parser allows users to mix raw HTML
* and Textile.
*
* @param string $text The Textile input to parse
* @return string Parsed Textile input
* @since 3.6.0
* @api
*/
public function parse($text)
{
$this->prepare();
$text = (string) $text;
if ($this->isRestrictedModeEnabled()) {
// Escape any raw HTML.
$text = $this->encodeHTML($text, false);
}
$text = $this->cleanWhiteSpace($text);
$text = $this->cleanUniqueTokens($text);
if ($this->isBlockTagEnabled()) {
if ($this->isLiteModeEnabled()) {
$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);
}
} else {
// Treat quoted quote as a special glyph.
$text = $this->glyphQuotedQuote($text);
// Inline markup (em, strong, sup, sub, del etc).
$text = $this->spans($text);
// Glyph level substitutions (mainly typographic -- " & ' => curly quotes, -- => em-dash etc.
$text = $this->glyphs($text);
}
$text = $this->retrieve($text);
$text = $this->replaceGlyphs($text);
$text = $this->retrieveTags($text);
$text = $this->retrieveURLs($text);
$text = str_replace("<br />", "<br />\n", $text);
return $text;
}
/**
* Parses the given Textile input in un-restricted mode.
*
* This method is deprecated, use Parser::parse() method instead.
* This method is equilavent of:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser->parse('h1. Hello World!');
*
* Additinal arguments can be passed with setter methods:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setLite(true)
* ->setImages(true)
* ->setLinkRelationShip('nofollow')
* ->parse('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 $text
* @see Parser::parse()
* @deprecated in 3.6.0
* @api
*/
public function textileThis($text, $lite = false, $encode = false, $noimage = false, $strict = false, $rel = '')
{
if ($encode) {
trigger_error(
'$encode argument is deprecated. Use Parser::textileEncode() instead.',
E_USER_DEPRECATED
);
return $this->textileEncode($text);
}
trigger_error(
'Parser::textileThis() is deprecated. Use Parser::parse() instead.',
E_USER_DEPRECATED
);
return $this
->setRestricted(false)
->setLite($lite)
->setBlockTags(true)
->setImages(!$noimage)
->setLinkRelationShip($rel)
->parse($text);
}
/**
* Parses the given Textile input in restricted mode.
*
* This method is deprecated, use Parser::parse() method with
* Parser::setRestricted() and Parser::setLite() enabled, and
* Parser::setImages() disabled.
*
* This method's defaults are identical to:
*
* bc. $parser = new \Netcarver\Textile\Parser();
* echo $parser
* ->setRestricted(true)
* ->setLite(true)
* ->setImages(false)
* ->setLinkRelationShip('nofollow')
* ->parse('h1. Hello World!');
*
* As in the above, restricted mode should be used when parsing any
* untrusted user input, including comments or forum posts.
*
* @param string $text The Textile input to parse
* @param bool $lite Controls lite mode, allowing extra formatting
* @param bool $noimage Allow images
* @param string $rel Relationship attribute applied to generated links
* @return string Parsed input
* @see Parser::setRestricted()
* @see Parser::setLite()
* @see Parser::setImages()
* @see Parser::setLinkRelationShip()
* @see Parser::parse()
* @deprecated in 3.6.0
* @api
*/
public function textileRestricted($text, $lite = true, $noimage = true, $rel = 'nofollow')
{
trigger_error(
'Parser::textileRestricted() is deprecated. Use Parser::parse() with Parser::setRestricted() instead.',
E_USER_DEPRECATED
);
return $this
->setRestricted(true)
->setLite($lite)
->setBlockTags(true)
->setImages(!$noimage)
->setLinkRelationShip($rel)
->parse($text);
}
/**
* Parses Textile syntax.
*
* This method performs common parse actions.
*
* @param string $text The input to parse
* @param bool $lite Enables lite mode
* @return string Parsed input
* @deprecated in 3.6.0
*/
protected function textileCommon($text, $lite)
{
trigger_error('Parser::textileCommon() is deprecated.', E_USER_DEPRECATED);
return $this->setLite($lite)->parse($text);
}
/**
* Prepares the glyph patterns from the symbol table.
*
* @see Parser::setSymbol()
* @see Parser::getSymbol()
*/
protected function prepGlyphs()
{
if ($this->rebuild_glyphs === false) {
return;
}
$pnc = '[[:punct:]]';
$cur = '';
if ($this->regex_snippets['cur']) {
$cur = '(?:['.$this->regex_snippets['cur'].']'.$this->regex_snippets['space'].'*)?';
}
$this->glyph_search = array();
$this->glyph_replace = array();
// Dimension sign
if ($this->symbols['dimension'] !== false) {
$this->glyph_search[] = '/(?<=\b|x)([0-9]++[\])]?[\'"]? ?)[x]( ?[\[(]?)(?=[+-]?'.$cur.'[0-9]*\.?[0-9]++)/i'.
$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['dimension'].'$2';
}
// Apostrophe
if ($this->symbols['apostrophe'] !== false) {
$this->glyph_search[] = '/('.$this->regex_snippets['wrd'].'|\))\''.
'('.$this->regex_snippets['wrd'].')/'.$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['apostrophe'].'$2';
// Back in '88/the '90s but not in his '90s', '1', '1.' '10m' or '5.png'
$this->glyph_search[] = '/('.$this->regex_snippets['space'].')\''.
'(\d+'.$this->regex_snippets['wrd'].'?)\b(?![.]?['.$this->regex_snippets['wrd'].']*?\')/'.
$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['apostrophe'].'$2';
}
// Single open following open bracket
if ($this->symbols['quote_single_open'] !== false) {
$this->glyph_search[] = "/([([{])'(?=\S)/".$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['quote_single_open'];
}
// Single closing
if ($this->symbols['quote_single_close'] !== false) {
$this->glyph_search[] = '/(\S)\'(?='.$this->regex_snippets['space'].'|'.$pnc.'|<|$)/'.
$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['quote_single_close'];
}
// Default single opening
if ($this->symbols['quote_single_open'] !== false) {
$this->glyph_search[] = "/'/";
$this->glyph_replace[] = $this->symbols['quote_single_open'];
}
// Double open following an open bracket. Allows things like Hello ["(Mum) & dad"]
if ($this->symbols['quote_double_open'] !== false) {
$this->glyph_search[] = '/([([{])"(?=\S)/'.$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['quote_double_open'];
}
// Double closing
if ($this->symbols['quote_double_close'] !== false) {
$this->glyph_search[] = '/(\S)"(?='.$this->regex_snippets['space'].'|'.$pnc.'|<|$)/'.
$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['quote_double_close'];
}
// Default double opening
if ($this->symbols['quote_double_open'] !== false) {
$this->glyph_search[] = '/"/';
$this->glyph_replace[] = $this->symbols['quote_double_open'];
}
if ($this->symbols['acronym'] === null) {
if ($this->getDocumentType() === 'html5') {
$acronym = '<abbr title="{title}">{content}</abbr>';
} else {
$acronym = '<acronym title="{title}">{content}</acronym>';
}
} else {
$acronym = $this->symbols['acronym'];
}
// 3+ uppercase acronym
if ($acronym !== false) {
$this->glyph_search[] = '/\b(['.$this->regex_snippets['abr'].']['.
$this->regex_snippets['acr'].']{2,})\b(?:[(]([^)]*)[)])/'.$this->regex_snippets['mod'];
$this->glyph_replace[] = $this->replaceMarkers($acronym, array(
'title' => '$2',
'content' => '$1',
));
}
// 3+ uppercase
if ($this->symbols['caps'] !== false) {
$this->glyph_search[] = '/('.$this->regex_snippets['space'].'|^|[>(;-])'.
'(['.$this->regex_snippets['abr'].']{3,})'.
'(['.$this->regex_snippets['nab'].']*)(?='.
$this->regex_snippets['space'].'|'.$pnc.'|<|$)'.
'(?=[^">]*?(<|$))/'.$this->regex_snippets['mod'];
$this->glyph_replace[] = $this->replaceMarkers('$1'.$this->symbols['caps'].'$3', array(
'content' => $this->uid.':glyph:$2',
));
}
// Ellipsis
if ($this->symbols['ellipsis'] !== false) {
$this->glyph_search[] = '/([^.]?)\.{3}/';
$this->glyph_replace[] = '$1'.$this->symbols['ellipsis'];
}
// em dash
if ($this->symbols['emdash'] !== false) {
$this->glyph_search[] = '/--/';
$this->glyph_replace[] = $this->symbols['emdash'];
}
// en dash
if ($this->symbols['endash'] !== false) {
$this->glyph_search[] = '/ - /';
$this->glyph_replace[] = ' '.$this->symbols['endash'].' ';
}
// Trademark
if ($this->symbols['trademark'] !== false) {
$this->glyph_search[] = '/(\b ?|'.$this->regex_snippets['space'].'|^)[([]TM[])]/i'.
$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['trademark'];
}
// Registered
if ($this->symbols['registered'] !== false) {
$this->glyph_search[] = '/(\b ?|'.$this->regex_snippets['space'].'|^)[([]R[])]/i'.
$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['registered'];
}
// Copyright
if ($this->symbols['copyright'] !== false) {
$this->glyph_search[] = '/(\b ?|'.$this->regex_snippets['space'].'|^)[([]C[])]/i'.
$this->regex_snippets['mod'];
$this->glyph_replace[] = '$1'.$this->symbols['copyright'];
}
// 1/4
if ($this->symbols['quarter'] !== false) {
$this->glyph_search[] = '/[([]1\/4[])]/';
$this->glyph_replace[] = $this->symbols['quarter'];
}
// 1/2
if ($this->symbols['half'] !== false) {
$this->glyph_search[] = '/[([]1\/2[])]/';
$this->glyph_replace[] = $this->symbols['half'];
}
// 3/4
if ($this->symbols['threequarters'] !== false) {
$this->glyph_search[] = '/[([]3\/4[])]/';
$this->glyph_replace[] = $this->symbols['threequarters'];
}
// Degrees -- that's a small 'oh'
if ($this->symbols['degrees'] !== false) {
$this->glyph_search[] = '/[([]o[])]/';
$this->glyph_replace[] = $this->symbols['degrees'];
}
// Plus minus
if ($this->symbols['plusminus'] !== false) {
$this->glyph_search[] = '/[([]\+\/-[])]/';
$this->glyph_replace[] = $this->symbols['plusminus'];
}
// No need to rebuild next run unless a symbol is redefined
$this->rebuild_glyphs = false;
}
/**
* Gets the maximum allowed link index.
*
* @return int Maximum link index
* @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|null $lite Controls lite mode
* @param bool|null $noimage Disallow images
* @param string|null $rel A relationship attribute applied to links
*/
protected function prepare($lite = null, $noimage = null, $rel = null)
{
if ($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();
$this->note_index = 1;
if ($lite !== null) {
trigger_error(
'$lite argument is deprecated. Use Parser::setLite() instead.',
E_USER_DEPRECATED
);
$this->setLite($lite);
}
if ($noimage !== null) {
trigger_error(
'$noimage argument is deprecated. Use Parser::setImages() instead.',
E_USER_DEPRECATED
);
$this->setImages(!$noimage);
}
if ($rel !== null) {
trigger_error(
'$rel argument is deprecated. Use Parser::setRelative() instead.',
E_USER_DEPRECATED
);
$this->setLinkRelationShip($rel);
}
if ($this->patterns === null) {
$block = implode('|', $this->blockContent);
$divider = implode('|', $this->dividerContent);
$phrasing = implode('|', $this->phrasingContent);
$this->patterns = array(
'block' => '/^(?:'.$block.')$/i',
'contained' => '/^<\/?(?P<open>[^\s<>\/]+)(?:\s.*|\/?>.*|)>$/si',
'divider' => '/^(?:<\/?('.$divider.')(?:\s[^<>]*?|\/?)>(?:<\/\1\s*?>)?)+$/smi',
'phrasing' => '/^(?:'.$phrasing.')$/i',
'wrapped' => '/^<\/?(?P<open>[^\s<>\/]+)[^<>]*?>(?:.*<\/\1\s*?>)?$/si',
'unwrappable' => '/<\/?(?:'.$block.')(?:\s[^<>]*?|\/?)>/smi',
);
}
$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 = array();
$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], '', $matched);
if ($element === 'code' && preg_match("/\[([a-zA-Z0-9_-]+)\]/U", $lng[0], $lng1)) {
$lang = $lng1[1];
} elseif (preg_match("/\[([a-zA-Z]{2}(?:[\-\_][a-zA-Z]{2})?)\]/U", $lng[0], $lng2)) {
$lang = $lng2[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->isRestrictedModeEnabled()) {
$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 (!empty($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 $p) {
if ($p) {
$so .= $p.';';
}
}
$o['style'] = trim(str_replace(array("\n", ';;'), array('', ';'), $so));
}
if ($width) {
$o['width'] = $this->cleanAttribs($width);
}
ksort($o);
return $o;
}
/**
* Checks whether the text block should be wrapped in a paragraph.
*
* @param string $text The input string
* @return bool TRUE if the text can be wrapped, FALSE otherwise
*/
protected function hasRawText($text)
{
if (preg_match($this->patterns['unwrappable'], $text)) {
return false;
}
if (preg_match($this->patterns['divider'], $text)) {
return false;
}
if (preg_match($this->patterns['wrapped'], $text, $m)) {
if (preg_match($this->patterns['phrasing'], $m['open'])) {
return true;
}
return false;
}
return true;
}
/**
* Parses textile table structures into HTML.
*
* @param string $text The textile input
* @return string The parsed text
*/
protected function tables($text)
{
$text = $text . "\n\n";
return (string)preg_replace_callback(
"/^(?:table(?P<tatts>_?{$this->s}{$this->a}{$this->cls})\.".
"(?P<summary>.*)?\n)?^(?P<rows>{$this->a}{$this->cls}\.? ?\|.*\|){$this->regex_snippets['space']}*\n\n/smU",
array($this, "fTable"),
$text
);
}
/**
* Constructs a HTML table from a textile table structure.
*
* This method is used by Parser::tables() to process
* found table structures.
*
* @param array $matches
* @return string HTML table
* @see Parser::tables()
*/
protected function fTable($matches)
{
$tatts = $this->parseAttribs($matches['tatts'], 'table');
$space = $this->regex_snippets['space'];
$cap = '';
$colgrp = '';
$last_rgrp = '';
$c_row = 1;
$sum = '';
$rows = array();
$summary = trim($matches['summary']);
if ($summary !== '') {
$sum = ' summary="'.htmlspecialchars($summary, ENT_QUOTES, 'UTF-8').'"';
}
foreach (preg_split("/\|{$space}*?$/m", $matches['rows'], -1, PREG_SPLIT_NO_EMPTY) as $row) {
$row = ltrim($row);
// Caption -- can only occur on row 1, otherwise treat '|=. foo |...'
// as a normal center-aligned cell.
if (($c_row <= 1) && preg_match(
"/^\|\=(?P<capts>$this->s$this->a$this->cls)\. (?P<cap>[^\n]*)(?P<row>.*)/s",
ltrim($row),
$cmtch
)) {
$capts = $this->parseAttribs($cmtch['capts']);
$cap = "\t<caption".$capts.">".trim($cmtch['cap'])."</caption>\n";
$row = ltrim($cmtch['row']);
if (!$row) {
continue;
}
}
$c_row += 1;
// Colgroup
if (preg_match("/^\|:(?P<cols>$this->s$this->a$this->cls\. .*)/m", ltrim($row), $gmtch)) {
// Is this colgroup def missing a closing pipe? If so, there
// will be a newline in the middle of $row somewhere.
$nl = strpos($row, "\n");
$idx = 0;
foreach (explode('|', str_replace('.', '', $gmtch['cols'])) as $col) {
$gatts = $this->parseAttribs(trim($col), 'col');
$colgrp .= "\t<col".(($idx==0) ? "group".$gatts.">" : $gatts." />")."\n";
$idx++;
}
$colgrp .= "\t</colgroup>\n";
if ($nl === false) {
continue;
} else {
// Recover from our missing pipe and process the rest of the line.
$row = ltrim(substr($row, $nl));
}
}
// Row group
$rgrpatts = $rgrp = '';
if (preg_match(
"/(:?^\|(?P<part>$this->vlgn)(?P<rgrpatts>$this->s$this->a$this->cls)\.{$space}*$\n)?^(?P<row>.*)/sm",
ltrim($row),
$grpmatch
)) {
if (isset($grpmatch['part'])) {
if ($grpmatch['part'] === '^') {
$rgrp = 'head';
} elseif ($grpmatch['part'] === '~') {
$rgrp = 'foot';
} elseif ($grpmatch['part'] === '-') {
$rgrp = 'body';
}
}
if (isset($grpmatch['part'])) {
$rgrpatts = $this->parseAttribs($grpmatch['rgrpatts']);
}
if (isset($grpmatch['row'])) {
$row = $grpmatch['row'];
}
}
if (preg_match("/^(?P<ratts>$this->a$this->cls\. )(?P<row>.*)/m", ltrim($row), $rmtch)) {
$ratts = $this->parseAttribs($rmtch['ratts'], 'tr');
$row = $rmtch['row'];
} else {
$ratts = '';
}
$cells = array();
$cellctr = 0;
foreach (explode("|", $row) as $cell) {
$ctyp = "d";
if (preg_match("/^_(?=[{$this->regex_snippets['space']}[:punct:]])/", $cell)) {
$ctyp = "h";
}
if (preg_match("/^(?P<catts>_?$this->s$this->a$this->cls\. )(?P<cell>.*)/s", $cell, $cmtch)) {
$catts = $this->parseAttribs($cmtch['catts'], 'td');
$cell = $cmtch['cell'];
} else {
$catts = '';
}
if (!$this->isLiteModeEnabled()) {
$a = array();
if (preg_match('/(?<space>'.$this->regex_snippets['space'].'*)(?P<cell>.*)/s', $cell, $a)) {
$cell = $this->redclothLists($a['cell']);
$cell = $this->textileLists($cell);
$cell = $a['space'] . $cell;
}
}
if ($cellctr > 0) {
// Ignore first 'cell': it precedes the opening pipe
$cells[] = $this->doTagBr("t$ctyp", "\t\t\t<t$ctyp$catts>$cell</t$ctyp>");
}
$cellctr++;
}
$grp = '';
if ($rgrp && $last_rgrp) {
$grp .= "\t</t".$last_rgrp.">\n";
}
if ($rgrp) {
$grp .= "\t<t".$rgrp.$rgrpatts.">\n";
}
$last_rgrp = ($rgrp) ? $rgrp : $last_rgrp;
$rows[] = $grp."\t\t<tr$ratts>\n" . join("\n", $cells) . ($cells ? "\n" : "") . "\t\t</tr>";
unset($cells, $catts);
}
$rows = join("\n", $rows) . "\n";
$close = '';
if ($last_rgrp) {
$close = "\t</t".$last_rgrp.">\n";
}
return "<table{$tatts}{$sum}>\n".$cap.$colgrp.$rows.$close."</table>\n\n";
}
/**
* Parses RedCloth-style definition lists into HTML.
*
* @param string $text The textile input
* @return string The parsed text
*/
protected function redclothLists($text)
{
return (string)preg_replace_callback(
"/^([-]+$this->cls[ .].*:=.*)$(?![^-])/smU",
array($this, "fRedclothList"),
$text
);
}
/**
* Constructs a HTML definition list from a RedCloth-style definition structure.
*
* This method is used by Parser::redclothLists() to process
* found definition list structures.
*
* @param array $m
* @return string HTML definition list
* @see Parser::redclothLists()
*/
protected function fRedclothList($m)
{
$in = $m[0];
$out = array();
$text = preg_split('/\n(?=[-])/m', $in);
foreach ($text as $line) {
$m = array();
if (preg_match("/^[-]+(?P<atts>$this->cls)\.? (?P<content>.*)$/s", $line, $m)) {
$content = trim($m['content']);
$atts = $this->parseAttribs($m['atts']);
if (!preg_match(
"/^(.*?){$this->regex_snippets['space']}*:=(.*?)".
"{$this->regex_snippets['space']}*(=:|:=)?".
"{$this->regex_snippets['space']}*$/s",
$content,
$xm
)) {
$xm = array( $content, $content, '' );
}
list(, $term, $def,) = $xm;
$term = trim($term);
$def = trim($def, ' ');
if (!$out) {
if ($def === '') {
$out[] = "<dl$atts>";
} else {
$out[] = '<dl>';
}
}
if ($term !== '') {
$pos = strpos($def, "\n");
$def = trim($def);
if ($this->isLineWrapEnabled()) {
$def = str_replace("\n", "<br />", $def);
}
if ($pos === 0) {
$def = '<p>' . $def . '</p>';
}
if ($this->isLineWrapEnabled()) {
$term = str_replace("\n", "<br />", $term);
}
$term = $this->graf($term);
$def = $this->graf($def);
$out[] = "\t<dt$atts>$term</dt>";
if ($def !== '') {
$out[] = "\t<dd>$def</dd>";
}
}
}
}
$out[] = '</dl>';
return implode("\n", $out);
}
/**
* Parses Textile list structures into HTML.
*
* Searches for ordered, un-ordered and definition lists in the
* textile input and generates HTML lists for them.
*
* @param string $text The input
* @return string The parsed text
*/
protected function textileLists($text)
{
return (string)preg_replace_callback(
"/^((?:[*;:]+|[*;:#]*#(?:_|\d+)?)$this->cls[ .].*)$(?![^#*;:])/smU",
array($this, "fTextileList"),
$text
);
}
/**
* Constructs a HTML list from a Textile list structure.
*
* This method is used by Parser::textileLists() to process
* found list structures.
*
* @param array $m
* @return string HTML list
* @see Parser::textileLists()
*/
protected function fTextileList($m)
{
$text = $m[0];
$lines = preg_split('/\n(?=[*#;:])/m', $m[0]);
$list = array();
$prev = false;
$out = array();
$lists = array();
if ($lines === false) {
return '';
}
foreach ($lines as $line) {
$match = preg_match(
"/^(?P<tl>[#*;:]+)(?P<st>_|\d+)?(?P<atts>$this->cls)[ .](?P<content>.*)$/s",
$line,
$m
);
if ($match) {
$list[] = array_merge($m, array(
'level' => strlen($m['tl']),
));
} else {
$list[count($list) - 1]['content'] .= "\n" . $line;
}
}
if (!$list || $list[0]['level'] > 1) {
return $text;
}
foreach ($list as $index => $m) {
$start = '';
$content = trim($m['content']);
$ltype = $this->liType($m['tl']);
if (isset($list[$index + 1])) {
$next = $list[$index + 1];
} else {
$next = false;
}
if (strpos($m['tl'], ';') !== false) {
$litem = 'dt';
} elseif (strpos($m['tl'], ':') !== false) {
$litem = 'dd';
} else {
$litem = 'li';
}
$showitem = ($content !== '');
if ('o' === $ltype) {
if (!isset($this->olstarts[$m['tl']])) {
$this->olstarts[$m['tl']] = 1;
}
if (!$prev || $m['level'] > $prev['level']) {
if ($m['st'] === '') {
$this->olstarts[$m['tl']] = 1;
} elseif ($m['st'] !== '_') {
$this->olstarts[$m['tl']] = (int) $m['st'];
}
}
if ($m['level'] > $prev['level'] && $m['st'] !== '') {
$start = ' start="' . $this->olstarts[$m['tl']] . '"';
}
if ($showitem) {
$this->olstarts[$m['tl']] += 1;
}
}
if (strpos($prev['tl'], ';') !== false && strpos($m['tl'], ':') !== false) {
$lists[$m['tl']] = 2;
}
$tabs = str_repeat("\t", $m['level'] - 1);
$atts = $this->parseAttribs($m['atts']);
if (!isset($lists[$m['tl']])) {
$lists[$m['tl']] = 1;
$line = $tabs.'<'.$ltype.'l'.$atts.$start.'>';
if ($showitem) {
$line .= "\n$tabs\t<$litem>$content";
}
} elseif ($showitem) {
$line = "$tabs\t<$litem$atts>$content";
}
if ((!$next || $next['level'] <= $m['level']) && $showitem) {
$line .= "</$litem>";
}
foreach (array_reverse($lists) as $k => $v) {
$indent = strlen($k);
if (!$next || $indent > $next['level']) {
if ($v !== 2) {
$line .= "\n$tabs</" . $this->liType($k) . "l>";
}
if ($v !== 2 && $indent > 1) {
$line .= "</".$litem.">";
}
unset($lists[$k]);
}
}
$prev = $m;
$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<type>[#*]+)/', $in, $m)) {
$type = ('#' === substr($m['type'], -1)) ? 'o' : 'u';
}
return $type;
}
/**
* Adds br tags within the specified container tag.
*
* @param string $tag The tag
* @param string $in The input
* @return string
*/
protected function doTagBr($tag, $in)
{
return (string)preg_replace_callback(
'@<(?P<tag>'.preg_quote($tag).')(?P<atts>[^>]*?)>(?P<content>.*)(?P<closetag></\1>)@s',
array($this, 'fBr'),
$in
);
}
/**
* Adds br tags to paragraphs and headings.
*
* @param string $in The input
* @return string
*/
protected function doPBr($in)
{
return (string)preg_replace_callback(
'@<(?P<tag>p|h[1-6])(?P<atts>[^>]*?)>(?P<content>.*)(?P<closetag></\1>)@s',
array($this, 'fPBr'),
$in
);
}
/**
* Less restrictive version of fBr method.
*
* Used only in paragraphs and headings where the next row may
* start with a smiley or perhaps something like '#8 bolt...'
* or '*** stars...'.
*
* @param array $m The input
* @return string
*/
protected function fPBr($m)
{
if ($this->isLineWrapEnabled()) {
// Replaces <br/>\n instances that are not followed by white-space,
// or at end, with single LF.
$m['content'] = preg_replace(
"~<br[ ]*/?>{$this->regex_snippets['space']}*\n(?![{$this->regex_snippets['space']}|])~i",
"\n",
$m['content']
);
}
// Replaces those LFs that aren't followed by white-space, or at end, with <br /> or a space.
$m['content'] = preg_replace(
"/\n(?![\s|])/",
$this->isLineWrapEnabled() ? '<br />' : ' ',
$m['content']
);
return '<'.$m['tag'].$m['atts'].'>'.$m['content'].$m['closetag'];
}
/**
* Formats line breaks.
*
* @param array $m The input
* @return string
*/
protected function fBr($m)
{
$content = preg_replace(
"@(.+)(?<!<br>|<br />|</li>|</dd>|</dt>)\n(?![\s|])@",
$this->isLineWrapEnabled() ? '$1<br />' : '$1 ',
$m['content']
);
return '<'.$m['tag'].$m['atts'].'>'.$content.$m['closetag'];
}
/**
* Splits the given input into blocks.
*
* Blocks are separated by double line-break b