diff options
author | zotlabs <mike@macgirvin.com> | 2017-03-05 18:22:37 +1100 |
---|---|---|
committer | GitHub <noreply@github.com> | 2017-03-05 18:22:37 +1100 |
commit | 215bd07f0b23c03f8af0be7bee1d393f1821087d (patch) | |
tree | 26809ee07eeee05240878bd08cfb4fdcf4bb450a | |
parent | 8e1716065ee01959fc799fa14ba627392a876afa (diff) | |
parent | 6c79e0c077971029343b2dff30017571ea118438 (diff) | |
download | volse-hubzilla-215bd07f0b23c03f8af0be7bee1d393f1821087d.tar.gz volse-hubzilla-215bd07f0b23c03f8af0be7bee1d393f1821087d.tar.bz2 volse-hubzilla-215bd07f0b23c03f8af0be7bee1d393f1821087d.zip |
Merge pull request #688 from dawnbreak/markdown
:arrow_up: :hammer: Upgrade Markdownify library.
-rw-r--r-- | composer.json | 3 | ||||
-rw-r--r-- | composer.lock | 58 | ||||
-rw-r--r-- | include/markdown.php | 10 | ||||
-rw-r--r-- | library/markdownify/TODO | 29 | ||||
-rw-r--r-- | library/markdownify/example.php | 51 | ||||
-rw-r--r-- | library/markdownify/markdownify.php | 1197 | ||||
-rwxr-xr-x | library/markdownify/markdownify_cli.php | 33 | ||||
-rw-r--r-- | library/markdownify/markdownify_extra.php | 489 | ||||
-rw-r--r-- | library/markdownify/parsehtml/parsehtml.php | 618 | ||||
-rw-r--r-- | vendor/composer/autoload_classmap.php | 9 | ||||
-rw-r--r-- | vendor/composer/autoload_psr4.php | 2 | ||||
-rw-r--r-- | vendor/composer/autoload_static.php | 25 | ||||
-rw-r--r-- | vendor/composer/installed.json | 58 | ||||
-rw-r--r-- | vendor/pixel418/markdownify/CHANGELOG.md | 76 | ||||
-rw-r--r-- | vendor/pixel418/markdownify/LICENSE (renamed from library/markdownify/LICENSE_LGPL.txt) | 0 | ||||
-rw-r--r-- | vendor/pixel418/markdownify/README.md | 63 | ||||
-rw-r--r-- | vendor/pixel418/markdownify/composer.json | 38 | ||||
-rw-r--r-- | vendor/pixel418/markdownify/src/Converter.php | 1400 | ||||
-rw-r--r-- | vendor/pixel418/markdownify/src/ConverterExtra.php | 573 | ||||
-rw-r--r-- | vendor/pixel418/markdownify/src/Parser.php | 564 |
20 files changed, 2869 insertions, 2427 deletions
diff --git a/composer.json b/composer.json index d977987eb..b59c81bba 100644 --- a/composer.json +++ b/composer.json @@ -29,7 +29,8 @@ "ext-xml" : "*", "ext-openssl" : "*", "sabre/dav" : "~3.2", - "michelf/php-markdown" : "^1.7" + "michelf/php-markdown" : "^1.7", + "pixel418/markdownify": "^2.2" }, "require-dev" : { "php" : ">=5.6", diff --git a/composer.lock b/composer.lock index cf348bdae..06fb17410 100644 --- a/composer.lock +++ b/composer.lock @@ -4,7 +4,7 @@ "Read more about it at https://getcomposer.org/doc/01-basic-usage.md#composer-lock-the-lock-file", "This file is @generated automatically" ], - "content-hash": "4b24468c1f8babe7c8804fba8ee602f7", + "content-hash": "c0cafbf9fd702be588f6b392b9742cb6", "packages": [ { "name": "michelf/php-markdown", @@ -58,6 +58,62 @@ "time": "2016-10-29T18:58:20+00:00" }, { + "name": "pixel418/markdownify", + "version": "v2.2.1", + "source": { + "type": "git", + "url": "https://github.com/Elephant418/Markdownify.git", + "reference": "0160677f04c784550dd10fd72fdf3994967db848" + }, + "dist": { + "type": "zip", + "url": "https://api.github.com/repos/Elephant418/Markdownify/zipball/0160677f04c784550dd10fd72fdf3994967db848", + "reference": "0160677f04c784550dd10fd72fdf3994967db848", + "shasum": "" + }, + "require": { + "php": ">=5.3.0" + }, + "require-dev": { + "phpunit/phpunit": "^4.8" + }, + "type": "lib", + "autoload": { + "psr-4": { + "Markdownify\\": "src", + "Test\\Markdownify\\": "test" + } + }, + "notification-url": "https://packagist.org/downloads/", + "license": [ + "LGPL" + ], + "authors": [ + { + "name": "Peter Kruithof", + "email": "pkruithof@gmail.com", + "homepage": "http://pkruithof.tumblr.com/" + }, + { + "name": "Milian Wolff", + "email": "mail@milianw.de", + "homepage": "http://milianw.de" + }, + { + "name": "Thomas Zilliox", + "email": "hello@tzi.fr", + "homepage": "http://tzi.fr" + } + ], + "description": "The HTML to Markdown converter for PHP ", + "homepage": "https://github.com/elephant418/Markdownify", + "keywords": [ + "markdown", + "markdownify" + ], + "time": "2016-09-21T13:01:43+00:00" + }, + { "name": "psr/log", "version": "1.0.2", "source": { diff --git a/include/markdown.php b/include/markdown.php index 43d973808..7e1f1c42c 100644 --- a/include/markdown.php +++ b/include/markdown.php @@ -5,12 +5,12 @@ */ use Michelf\MarkdownExtra; +use Markdownify\Converter; require_once("include/oembed.php"); require_once("include/event.php"); require_once("include/html2bbcode.php"); require_once("include/bbcode.php"); -require_once("library/markdownify/markdownify.php"); function get_bb_tag_pos($s, $name, $occurance = 1) { @@ -367,7 +367,6 @@ function bb2diaspora_itemwallwall(&$item,$uplink = false) { function bb2diaspora_itembody($item, $force_update = false, $have_channel = false, $uplink = false) { - if(! get_iconfig($item,'diaspora','fields')) { $force_update = true; } @@ -454,7 +453,7 @@ function bb2diaspora_itembody($item, $force_update = false, $have_channel = fals return html_entity_decode($body); } -function bb2diaspora($Text,$preserve_nl = false, $fordiaspora = true) { +function bb2diaspora($Text, $preserve_nl = false, $fordiaspora = true) { // Re-enabling the converter again. // The bbcode parser now handles youtube-links (and the other stuff) correctly. @@ -496,11 +495,10 @@ function bb2diaspora($Text,$preserve_nl = false, $fordiaspora = true) { $Text = str_replace(array('<','>','&'),array('&_lt_;','&_gt_;','&_amp_;'),$Text); // Now convert HTML to Markdown - $md = new Markdownify(false, false, false); + $md = new Converter(Converter::LINK_AFTER_CONTENT, false, false); $Text = $md->parseString($Text); - // It also adds backslashes to our attempt at getting around the html entity preservation for some weird reason. $Text = str_replace(array('&\\_lt\\_;','&\\_gt\\_;','&\\_amp\\_;'),array('<','>','&'),$Text); @@ -522,7 +520,7 @@ function bb2diaspora($Text,$preserve_nl = false, $fordiaspora = true) { $Text = trim($Text); - call_hooks('bb2diaspora',$Text); + call_hooks('bb2diaspora', $Text); return $Text; } diff --git a/library/markdownify/TODO b/library/markdownify/TODO deleted file mode 100644 index 06ec8508b..000000000 --- a/library/markdownify/TODO +++ /dev/null @@ -1,29 +0,0 @@ -Markdownify -=========== -* handle non-markdownifiable lists (i.e. `<ul><li id="foobar">asdf</li></ul>`) -* organize methods better (i.e. flushlinebreaks & setlinebreaks close to each other) -* take a look at function names etc. -* is the new (in rev. 93) lastclosedtag property needed? -* word wrapping (some work is done but it's still very buggy) - - -Markdownify Extra -================= - -* handle table alignment with KEEP_HTML=false -* handle tables without headings when KEEP_HTML=false is set -* handle Markdown inside non-markdownable tags - - -Implementation Thoughts -======================= -* non-markdownifiable lists and markdown inside non-markdownable tags as well as the current - table implementation could be rewritten by using a rollback mechanism. - - example: - - <ul><li>asdf</li><li id="foobar">asdf</li></ul> - - we come to `<ul>`, know that this might fail and create a snapshot of our current parser - we keep on parsing and when we reach `<li id="foobar">` we gotta rollback and keep this - list in HTML format. diff --git a/library/markdownify/example.php b/library/markdownify/example.php deleted file mode 100644 index ef86dca83..000000000 --- a/library/markdownify/example.php +++ /dev/null @@ -1,51 +0,0 @@ -<?php - error_reporting(E_ALL); - if (!empty($_POST['input'])) { - include 'markdownify_extra.php'; - if (!isset($_POST['leap'])) { - $leap = MDFY_LINKS_EACH_PARAGRAPH; - } else { - $leap = $_POST['leap']; - } - - if (!isset($_POST['keepHTML'])) { - $keephtml = MDFY_KEEPHTML; - } else { - $keephtml = $_POST['keepHTML']; - } - if (!empty($_POST['extra'])) { - $md = new Markdownify_Extra($leap, MDFY_BODYWIDTH, $keephtml); - } else { - $md = new Markdownify($leap, MDFY_BODYWIDTH, $keephtml); - } - if (ini_get('magic_quotes_gpc')) { - $_POST['input'] = stripslashes($_POST['input']); - } - $output = $md->parseString($_POST['input']); - } else { - $_POST['input'] = ''; - } -?><!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> -<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en-US" lang="en-US"> - <head> - <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/> - <title>HTML to Markdown Converter</title> - </head> - <body> - <?php if (empty($_POST['input'])): ?> - <form action="<?php echo $_SERVER['PHP_SELF']; ?>" method="post"> - <fieldset> - <legend>HTML Input</legend> - <textarea style="width:100%;" cols="85" rows="40" name="input"><?php echo htmlspecialchars($_POST['input'], ENT_NOQUOTES, 'UTF-8'); ?></textarea> - </fieldset> - <label for="extra">Markdownify Extra: <input name="extra" checked="checked" id="extra" type="checkbox" value="1" /></label> - <label for="leap">Links after each block elem: <input name="leap" id="leap" type="checkbox" value="1" /></label> - <label for="keepHTML">keep HTML: <input name="keepHTML" id="keepHTML" type="checkbox" value="1" checked="checked" /></label> - <input type="submit" name="submit" value="submit" /> - </form> - <?php else: ?> - <h1 style="text-align:right;"><a href="<?php echo $_SERVER['PHP_SELF']; ?>">BACK</a></h1> - <pre><?php echo htmlspecialchars($output, ENT_NOQUOTES, 'UTF-8'); ?></pre> - <?php endif; ?> - </body> -</html>
\ No newline at end of file diff --git a/library/markdownify/markdownify.php b/library/markdownify/markdownify.php deleted file mode 100644 index 0d4429a01..000000000 --- a/library/markdownify/markdownify.php +++ /dev/null @@ -1,1197 +0,0 @@ -<?php -/** - * Markdownify converts HTML Markup to [Markdown][1] (by [John Gruber][2]. It - * also supports [Markdown Extra][3] by [Michel Fortin][4] via Markdownify_Extra. - * - * It all started as `html2text.php` - a port of [Aaron Swartz'][5] [`html2text.py`][6] - but - * got a long way since. This is far more than a mere port now! - * Starting with version 2.0.0 this is a complete rewrite and cannot be - * compared to Aaron Swatz' `html2text.py` anylonger. I'm now using a HTML parser - * (see `parsehtml.php` which I also wrote) which makes most of the evil - * RegEx magic go away and additionally it gives a much cleaner class - * structure. Also notably is the fact that I now try to prevent regressions by - * utilizing testcases of Michel Fortin's [MDTest][7]. - * - * [1]: http://daringfireball.com/projects/markdown - * [2]: http://daringfireball.com/ - * [3]: http://www.michelf.com/projects/php-markdown/extra/ - * [4]: http://www.michelf.com/ - * [5]: http://www.aaronsw.com/ - * [6]: http://www.aaronsw.com/2002/html2text/ - * [7]: http://article.gmane.org/gmane.text.markdown.general/2540 - * - * @version 2.0.0 alpha - * @author Milian Wolff (<mail@milianw.de>, <http://milianw.de>) - * @license LGPL, see LICENSE_LGPL.txt and the summary below - * @copyright (C) 2007 Milian Wolff - * - * This library is free software; you can redistribute it and/or - * modify it under the terms of the GNU Lesser General Public - * License as published by the Free Software Foundation; either - * version 2.1 of the License, or (at your option) any later version. - * - * This library is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * Lesser General Public License for more details. - * - * You should have received a copy of the GNU Lesser General Public - * License along with this library; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA - */ - -/** - * HTML Parser, see http://sf.net/projects/parseHTML - */ -require_once dirname(__FILE__).'/parsehtml/parsehtml.php'; - -/** - * default configuration - */ -define('MDFY_LINKS_EACH_PARAGRAPH', false); -define('MDFY_BODYWIDTH', false); -define('MDFY_KEEPHTML', true); - -/** - * HTML to Markdown converter class - */ -class Markdownify { - /** - * html parser object - * - * @var parseHTML - */ - var $parser; - /** - * markdown output - * - * @var string - */ - var $output; - /** - * stack with tags which where not converted to html - * - * @var array<string> - */ - var $notConverted = array(); - /** - * skip conversion to markdown - * - * @var bool - */ - var $skipConversion = false; - /* options */ - /** - * keep html tags which cannot be converted to markdown - * - * @var bool - */ - var $keepHTML = false; - /** - * wrap output, set to 0 to skip wrapping - * - * @var int - */ - var $bodyWidth = 0; - /** - * minimum body width - * - * @var int - */ - var $minBodyWidth = 25; - /** - * display links after each paragraph - * - * @var bool - */ - var $linksAfterEachParagraph = false; - /** - * constructor, set options, setup parser - * - * @param bool $linksAfterEachParagraph wether or not to flush stacked links after each paragraph - * defaults to false - * @param int $bodyWidth wether or not to wrap the output to the given width - * defaults to false - * @param bool $keepHTML wether to keep non markdownable HTML or to discard it - * defaults to true (HTML will be kept) - * @return void - */ - function Markdownify($linksAfterEachParagraph = MDFY_LINKS_EACH_PARAGRAPH, $bodyWidth = MDFY_BODYWIDTH, $keepHTML = MDFY_KEEPHTML) { - $this->linksAfterEachParagraph = $linksAfterEachParagraph; - $this->keepHTML = $keepHTML; - - if ($bodyWidth > $this->minBodyWidth) { - $this->bodyWidth = intval($bodyWidth); - } else { - $this->bodyWidth = false; - } - - $this->parser = new parseHTML; - $this->parser->noTagsInCode = true; - - # we don't have to do this every time - $search = array(); - $replace = array(); - foreach ($this->escapeInText as $s => $r) { - array_push($search, '#(?<!\\\)'.$s.'#U'); - array_push($replace, $r); - } - $this->escapeInText = array( - 'search' => $search, - 'replace' => $replace - ); - } - /** - * parse a HTML string - * - * @param string $html - * @return string markdown formatted - */ - function parseString($html) { - $this->parser->html = $html; - $this->parse(); - return $this->output; - } - /** - * tags with elements which can be handled by markdown - * - * @var array<string> - */ - var $isMarkdownable = array( - 'p' => array(), - 'ul' => array(), - 'ol' => array(), - 'li' => array(), - 'br' => array(), - 'blockquote' => array(), - 'code' => array(), - 'pre' => array(), - 'a' => array( - 'href' => 'required', - 'title' => 'optional', - ), - 'strong' => array(), - 'b' => array(), - 'em' => array(), - 'i' => array(), - 'img' => array( - 'src' => 'required', - 'alt' => 'optional', - 'title' => 'optional', - ), - 'h1' => array(), - 'h2' => array(), - 'h3' => array(), - 'h4' => array(), - 'h5' => array(), - 'h6' => array(), - 'hr' => array(), - ); - /** - * html tags to be ignored (contents will be parsed) - * - * @var array<string> - */ - var $ignore = array( - 'html', - 'body', - ); - /** - * html tags to be dropped (contents will not be parsed!) - * - * @var array<string> - */ - var $drop = array( - 'script', - 'head', - 'style', - 'form', - 'area', - 'object', - 'param', - 'iframe', - ); - /** - * Markdown indents which could be wrapped - * @note: use strings in regex format - * - * @var array<string> - */ - var $wrappableIndents = array( - '\* ', # ul - '\d. ', # ol - '\d\d. ', # ol - '> ', # blockquote - '', # p - ); - /** - * list of chars which have to be escaped in normal text - * @note: use strings in regex format - * - * @var array - * - * TODO: what's with block chars / sequences at the beginning of a block? - */ - var $escapeInText = array( - '([-*_])([ ]{0,2}\1){2,}' => '\\\\$0|', # hr - '\*\*([^*\s]+)\*\*' => '\*\*$1\*\*', # strong - '\*([^*\s]+)\*' => '\*$1\*', # em - '__(?! |_)(.+)(?!<_| )__' => '\_\_$1\_\_', # em - '_(?! |_)(.+)(?!<_| )_' => '\_$1\_', # em - '`(.+)`' => '\`$1\`', # code - '\[(.+)\](\s*\()' => '\[$1\]$2', # links: [text] (url) => [text\] (url) - '\[(.+)\](\s*)\[(.*)\]' => '\[$1\]$2\[$3\]', # links: [text][id] => [text\][id\] - ); - /** - * wether last processed node was a block tag or not - * - * @var bool - */ - var $lastWasBlockTag = false; - /** - * name of last closed tag - * - * @var string - */ - var $lastClosedTag = ''; - /** - * iterate through the nodes and decide what we - * shall do with the current node - * - * @param void - * @return void - */ - function parse() { - $this->output = ''; - # drop tags - $this->parser->html = preg_replace('#<('.implode('|', $this->drop).')[^>]*>.*</\\1>#sU', '', $this->parser->html); - while ($this->parser->nextNode()) { - switch ($this->parser->nodeType) { - case 'doctype': - break; - case 'pi': - case 'comment': - if ($this->keepHTML) { - $this->flushLinebreaks(); - $this->out($this->parser->node); - $this->setLineBreaks(2); - } - # else drop - break; - case 'text': - $this->handleText(); - break; - case 'tag': - if (in_array($this->parser->tagName, $this->ignore)) { - break; - } - if ($this->parser->isStartTag) { - $this->flushLinebreaks(); - } - if ($this->skipConversion) { - $this->isMarkdownable(); # update notConverted - $this->handleTagToText(); - continue; - } - if (!$this->parser->keepWhitespace && $this->parser->isBlockElement && $this->parser->isStartTag) { - $this->parser->html = ltrim($this->parser->html); - } - if ($this->isMarkdownable()) { - if ($this->parser->isBlockElement && $this->parser->isStartTag && !$this->lastWasBlockTag && !empty($this->output)) { - if (!empty($this->buffer)) { - $str =& $this->buffer[count($this->buffer) -1]; - } else { - $str =& $this->output; - } - if (substr($str, -strlen($this->indent)-1) != "\n".$this->indent) { - $str .= "\n".$this->indent; - } - } - $func = 'handleTag_'.$this->parser->tagName; - $this->$func(); - if ($this->linksAfterEachParagraph && $this->parser->isBlockElement && !$this->parser->isStartTag && empty($this->parser->openTags)) { - $this->flushStacked(); - } - if (!$this->parser->isStartTag) { - $this->lastClosedTag = $this->parser->tagName; - } - } else { - $this->handleTagToText(); - $this->lastClosedTag = ''; - } - break; - default: - trigger_error('invalid node type', E_USER_ERROR); - break; - } - $this->lastWasBlockTag = $this->parser->nodeType == 'tag' && $this->parser->isStartTag && $this->parser->isBlockElement; - } - if (!empty($this->buffer)) { - trigger_error('buffer was not flushed, this is a bug. please report!', E_USER_WARNING); - while (!empty($this->buffer)) { - $this->out($this->unbuffer()); - } - } - ### cleanup - $this->output = rtrim(str_replace('&', '&', str_replace('<', '<', str_replace('>', '>', $this->output)))); - # end parsing, flush stacked tags - $this->flushStacked(); - $this->stack = array(); - } - /** - * check if current tag can be converted to Markdown - * - * @param void - * @return bool - */ - function isMarkdownable() { - if (!isset($this->isMarkdownable[$this->parser->tagName])) { - # simply not markdownable - return false; - } - if ($this->parser->isStartTag) { - $return = true; - if ($this->keepHTML) { - $diff = array_diff(array_keys($this->parser->tagAttributes), array_keys($this->isMarkdownable[$this->parser->tagName])); - if (!empty($diff)) { - # non markdownable attributes given - $return = false; - } - } - if ($return) { - foreach ($this->isMarkdownable[$this->parser->tagName] as $attr => $type) { - if ($type == 'required' && !isset($this->parser->tagAttributes[$attr])) { - # required markdown attribute not given - $return = false; - break; - } - } - } - if (!$return) { - array_push($this->notConverted, $this->parser->tagName.'::'.implode('/', $this->parser->openTags)); - } - return $return; - } else { - if (!empty($this->notConverted) && end($this->notConverted) === $this->parser->tagName.'::'.implode('/', $this->parser->openTags)) { - array_pop($this->notConverted); - return false; - } - return true; - } - } - /** - * output all stacked tags - * - * @param void - * @return void - */ - function flushStacked() { - # links - foreach ($this->stack as $tag => $a) { - if (!empty($a)) { - call_user_func(array(&$this, 'flushStacked_'.$tag)); - } - } - } - /** - * output link references (e.g. [1]: http://example.com "title"); - * - * @param void - * @return void - */ - function flushStacked_a() { - $out = false; - foreach ($this->stack['a'] as $k => $tag) { - if (!isset($tag['unstacked'])) { - if (!$out) { - $out = true; - $this->out("\n\n", true); - } else { - $this->out("\n", true); - } - $this->out(' ['.$tag['linkID'].']: '.$tag['href'].(isset($tag['title']) ? ' "'.$tag['title'].'"' : ''), true); - $tag['unstacked'] = true; - $this->stack['a'][$k] = $tag; - } - } - } - /** - * flush enqued linebreaks - * - * @param void - * @return void - */ - function flushLinebreaks() { - if ($this->lineBreaks && !empty($this->output)) { - $this->out(str_repeat("\n".$this->indent, $this->lineBreaks), true); - } - $this->lineBreaks = 0; - } - /** - * handle non Markdownable tags - * - * @param void - * @return void - */ - function handleTagToText() { - if (!$this->keepHTML) { - if (!$this->parser->isStartTag && $this->parser->isBlockElement) { - $this->setLineBreaks(2); - } - } else { - # dont convert to markdown inside this tag - /** TODO: markdown extra **/ - if (!$this->parser->isEmptyTag) { - if ($this->parser->isStartTag) { - if (!$this->skipConversion) { - $this->skipConversion = $this->parser->tagName.'::'.implode('/', $this->parser->openTags); - } - } else { - if ($this->skipConversion == $this->parser->tagName.'::'.implode('/', $this->parser->openTags)) { - $this->skipConversion = false; - } - } - } - - if ($this->parser->isBlockElement) { - if ($this->parser->isStartTag) { - if (in_array($this->parent(), array('ins', 'del'))) { - # looks like ins or del are block elements now - $this->out("\n", true); - $this->indent(' '); - } - if ($this->parser->tagName != 'pre') { - $this->out($this->parser->node."\n".$this->indent); - if (!$this->parser->isEmptyTag) { - $this->indent(' '); - } else { - $this->setLineBreaks(1); - } - $this->parser->html = ltrim($this->parser->html); - } else { - # don't indent inside <pre> tags - $this->out($this->parser->node); - static $indent; - $indent = $this->indent; - $this->indent = ''; - } - } else { - if (!$this->parser->keepWhitespace) { - $this->output = rtrim($this->output); - } - if ($this->parser->tagName != 'pre') { - $this->indent(' '); - $this->out("\n".$this->indent.$this->parser->node); - } else { - # reset indentation - $this->out($this->parser->node); - static $indent; - $this->indent = $indent; - } - - if (in_array($this->parent(), array('ins', 'del'))) { - # ins or del was block element - $this->out("\n"); - $this->indent(' '); - } - if ($this->parser->tagName == 'li') { - $this->setLineBreaks(1); - } else { - $this->setLineBreaks(2); - } - } - } else { - $this->out($this->parser->node); - } - if (in_array($this->parser->tagName, array('code', 'pre'))) { - if ($this->parser->isStartTag) { - $this->buffer(); - } else { - # add stuff so cleanup just reverses this - $this->out(str_replace('<', '&lt;', str_replace('>', '&gt;', $this->unbuffer()))); - } - } - } - } - /** - * handle plain text - * - * @param void - * @return void - */ - function handleText() { - if ($this->hasParent('pre') && strpos($this->parser->node, "\n") !== false) { - $this->parser->node = str_replace("\n", "\n".$this->indent, $this->parser->node); - } - if (!$this->hasParent('code') && !$this->hasParent('pre')) { - # entity decode - $this->parser->node = $this->decode($this->parser->node); - if (!$this->skipConversion) { - # escape some chars in normal Text - $this->parser->node = preg_replace($this->escapeInText['search'], $this->escapeInText['replace'], $this->parser->node); - } - } else { - $this->parser->node = str_replace(array('"', '&apos'), array('"', '\''), $this->parser->node); - } - $this->out($this->parser->node); - $this->lastClosedTag = ''; - } - /** - * handle <em> and <i> tags - * - * @param void - * @return void - */ - function handleTag_em() { - $this->out('*', true); - } - function handleTag_i() { - $this->handleTag_em(); - } - /** - * handle <strong> and <b> tags - * - * @param void - * @return void - */ - function handleTag_strong() { - $this->out('**', true); - } - function handleTag_b() { - $this->handleTag_strong(); - } - /** - * handle <h1> tags - * - * @param void - * @return void - */ - function handleTag_h1() { - $this->handleHeader(1); - } - /** - * handle <h2> tags - * - * @param void - * @return void - */ - function handleTag_h2() { - $this->handleHeader(2); - } - /** - * handle <h3> tags - * - * @param void - * @return void - */ - function handleTag_h3() { - $this->handleHeader(3); - } - /** - * handle <h4> tags - * - * @param void - * @return void - */ - function handleTag_h4() { - $this->handleHeader(4); - } - /** - * handle <h5> tags - * - * @param void - * @return void - */ - function handleTag_h5() { - $this->handleHeader(5); - } - /** - * handle <h6> tags - * - * @param void - * @return void - */ - function handleTag_h6() { - $this->handleHeader(6); - } - /** - * number of line breaks before next inline output - */ - var $lineBreaks = 0; - /** - * handle header tags (<h1> - <h6>) - * - * @param int $level 1-6 - * @return void - */ - function handleHeader($level) { - if ($this->parser->isStartTag) { - $this->out(str_repeat('#', $level).' ', true); - } else { - $this->setLineBreaks(2); - } - } - /** - * handle <p> tags - * - * @param void - * @return void - */ - function handleTag_p() { - if (!$this->parser->isStartTag) { - $this->setLineBreaks(2); - } - } - /** - * handle <a> tags - * - * @param void - * @return void - */ - function handleTag_a() { - if ($this->parser->isStartTag) { - $this->buffer(); - if (isset($this->parser->tagAttributes['title'])) { - $this->parser->tagAttributes['title'] = $this->decode($this->parser->tagAttributes['title']); - } else { - $this->parser->tagAttributes['title'] = null; - } - $this->parser->tagAttributes['href'] = $this->decode(trim($this->parser->tagAttributes['href'])); - $this->stack(); - } else { - $tag = $this->unstack(); - $buffer = $this->unbuffer(); - - if (empty($tag['href']) && empty($tag['title'])) { - # empty links... testcase mania, who would possibly do anything like that?! - $this->out('['.$buffer.']()', true); - return; - } - - if ($buffer == $tag['href'] && empty($tag['title'])) { - # <http://example.com> - $this->out('<'.$buffer.'>', true); - return; - } - - $bufferDecoded = $this->decode(trim($buffer)); - if (substr($tag['href'], 0, 7) == 'mailto:' && 'mailto:'.$bufferDecoded == $tag['href']) { - if (is_null($tag['title'])) { - # <mail@example.com> - $this->out('<'.$bufferDecoded.'>', true); - return; - } - # [mail@example.com][1] - # ... - # [1]: mailto:mail@example.com Title - $tag['href'] = 'mailto:'.$bufferDecoded; - } - - $this->out('['.$buffer.']('.$tag['href'].' "'.$tag['title'].'")', true); - -/* - # [This link][id] - foreach ($this->stack['a'] as $tag2) { - if ($tag2['href'] == $tag['href'] && $tag2['title'] === $tag['title']) { - $tag['linkID'] = $tag2['linkID']; - break; - } - } - if (!isset($tag['linkID'])) { - $tag['linkID'] = count($this->stack['a']) + 1; - array_push($this->stack['a'], $tag); - } - - $this->out('['.$buffer.']['.$tag['linkID'].']', true); -*/ - } - } - /** - * handle <img /> tags - * - * @param void - * @return void - */ - function handleTag_img() { - if (!$this->parser->isStartTag) { - return; # just to be sure this is really an empty tag... - } - - if (isset($this->parser->tagAttributes['title'])) { - $this->parser->tagAttributes['title'] = $this->decode($this->parser->tagAttributes['title']); - } else { - $this->parser->tagAttributes['title'] = null; - } - if (isset($this->parser->tagAttributes['alt'])) { - $this->parser->tagAttributes['alt'] = $this->decode($this->parser->tagAttributes['alt']); - } else { - $this->parser->tagAttributes['alt'] = null; - } - - if (empty($this->parser->tagAttributes['src'])) { - # support for "empty" images... dunno if this is really needed - # but there are some testcases which do that... - if (!empty($this->parser->tagAttributes['title'])) { - $this->parser->tagAttributes['title'] = ' '.$this->parser->tagAttributes['title'].' '; - } - $this->out('!['.$this->parser->tagAttributes['alt'].']('.$this->parser->tagAttributes['title'].')', true); - return; - } else { - $this->parser->tagAttributes['src'] = $this->decode($this->parser->tagAttributes['src']); - } - -// ![Alt text](/path/to/img.jpg "Optional title") - if ($this->parser->tagAttributes['title'] != "") - $this->out('!['.$this->parser->tagAttributes['alt'].']('.$this->parser->tagAttributes['src'].' "'.$this->parser->tagAttributes['title'].'")', true); - else - $this->out('!['.$this->parser->tagAttributes['alt'].']('.$this->parser->tagAttributes['src'].')', true); - -/* - # [This link][id] - $link_id = false; - if (!empty($this->stack['a'])) { - foreach ($this->stack['a'] as $tag) { - if ($tag['href'] == $this->parser->tagAttributes['src'] - && $tag['title'] === $this->parser->tagAttributes['title']) { - $link_id = $tag['linkID']; - break; - } - } - } else { - $this->stack['a'] = array(); - } - if (!$link_id) { - $link_id = count($this->stack['a']) + 1; - $tag = array( - 'href' => $this->parser->tagAttributes['src'], - 'linkID' => $link_id, - 'title' => $this->parser->tagAttributes['title'] - ); - array_push($this->stack['a'], $tag); - } - - $this->out('!['.$this->parser->tagAttributes['alt'].']['.$link_id.']', true); -*/ - } - /** - * handle <code> tags - * - * @param void - * @return void - */ - function handleTag_code() { - if ($this->hasParent('pre')) { - # ignore code blocks inside <pre> - return; - } - if ($this->parser->isStartTag) { - $this->buffer(); - } else { - $buffer = $this->unbuffer(); - # use as many backticks as needed - preg_match_all('#`+#', $buffer, $matches); - if (!empty($matches[0])) { - rsort($matches[0]); - - $ticks = '`'; - while (true) { - if (!in_array($ticks, $matches[0])) { - break; - } - $ticks .= '`'; - } - } else { - $ticks = '`'; - } - if ($buffer[0] == '`' || substr($buffer, -1) == '`') { - $buffer = ' '.$buffer.' '; - } - $this->out($ticks.$buffer.$ticks, true); - } - } - /** - * handle <pre> tags - * - * @param void - * @return void - */ - function handleTag_pre() { - if ($this->keepHTML && $this->parser->isStartTag) { - # check if a simple <code> follows - if (!preg_match('#^\s*<code\s*>#Us', $this->parser->html)) { - # this is no standard markdown code block - $this->handleTagToText(); - return; - } - } - $this->indent(' '); - if (!$this->parser->isStartTag) { - $this->setLineBreaks(2); - } else { - $this->parser->html = ltrim($this->parser->html); - } - } - /** - * handle <blockquote> tags - * - * @param void - * @return void - */ - function handleTag_blockquote() { - $this->indent('> '); - } - /** - * handle <ul> tags - * - * @param void - * @return void - */ - function handleTag_ul() { - if ($this->parser->isStartTag) { - $this->stack(); - if (!$this->keepHTML && $this->lastClosedTag == $this->parser->tagName) { - $this->out("\n".$this->indent.'<!-- -->'."\n".$this->indent."\n".$this->indent); - } - } else { - $this->unstack(); - if ($this->parent() != 'li' || preg_match('#^\s*(</li\s*>\s*<li\s*>\s*)?<(p|blockquote)\s*>#sU', $this->parser->html)) { - # dont make Markdown add unneeded paragraphs - $this->setLineBreaks(2); - } - } - } - /** - * handle <ul> tags - * - * @param void - * @return void - */ - function handleTag_ol() { - # same as above - $this->parser->tagAttributes['num'] = 0; - $this->handleTag_ul(); - } - /** - * handle <li> tags - * - * @param void - * @return void - */ - function handleTag_li() { - if ($this->parent() == 'ol') { - $parent =& $this->getStacked('ol'); - if ($this->parser->isStartTag) { - $parent['num']++; - $this->out($parent['num'].'.'.str_repeat(' ', 3 - strlen($parent['num'])), true); - } - $this->indent(' ', false); - } else { - if ($this->parser->isStartTag) { - $this->out('* ', true); - } - $this->indent(' ', false); - } - if (!$this->parser->isStartTag) { - $this->setLineBreaks(1); - } - } - /** - * handle <hr /> tags - * - * @param void - * @return void - */ - function handleTag_hr() { - if (!$this->parser->isStartTag) { - return; # just to be sure this really is an empty tag - } - $this->out('* * *', true); - $this->setLineBreaks(2); - } - /** - * handle <br /> tags - * - * @param void - * @return void - */ - function handleTag_br() { - $this->out(" \n".$this->indent, true); - $this->parser->html = ltrim($this->parser->html); - } - /** - * node stack, e.g. for <a> and <abbr> tags - * - * @var array<array> - */ - var $stack = array(); - /** - * add current node to the stack - * this only stores the attributes - * - * @param void - * @return void - */ - function stack() { - if (!isset($this->stack[$this->parser->tagName])) { - $this->stack[$this->parser->tagName] = array(); - } - array_push($this->stack[$this->parser->tagName], $this->parser->tagAttributes); - } - /** - * remove current tag from stack - * - * @param void - * @return array - */ - function unstack() { - if (!isset($this->stack[$this->parser->tagName]) || !is_array($this->stack[$this->parser->tagName])) { - trigger_error('Trying to unstack from empty stack. This must not happen.', E_USER_ERROR); - } - return array_pop($this->stack[$this->parser->tagName]); - } - /** - * get last stacked element of type $tagName - * - * @param string $tagName - * @return array - */ - function & getStacked($tagName) { - // no end() so it can be referenced - return $this->stack[$tagName][count($this->stack[$tagName])-1]; - } - /** - * set number of line breaks before next start tag - * - * @param int $number - * @return void - */ - function setLineBreaks($number) { - if ($this->lineBreaks < $number) { - $this->lineBreaks = $number; - } - } - /** - * stores current buffers - * - * @var array<string> - */ - var $buffer = array(); - /** - * buffer next parser output until unbuffer() is called - * - * @param void - * @return void - */ - function buffer() { - array_push($this->buffer, ''); - } - /** - * end current buffer and return buffered output - * - * @param void - * @return string - */ - function unbuffer() { - return array_pop($this->buffer); - } - /** - * append string to the correct var, either - * directly to $this->output or to the current - * buffers - * - * @param string $put - * @return void - */ - function out($put, $nowrap = false) { - if (empty($put)) { - return; - } - if (!empty($this->buffer)) { - $this->buffer[count($this->buffer) - 1] .= $put; - } else { - if ($this->bodyWidth && !$this->parser->keepWhitespace) { # wrap lines - // get last line - $pos = strrpos($this->output, "\n"); - if ($pos === false) { - $line = $this->output; - } else { - $line = substr($this->output, $pos); - } - - if ($nowrap) { - if ($put[0] != "\n" && $this->strlen($line) + $this->strlen($put) > $this->bodyWidth) { - $this->output .= "\n".$this->indent.$put; - } else { - $this->output .= $put; - } - return; - } else { - $put .= "\n"; # make sure we get all lines in the while below - $lineLen = $this->strlen($line); - while ($pos = strpos($put, "\n")) { - $putLine = substr($put, 0, $pos+1); - $put = substr($put, $pos+1); - $putLen = $this->strlen($putLine); - if ($lineLen + $putLen < $this->bodyWidth) { - $this->output .= $putLine; - $lineLen = $putLen; - } else { - $split = preg_split('#^(.{0,'.($this->bodyWidth - $lineLen).'})\b#', $putLine, 2, PREG_SPLIT_OFFSET_CAPTURE | PREG_SPLIT_DELIM_CAPTURE); - $this->output .= rtrim($split[1][0])."\n".$this->indent.$this->wordwrap(ltrim($split[2][0]), $this->bodyWidth, "\n".$this->indent, false); - } - } - $this->output = substr($this->output, 0, -1); - return; - } - } else { - $this->output .= $put; - } - } - } - /** - * current indentation - * - * @var string - */ - var $indent = ''; - /** - * indent next output (start tag) or unindent (end tag) - * - * @param string $str indentation - * @param bool $output add indendation to output - * @return void - */ - function indent($str, $output = true) { - if ($this->parser->isStartTag) { - $this->indent .= $str; - if ($output) { - $this->out($str, true); - } - } else { - $this->indent = substr($this->indent, 0, -strlen($str)); - } - } - /** - * decode email addresses - * - * @author derernst@gmx.ch <http://www.php.net/manual/en/function.html-entity-decode.php#68536> - * @author Milian Wolff <http://milianw.de> - */ - function decode($text, $quote_style = ENT_QUOTES) { - if (version_compare(PHP_VERSION, '5', '>=')) { - # UTF-8 is only supported in PHP 5.x.x and above - $text = html_entity_decode($text, $quote_style, 'UTF-8'); - } else { - if (function_exists('html_entity_decode')) { - $text = html_entity_decode($text, $quote_style, 'ISO-8859-1'); - } else { - static $trans_tbl; - if (!isset($trans_tbl)) { - $trans_tbl = array_flip(get_html_translation_table(HTML_ENTITIES, $quote_style)); - } - $text = strtr($text, $trans_tbl); - } - $text = preg_replace_callback('~&#x([0-9a-f]+);~i', array(&$this, '_decode_hex'), $text); - $text = preg_replace_callback('~&#(\d{2,5});~', array(&$this, '_decode_numeric'), $text); - } - return $text; - } - /** - * callback for decode() which converts a hexadecimal entity to UTF-8 - * - * @param array $matches - * @return string UTF-8 encoded - */ - function _decode_hex($matches) { - return $this->unichr(hexdec($matches[1])); - } - /** - * callback for decode() which converts a numerical entity to UTF-8 - * - * @param array $matches - * @return string UTF-8 encoded - */ - function _decode_numeric($matches) { - return $this->unichr($matches[1]); - } - /** - * UTF-8 chr() which supports numeric entities - * - * @author grey - greywyvern - com <http://www.php.net/manual/en/function.chr.php#55978> - * @param array $matches - * @return string UTF-8 encoded - */ - function unichr($dec) { - if ($dec < 128) { - $utf = chr($dec); - } else if ($dec < 2048) { - $utf = chr(192 + (($dec - ($dec % 64)) / 64)); - $utf .= chr(128 + ($dec % 64)); - } else { - $utf = chr(224 + (($dec - ($dec % 4096)) / 4096)); - $utf .= chr(128 + ((($dec % 4096) - ($dec % 64)) / 64)); - $utf .= chr(128 + ($dec % 64)); - } - return $utf; - } - /** - * UTF-8 strlen() - * - * @param string $str - * @return int - * - * @author dtorop 932 at hotmail dot com <http://www.php.net/manual/en/function.strlen.php#37975> - * @author Milian Wolff <http://milianw.de> - */ - function strlen($str) { - if (function_exists('mb_strlen')) { - return mb_strlen($str, 'UTF-8'); - } else { - return preg_match_all('/[\x00-\x7F\xC0-\xFD]/', $str, $var_empty); - } - } - /** - * wordwrap for utf8 encoded strings - * - * @param string $str - * @param integer $len - * @param string $what - * @return string - */ - function wordwrap($str, $width, $break, $cut = false){ - if (!$cut) { - $regexp = '#^(?:[\x00-\x7F]|[\xC0-\xFF][\x80-\xBF]+){1,'.$width.'}\b#'; - } else { - $regexp = '#^(?:[\x00-\x7F]|[\xC0-\xFF][\x80-\xBF]+){'.$width.'}#'; - } - $return = ''; - while (preg_match($regexp, $str, $matches)) { - $string = $matches[0]; - $str = ltrim(substr($str, strlen($string))); - if (!$cut && isset($str[0]) && in_array($str[0], array('.', '!', ';', ':', '?', ','))) { - $string .= $str[0]; - $str = ltrim(substr($str, 1)); - } - $return .= $string.$break; - } - return $return.ltrim($str); - } - /** - * check if current node has a $tagName as parent (somewhere, not only the direct parent) - * - * @param string $tagName - * @return bool - */ - function hasParent($tagName) { - return in_array($tagName, $this->parser->openTags); - } - /** - * get tagName of direct parent tag - * - * @param void - * @return string $tagName - */ - function parent() { - return end($this->parser->openTags); - } -} diff --git a/library/markdownify/markdownify_cli.php b/library/markdownify/markdownify_cli.php deleted file mode 100755 index b3fffbd5c..000000000 --- a/library/markdownify/markdownify_cli.php +++ /dev/null @@ -1,33 +0,0 @@ -#!/usr/bin/php -<?php -require dirname(__FILE__) .'/markdownify_extra.php'; - -function param($name, $default = false) { - if (!in_array('--'.$name, $_SERVER['argv'])) - return $default; - reset($_SERVER['argv']); - while (each($_SERVER['argv'])) { - if (current($_SERVER['argv']) == '--'.$name) - break; - } - $value = next($_SERVER['argv']); - if ($value === false || substr($value, 0, 2) == '--') - return true; - else - return $value; -} - - -$input = stream_get_contents(STDIN); - -$linksAfterEachParagraph = param('links'); -$bodyWidth = param('width'); -$keepHTML = param('html', true); - -if (param('no_extra')) { - $parser = new Markdownify($linksAfterEachParagraph, $bodyWidth, $keepHTML); -} else { - $parser = new Markdownify_Extra($linksAfterEachParagraph, $bodyWidth, $keepHTML); -} - -echo $parser->parseString($input) ."\n";
\ No newline at end of file diff --git a/library/markdownify/markdownify_extra.php b/library/markdownify/markdownify_extra.php deleted file mode 100644 index e978a1c8a..000000000 --- a/library/markdownify/markdownify_extra.php +++ /dev/null @@ -1,489 +0,0 @@ -<?php -/** - * Class to convert HTML to Markdown with PHP Markdown Extra syntax support. - * - * @version 1.0.0 alpha - * @author Milian Wolff (<mail@milianw.de>, <http://milianw.de>) - * @license LGPL, see LICENSE_LGPL.txt and the summary below - * @copyright (C) 2007 Milian Wolff - * - * This library is free software; you can redistribute it and/or - * modify it under the terms of the GNU Lesser General Public - * License as published by the Free Software Foundation; either - * version 2.1 of the License, or (at your option) any later version. - * - * This library is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * Lesser General Public License for more details. - * - * You should have received a copy of the GNU Lesser General Public - * License along with this library; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA - */ - -/** - * standard Markdownify class - */ -require_once dirname(__FILE__).'/markdownify.php'; - -class Markdownify_Extra extends Markdownify { - /** - * table data, including rows with content and the maximum width of each col - * - * @var array - */ - var $table = array(); - /** - * current col - * - * @var int - */ - var $col = -1; - /** - * current row - * - * @var int - */ - var $row = 0; - /** - * constructor, see Markdownify::Markdownify() for more information - */ - function Markdownify_Extra($linksAfterEachParagraph = MDFY_LINKS_EACH_PARAGRAPH, $bodyWidth = MDFY_BODYWIDTH, $keepHTML = MDFY_KEEPHTML) { - parent::Markdownify($linksAfterEachParagraph, $bodyWidth, $keepHTML); - - ### new markdownable tags & attributes - # header ids: # foo {bar} - $this->isMarkdownable['h1']['id'] = 'optional'; - $this->isMarkdownable['h2']['id'] = 'optional'; - $this->isMarkdownable['h3']['id'] = 'optional'; - $this->isMarkdownable['h4']['id'] = 'optional'; - $this->isMarkdownable['h5']['id'] = 'optional'; - $this->isMarkdownable['h6']['id'] = 'optional'; - # tables - $this->isMarkdownable['table'] = array(); - $this->isMarkdownable['th'] = array( - 'align' => 'optional', - ); - $this->isMarkdownable['td'] = array( - 'align' => 'optional', - ); - $this->isMarkdownable['tr'] = array(); - array_push($this->ignore, 'thead'); - array_push($this->ignore, 'tbody'); - array_push($this->ignore, 'tfoot'); - # definition lists - $this->isMarkdownable['dl'] = array(); - $this->isMarkdownable['dd'] = array(); - $this->isMarkdownable['dt'] = array(); - # footnotes - $this->isMarkdownable['fnref'] = array( - 'target' => 'required', - ); - $this->isMarkdownable['footnotes'] = array(); - $this->isMarkdownable['fn'] = array( - 'name' => 'required', - ); - $this->parser->blockElements['fnref'] = false; - $this->parser->blockElements['fn'] = true; - $this->parser->blockElements['footnotes'] = true; - # abbr - $this->isMarkdownable['abbr'] = array( - 'title' => 'required', - ); - # build RegEx lookahead to decide wether table can pe parsed or not - $inlineTags = array_keys($this->parser->blockElements, false); - $colContents = '(?:[^<]|<(?:'.implode('|', $inlineTags).'|[^a-z]))+'; - $this->tableLookaheadHeader = '{ - ^\s*(?:<thead\s*>)?\s* # open optional thead - <tr\s*>\s*(?: # start required row with headers - <th(?:\s+align=("|\')(?:left|center|right)\1)?\s*> # header with optional align - \s*'.$colContents.'\s* # contents - </th>\s* # close header - )+</tr> # close row with headers - \s*(?:</thead>)? # close optional thead - }sxi'; - $this->tdSubstitute = '\s*'.$colContents.'\s* # contents - </td>\s*'; - $this->tableLookaheadBody = '{ - \s*(?:<tbody\s*>)?\s* # open optional tbody - (?:<tr\s*>\s* # start row - %s # cols to be substituted - </tr>)+ # close row - \s*(?:</tbody>)? # close optional tbody - \s*</table> # close table - }sxi'; - } - /** - * handle header tags (<h1> - <h6>) - * - * @param int $level 1-6 - * @return void - */ - function handleHeader($level) { - static $id = null; - if ($this->parser->isStartTag) { - if (isset($this->parser->tagAttributes['id'])) { - $id = $this->parser->tagAttributes['id']; - } - } else { - if (!is_null($id)) { - $this->out(' {#'.$id.'}'); - $id = null; - } - } - parent::handleHeader($level); - } - /** - * handle <abbr> tags - * - * @param void - * @return void - */ - function handleTag_abbr() { - if ($this->parser->isStartTag) { - $this->stack(); - $this->buffer(); - } else { - $tag = $this->unstack(); - $tag['text'] = $this->unbuffer(); - $add = true; - foreach ($this->stack['abbr'] as $stacked) { - if ($stacked['text'] == $tag['text']) { - /** TODO: differing abbr definitions, i.e. different titles for same text **/ - $add = false; - break; - } - } - $this->out($tag['text']); - if ($add) { - array_push($this->stack['abbr'], $tag); - } - } - } - /** - * flush stacked abbr tags - * - * @param void - * @return void - */ - function flushStacked_abbr() { - $out = array(); - foreach ($this->stack['abbr'] as $k => $tag) { - if (!isset($tag['unstacked'])) { - array_push($out, ' *['.$tag['text'].']: '.$tag['title']); - $tag['unstacked'] = true; - $this->stack['abbr'][$k] = $tag; - } - } - if (!empty($out)) { - $this->out("\n\n".implode("\n", $out)); - } - } - /** - * handle <table> tags - * - * @param void - * @return void - */ - function handleTag_table() { - if ($this->parser->isStartTag) { - # check if upcoming table can be converted - if ($this->keepHTML) { - if (preg_match($this->tableLookaheadHeader, $this->parser->html, $matches)) { - # header seems good, now check body - # get align & number of cols - preg_match_all('#<th(?:\s+align=("|\')(left|right|center)\1)?\s*>#si', $matches[0], $cols); - $regEx = ''; - $i = 1; - $aligns = array(); - foreach ($cols[2] as $align) { - $align = strtolower($align); - array_push($aligns, $align); - if (empty($align)) { - $align = 'left'; # default value - } - $td = '\s+align=("|\')'.$align.'\\'.$i; - $i++; - if ($align == 'left') { - # look for empty align or left - $td = '(?:'.$td.')?'; - } - $td = '<td'.$td.'\s*>'; - $regEx .= $td.$this->tdSubstitute; - } - $regEx = sprintf($this->tableLookaheadBody, $regEx); - if (preg_match($regEx, $this->parser->html, $matches, null, strlen($matches[0]))) { - # this is a markdownable table tag! - $this->table = array( - 'rows' => array(), - 'col_widths' => array(), - 'aligns' => $aligns, - ); - $this->row = 0; - } else { - # non markdownable table - $this->handleTagToText(); - } - } else { - # non markdownable table - $this->handleTagToText(); - } - } else { - $this->table = array( - 'rows' => array(), - 'col_widths' => array(), - 'aligns' => array(), - ); - $this->row = 0; - } - } else { - # finally build the table in Markdown Extra syntax - $separator = array(); - # seperator with correct align identifikators - foreach($this->table['aligns'] as $col => $align) { - if (!$this->keepHTML && !isset($this->table['col_widths'][$col])) { - break; - } - $left = ' '; - $right = ' '; - switch ($align) { - case 'left': - $left = ':'; - break; - case 'center': - $right = ':'; - $left = ':'; - case 'right': - $right = ':'; - break; - } - array_push($separator, $left.str_repeat('-', $this->table['col_widths'][$col]).$right); - } - $separator = '|'.implode('|', $separator).'|'; - - $rows = array(); - # add padding - array_walk_recursive($this->table['rows'], array(&$this, 'alignTdContent')); - $header = array_shift($this->table['rows']); - array_push($rows, '| '.implode(' | ', $header).' |'); - array_push($rows, $separator); - foreach ($this->table['rows'] as $row) { - array_push($rows, '| '.implode(' | ', $row).' |'); - } - $this->out(implode("\n".$this->indent, $rows)); - $this->table = array(); - $this->setLineBreaks(2); - } - } - /** - * properly pad content so it is aligned as whished - * should be used with array_walk_recursive on $this->table['rows'] - * - * @param string &$content - * @param int $col - * @return void - */ - function alignTdContent(&$content, $col) { - switch ($this->table['aligns'][$col]) { - default: - case 'left': - $content .= str_repeat(' ', $this->table['col_widths'][$col] - $this->strlen($content)); - break; - case 'right': - $content = str_repeat(' ', $this->table['col_widths'][$col] - $this->strlen($content)).$content; - break; - case 'center': - $paddingNeeded = $this->table['col_widths'][$col] - $this->strlen($content); - $left = floor($paddingNeeded / 2); - $right = $paddingNeeded - $left; - $content = str_repeat(' ', $left).$content.str_repeat(' ', $right); - break; - } - } - /** - * handle <tr> tags - * - * @param void - * @return void - */ - function handleTag_tr() { - if ($this->parser->isStartTag) { - $this->col = -1; - } else { - $this->row++; - } - } - /** - * handle <td> tags - * - * @param void - * @return void - */ - function handleTag_td() { - if ($this->parser->isStartTag) { - $this->col++; - if (!isset($this->table['col_widths'][$this->col])) { - $this->table['col_widths'][$this->col] = 0; - } - $this->buffer(); - } else { - $buffer = trim($this->unbuffer()); - $this->table['col_widths'][$this->col] = max($this->table['col_widths'][$this->col], $this->strlen($buffer)); - $this->table['rows'][$this->row][$this->col] = $buffer; - } - } - /** - * handle <th> tags - * - * @param void - * @return void - */ - function handleTag_th() { - if (!$this->keepHTML && !isset($this->table['rows'][1]) && !isset($this->table['aligns'][$this->col+1])) { - if (isset($this->parser->tagAttributes['align'])) { - $this->table['aligns'][$this->col+1] = $this->parser->tagAttributes['align']; - } else { - $this->table['aligns'][$this->col+1] = ''; - } - } - $this->handleTag_td(); - } - /** - * handle <dl> tags - * - * @param void - * @return void - */ - function handleTag_dl() { - if (!$this->parser->isStartTag) { - $this->setLineBreaks(2); - } - } - /** - * handle <dt> tags - * - * @param void - * @return void - **/ - function handleTag_dt() { - if (!$this->parser->isStartTag) { - $this->setLineBreaks(1); - } - } - /** - * handle <dd> tags - * - * @param void - * @return void - */ - function handleTag_dd() { - if ($this->parser->isStartTag) { - if (substr(ltrim($this->parser->html), 0, 3) == '<p>') { - # next comes a paragraph, so we'll need an extra line - $this->out("\n".$this->indent); - } elseif (substr($this->output, -2) == "\n\n") { - $this->output = substr($this->output, 0, -1); - } - $this->out(': '); - $this->indent(' ', false); - } else { - # lookahead for next dt - if (substr(ltrim($this->parser->html), 0, 4) == '<dt>') { - $this->setLineBreaks(2); - } else { - $this->setLineBreaks(1); - } - $this->indent(' '); - } - } - /** - * handle <fnref /> tags (custom footnote references, see markdownify_extra::parseString()) - * - * @param void - * @return void - */ - function handleTag_fnref() { - $this->out('[^'.$this->parser->tagAttributes['target'].']'); - } - /** - * handle <fn> tags (custom footnotes, see markdownify_extra::parseString() - * and markdownify_extra::_makeFootnotes()) - * - * @param void - * @return void - */ - function handleTag_fn() { - if ($this->parser->isStartTag) { - $this->out('[^'.$this->parser->tagAttributes['name'].']:'); - $this->setLineBreaks(1); - } else { - $this->setLineBreaks(2); - } - $this->indent(' '); - } - /** - * handle <footnotes> tag (custom footnotes, see markdownify_extra::parseString() - * and markdownify_extra::_makeFootnotes()) - * - * @param void - * @return void - */ - function handleTag_footnotes() { - if (!$this->parser->isStartTag) { - $this->setLineBreaks(2); - } - } - /** - * parse a HTML string, clean up footnotes prior - * - * @param string $HTML input - * @return string Markdown formatted output - */ - function parseString($html) { - /** TODO: custom markdown-extra options, e.g. titles & classes **/ - # <sup id="fnref:..."><a href"#fn..." rel="footnote">...</a></sup> - # => <fnref target="..." /> - $html = preg_replace('@<sup id="fnref:([^"]+)">\s*<a href="#fn:\1" rel="footnote">\s*\d+\s*</a>\s*</sup>@Us', '<fnref target="$1" />', $html); - # <div class="footnotes"> - # <hr /> - # <ol> - # - # <li id="fn:...">...</li> - # ... - # - # </ol> - # </div> - # => - # <footnotes> - # <fn name="...">...</fn> - # ... - # </footnotes> - $html = preg_replace_callback('#<div class="footnotes">\s*<hr />\s*<ol>\s*(.+)\s*</ol>\s*</div>#Us', array(&$this, '_makeFootnotes'), $html); - return parent::parseString($html); - } - /** - * replace HTML representation of footnotes with something more easily parsable - * - * @note this is a callback to be used in parseString() - * - * @param array $matches - * @return string - */ - function _makeFootnotes($matches) { - # <li id="fn:1"> - # ... - # <a href="#fnref:block" rev="footnote">↩</a></p> - # </li> - # => <fn name="1">...</fn> - # remove footnote link - $fns = preg_replace('@\s*( \s*)?<a href="#fnref:[^"]+" rev="footnote"[^>]*>↩</a>\s*@s', '', $matches[1]); - # remove empty paragraph - $fns = preg_replace('@<p>\s*</p>@s', '', $fns); - # <li id="fn:1">...</li> -> <footnote nr="1">...</footnote> - $fns = str_replace('<li id="fn:', '<fn name="', $fns); - - $fns = '<footnotes>'.$fns.'</footnotes>'; - return preg_replace('#</li>\s*(?=(?:<fn|</footnotes>))#s', '</fn>$1', $fns); - } -}
\ No newline at end of file diff --git a/library/markdownify/parsehtml/parsehtml.php b/library/markdownify/parsehtml/parsehtml.php deleted file mode 100644 index 1a8ecacda..000000000 --- a/library/markdownify/parsehtml/parsehtml.php +++ /dev/null @@ -1,618 +0,0 @@ -<?php -/** - * parseHTML is a HTML parser which works with PHP 4 and above. - * It tries to handle invalid HTML to some degree. - * - * @version 1.0 beta - * @author Milian Wolff (mail@milianw.de, http://milianw.de) - * @license LGPL, see LICENSE_LGPL.txt and the summary below - * @copyright (C) 2007 Milian Wolff - * - * This library is free software; you can redistribute it and/or - * modify it under the terms of the GNU Lesser General Public - * License as published by the Free Software Foundation; either - * version 2.1 of the License, or (at your option) any later version. - * - * This library is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * Lesser General Public License for more details. - * - * You should have received a copy of the GNU Lesser General Public - * License along with this library; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA - */ -class parseHTML { - /** - * tags which are always empty (<br /> etc.) - * - * @var array<string> - */ - var $emptyTags = array( - 'br', - 'hr', - 'input', - 'img', - 'area', - 'link', - 'meta', - 'param', - ); - /** - * tags with preformatted text - * whitespaces wont be touched in them - * - * @var array<string> - */ - var $preformattedTags = array( - 'script', - 'style', - 'pre', - 'code', - ); - /** - * supress HTML tags inside preformatted tags (see above) - * - * @var bool - */ - var $noTagsInCode = false; - /** - * html to be parsed - * - * @var string - */ - var $html = ''; - /** - * node type: - * - * - tag (see isStartTag) - * - text (includes cdata) - * - comment - * - doctype - * - pi (processing instruction) - * - * @var string - */ - var $nodeType = ''; - /** - * current node content, i.e. either a - * simple string (text node), or something like - * <tag attrib="value"...> - * - * @var string - */ - var $node = ''; - /** - * wether current node is an opening tag (<a>) or not (</a>) - * set to NULL if current node is not a tag - * NOTE: empty tags (<br />) set this to true as well! - * - * @var bool | null - */ - var $isStartTag = null; - /** - * wether current node is an empty tag (<br />) or not (<a></a>) - * - * @var bool | null - */ - var $isEmptyTag = null; - /** - * tag name - * - * @var string | null - */ - var $tagName = ''; - /** - * attributes of current tag - * - * @var array (attribName=>value) | null - */ - var $tagAttributes = null; - /** - * wether the current tag is a block element - * - * @var bool | null - */ - var $isBlockElement = null; - - /** - * keep whitespace - * - * @var int - */ - var $keepWhitespace = 0; - /** - * list of open tags - * count this to get current depth - * - * @var array - */ - var $openTags = array(); - /** - * list of block elements - * - * @var array - * TODO: what shall we do with <del> and <ins> ?! - */ - var $blockElements = array ( - # tag name => <bool> is block - # block elements - 'address' => true, - 'blockquote' => true, - 'center' => true, - 'del' => true, - 'dir' => true, - 'div' => true, - 'dl' => true, - 'fieldset' => true, - 'form' => true, - 'h1' => true, - 'h2' => true, - 'h3' => true, - 'h4' => true, - 'h5' => true, - 'h6' => true, - 'hr' => true, - 'ins' => true, - 'isindex' => true, - 'menu' => true, - 'noframes' => true, - 'noscript' => true, - 'ol' => true, - 'p' => true, - 'pre' => true, - 'table' => true, - 'ul' => true, - # set table elements and list items to block as well - 'thead' => true, - 'tbody' => true, - 'tfoot' => true, - 'td' => true, - 'tr' => true, - 'th' => true, - 'li' => true, - 'dd' => true, - 'dt' => true, - # header items and html / body as well - 'html' => true, - 'body' => true, - 'head' => true, - 'meta' => true, - 'link' => true, - 'style' => true, - 'title' => true, - # unfancy media tags, when indented should be rendered as block - 'map' => true, - 'object' => true, - 'param' => true, - 'embed' => true, - 'area' => true, - # inline elements - 'a' => false, - 'abbr' => false, - 'acronym' => false, - 'applet' => false, - 'b' => false, - 'basefont' => false, - 'bdo' => false, - 'big' => false, - 'br' => false, - 'button' => false, - 'cite' => false, - 'code' => false, - 'del' => false, - 'dfn' => false, - 'em' => false, - 'font' => false, - 'i' => false, - 'img' => false, - 'ins' => false, - 'input' => false, - 'iframe' => false, - 'kbd' => false, - 'label' => false, - 'q' => false, - 'samp' => false, - 'script' => false, - 'select' => false, - 'small' => false, - 'span' => false, - 'strong' => false, - 'sub' => false, - 'sup' => false, - 'textarea' => false, - 'tt' => false, - 'var' => false, - ); - /** - * get next node, set $this->html prior! - * - * @param void - * @return bool - */ - function nextNode() { - if (empty($this->html)) { - # we are done with parsing the html string - return false; - } - static $skipWhitespace = true; - if ($this->isStartTag && !$this->isEmptyTag) { - array_push($this->openTags, $this->tagName); - if (in_array($this->tagName, $this->preformattedTags)) { - # dont truncate whitespaces for <code> or <pre> contents - $this->keepWhitespace++; - } - } - - if ($this->html[0] == '<') { - $token = substr($this->html, 0, 9); - if (substr($token, 0, 2) == '<?') { - # xml prolog or other pi's - /** TODO **/ - #trigger_error('this might need some work', E_USER_NOTICE); - $pos = strpos($this->html, '>'); - $this->setNode('pi', $pos + 1); - return true; - } - if (substr($token, 0, 4) == '<!--') { - # comment - $pos = strpos($this->html, '-->'); - if ($pos === false) { - # could not find a closing -->, use next gt instead - # this is firefox' behaviour - $pos = strpos($this->html, '>') + 1; - } else { - $pos += 3; - } - $this->setNode('comment', $pos); - - $skipWhitespace = true; - return true; - } - if ($token == '<!DOCTYPE') { - # doctype - $this->setNode('doctype', strpos($this->html, '>')+1); - - $skipWhitespace = true; - return true; - } - if ($token == '<![CDATA[') { - # cdata, use text node - - # remove leading <![CDATA[ - $this->html = substr($this->html, 9); - - $this->setNode('text', strpos($this->html, ']]>')+3); - - # remove trailing ]]> and trim - $this->node = substr($this->node, 0, -3); - $this->handleWhitespaces(); - - $skipWhitespace = true; - return true; - } - if ($this->parseTag()) { - # seems to be a tag - # handle whitespaces - if ($this->isBlockElement) { - $skipWhitespace = true; - } else { - $skipWhitespace = false; - } - return true; - } - } - if ($this->keepWhitespace) { - $skipWhitespace = false; - } - # when we get here it seems to be a text node - $pos = strpos($this->html, '<'); - if ($pos === false) { - $pos = strlen($this->html); - } - $this->setNode('text', $pos); - $this->handleWhitespaces(); - if ($skipWhitespace && $this->node == ' ') { - return $this->nextNode(); - } - $skipWhitespace = false; - return true; - } - /** - * parse tag, set tag name and attributes, see if it's a closing tag and so forth... - * - * @param void - * @return bool - */ - function parseTag() { - static $a_ord, $z_ord, $special_ords; - if (!isset($a_ord)) { - $a_ord = ord('a'); - $z_ord = ord('z'); - $special_ords = array( - ord(':'), // for xml:lang - ord('-'), // for http-equiv - ); - } - - $tagName = ''; - - $pos = 1; - $isStartTag = $this->html[$pos] != '/'; - if (!$isStartTag) { - $pos++; - } - # get tagName - while (isset($this->html[$pos])) { - $pos_ord = ord(strtolower($this->html[$pos])); - if (($pos_ord >= $a_ord && $pos_ord <= $z_ord) || (!empty($tagName) && is_numeric($this->html[$pos]))) { - $tagName .= $this->html[$pos]; - $pos++; - } else { - $pos--; - break; - } - } - - $tagName = strtolower($tagName); - if (empty($tagName) || !isset($this->blockElements[$tagName])) { - # something went wrong => invalid tag - $this->invalidTag(); - return false; - } - if ($this->noTagsInCode && end($this->openTags) == 'code' && !($tagName == 'code' && !$isStartTag)) { - # we supress all HTML tags inside code tags - $this->invalidTag(); - return false; - } - - # get tag attributes - /** TODO: in html 4 attributes do not need to be quoted **/ - $isEmptyTag = false; - $attributes = array(); - $currAttrib = ''; - while (isset($this->html[$pos+1])) { - $pos++; - # close tag - if ($this->html[$pos] == '>' || $this->html[$pos].$this->html[$pos+1] == '/>') { - if ($this->html[$pos] == '/') { - $isEmptyTag = true; - $pos++; - } - break; - } - - $pos_ord = ord(strtolower($this->html[$pos])); - if ( ($pos_ord >= $a_ord && $pos_ord <= $z_ord) || in_array($pos_ord, $special_ords)) { - # attribute name - $currAttrib .= $this->html[$pos]; - } elseif (in_array($this->html[$pos], array(' ', "\t", "\n"))) { - # drop whitespace - } elseif (in_array($this->html[$pos].$this->html[$pos+1], array('="', "='"))) { - # get attribute value - $pos++; - $await = $this->html[$pos]; # single or double quote - $pos++; - $value = ''; - while (isset($this->html[$pos]) && $this->html[$pos] != $await) { - $value .= $this->html[$pos]; - $pos++; - } - $attributes[$currAttrib] = $value; - $currAttrib = ''; - } else { - $this->invalidTag(); - return false; - } - } - if ($this->html[$pos] != '>') { - $this->invalidTag(); - return false; - } - - if (!empty($currAttrib)) { - # html 4 allows something like <option selected> instead of <option selected="selected"> - $attributes[$currAttrib] = $currAttrib; - } - if (!$isStartTag) { - if (!empty($attributes) || $tagName != end($this->openTags)) { - # end tags must not contain any attributes - # or maybe we did not expect a different tag to be closed - $this->invalidTag(); - return false; - } - array_pop($this->openTags); - if (in_array($tagName, $this->preformattedTags)) { - $this->keepWhitespace--; - } - } - $pos++; - $this->node = substr($this->html, 0, $pos); - $this->html = substr($this->html, $pos); - $this->tagName = $tagName; - $this->tagAttributes = $attributes; - $this->isStartTag = $isStartTag; - $this->isEmptyTag = $isEmptyTag || in_array($tagName, $this->emptyTags); - if ($this->isEmptyTag) { - # might be not well formed - $this->node = preg_replace('# */? *>$#', ' />', $this->node); - } - $this->nodeType = 'tag'; - $this->isBlockElement = $this->blockElements[$tagName]; - return true; - } - /** - * handle invalid tags - * - * @param void - * @return void - */ - function invalidTag() { - $this->html = substr_replace($this->html, '<', 0, 1); - } - /** - * update all vars and make $this->html shorter - * - * @param string $type see description for $this->nodeType - * @param int $pos to which position shall we cut? - * @return void - */ - function setNode($type, $pos) { - if ($this->nodeType == 'tag') { - # set tag specific vars to null - # $type == tag should not be called here - # see this::parseTag() for more - $this->tagName = null; - $this->tagAttributes = null; - $this->isStartTag = null; - $this->isEmptyTag = null; - $this->isBlockElement = null; - - } - $this->nodeType = $type; - $this->node = substr($this->html, 0, $pos); - $this->html = substr($this->html, $pos); - } - /** - * check if $this->html begins with $str - * - * @param string $str - * @return bool - */ - function match($str) { - return substr($this->html, 0, strlen($str)) == $str; - } - /** - * truncate whitespaces - * - * @param void - * @return void - */ - function handleWhitespaces() { - if ($this->keepWhitespace) { - # <pre> or <code> before... - return; - } - # truncate multiple whitespaces to a single one - $this->node = preg_replace('#\s+#s', ' ', $this->node); - } - /** - * normalize self::node - * - * @param void - * @return void - */ - function normalizeNode() { - $this->node = '<'; - if (!$this->isStartTag) { - $this->node .= '/'.$this->tagName.'>'; - return; - } - $this->node .= $this->tagName; - foreach ($this->tagAttributes as $name => $value) { - $this->node .= ' '.$name.'="'.str_replace('"', '"', $value).'"'; - } - if ($this->isEmptyTag) { - $this->node .= ' /'; - } - $this->node .= '>'; - } -} - -/** - * indent a HTML string properly - * - * @param string $html - * @param string $indent optional - * @return string - */ -function indentHTML($html, $indent = " ", $noTagsInCode = false) { - $parser = new parseHTML; - $parser->noTagsInCode = $noTagsInCode; - $parser->html = $html; - $html = ''; - $last = true; # last tag was block elem - $indent_a = array(); - while($parser->nextNode()) { - if ($parser->nodeType == 'tag') { - $parser->normalizeNode(); - } - if ($parser->nodeType == 'tag' && $parser->isBlockElement) { - $isPreOrCode = in_array($parser->tagName, array('code', 'pre')); - if (!$parser->keepWhitespace && !$last && !$isPreOrCode) { - $html = rtrim($html)."\n"; - } - if ($parser->isStartTag) { - $html .= implode($indent_a); - if (!$parser->isEmptyTag) { - array_push($indent_a, $indent); - } - } else { - array_pop($indent_a); - if (!$isPreOrCode) { - $html .= implode($indent_a); - } - } - $html .= $parser->node; - if (!$parser->keepWhitespace && !($isPreOrCode && $parser->isStartTag)) { - $html .= "\n"; - } - $last = true; - } else { - if ($parser->nodeType == 'tag' && $parser->tagName == 'br') { - $html .= $parser->node."\n"; - $last = true; - continue; - } elseif ($last && !$parser->keepWhitespace) { - $html .= implode($indent_a); - $parser->node = ltrim($parser->node); - } - $html .= $parser->node; - - if (in_array($parser->nodeType, array('comment', 'pi', 'doctype'))) { - $html .= "\n"; - } else { - $last = false; - } - } - } - return $html; -} -/* -# testcase / example -error_reporting(E_ALL); - -$html = '<p>Simple block on one line:</p> - -<div>foo</div> - -<p>And nested without indentation:</p> - -<div> -<div> -<div> -foo -</div> -<div style=">"/> -</div> -<div>bar</div> -</div> - -<p>And with attributes:</p> - -<div> - <div id="foo"> - </div> -</div> - -<p>This was broken in 1.0.2b7:</p> - -<div class="inlinepage"> -<div class="toggleableend"> -foo -</div> -</div>'; -#$html = '<a href="asdfasdf" title=\'asdf\' foo="bar">asdf</a>'; -echo indentHTML($html); -die(); -*/ diff --git a/vendor/composer/autoload_classmap.php b/vendor/composer/autoload_classmap.php index 8b8d3dd58..2adb2fe53 100644 --- a/vendor/composer/autoload_classmap.php +++ b/vendor/composer/autoload_classmap.php @@ -7,6 +7,9 @@ $baseDir = dirname($vendorDir); return array( 'Hubzilla\\Import\\Import' => $baseDir . '/include/Import/Importer.php', + 'Markdownify\\Converter' => $vendorDir . '/pixel418/markdownify/src/Converter.php', + 'Markdownify\\ConverterExtra' => $vendorDir . '/pixel418/markdownify/src/ConverterExtra.php', + 'Markdownify\\Parser' => $vendorDir . '/pixel418/markdownify/src/Parser.php', 'Michelf\\Markdown' => $vendorDir . '/michelf/php-markdown/Michelf/Markdown.php', 'Michelf\\MarkdownExtra' => $vendorDir . '/michelf/php-markdown/Michelf/MarkdownExtra.php', 'Michelf\\MarkdownInterface' => $vendorDir . '/michelf/php-markdown/Michelf/MarkdownInterface.php', @@ -18,6 +21,8 @@ return array( 'Psr\\Log\\LoggerInterface' => $vendorDir . '/psr/log/Psr/Log/LoggerInterface.php', 'Psr\\Log\\LoggerTrait' => $vendorDir . '/psr/log/Psr/Log/LoggerTrait.php', 'Psr\\Log\\NullLogger' => $vendorDir . '/psr/log/Psr/Log/NullLogger.php', + 'Psr\\Log\\Test\\DummyTest' => $vendorDir . '/psr/log/Psr/Log/Test/LoggerInterfaceTest.php', + 'Psr\\Log\\Test\\LoggerInterfaceTest' => $vendorDir . '/psr/log/Psr/Log/Test/LoggerInterfaceTest.php', 'Sabre\\CalDAV\\Backend\\AbstractBackend' => $vendorDir . '/sabre/dav/lib/CalDAV/Backend/AbstractBackend.php', 'Sabre\\CalDAV\\Backend\\BackendInterface' => $vendorDir . '/sabre/dav/lib/CalDAV/Backend/BackendInterface.php', 'Sabre\\CalDAV\\Backend\\NotificationSupport' => $vendorDir . '/sabre/dav/lib/CalDAV/Backend/NotificationSupport.php', @@ -268,7 +273,6 @@ return array( 'Sabre\\HTTP\\URLUtil' => $vendorDir . '/sabre/http/lib/URLUtil.php', 'Sabre\\HTTP\\Util' => $vendorDir . '/sabre/http/lib/Util.php', 'Sabre\\HTTP\\Version' => $vendorDir . '/sabre/http/lib/Version.php', - 'Sabre\\Uri\\InvalidUriException' => $vendorDir . '/sabre/uri/lib/InvalidUriException.php', 'Sabre\\Uri\\Version' => $vendorDir . '/sabre/uri/lib/Version.php', 'Sabre\\VObject\\BirthdayCalendarGenerator' => $vendorDir . '/sabre/vobject/lib/BirthdayCalendarGenerator.php', 'Sabre\\VObject\\Cli' => $vendorDir . '/sabre/vobject/lib/Cli.php', @@ -357,6 +361,9 @@ return array( 'Sabre\\Xml\\Writer' => $vendorDir . '/sabre/xml/lib/Writer.php', 'Sabre\\Xml\\XmlDeserializable' => $vendorDir . '/sabre/xml/lib/XmlDeserializable.php', 'Sabre\\Xml\\XmlSerializable' => $vendorDir . '/sabre/xml/lib/XmlSerializable.php', + 'Test\\Markdownify\\ConverterExtraTest' => $vendorDir . '/pixel418/markdownify/test/ConverterExtraTest.php', + 'Test\\Markdownify\\ConverterTest' => $vendorDir . '/pixel418/markdownify/test/ConverterTest.php', + 'Test\\Markdownify\\ConverterTestCase' => $vendorDir . '/pixel418/markdownify/test/ConverterTestCase.php', 'Zotlabs\\Access\\AccessList' => $baseDir . '/Zotlabs/Access/AccessList.php', 'Zotlabs\\Access\\PermissionLimits' => $baseDir . '/Zotlabs/Access/PermissionLimits.php', 'Zotlabs\\Access\\PermissionRoles' => $baseDir . '/Zotlabs/Access/PermissionRoles.php', diff --git a/vendor/composer/autoload_psr4.php b/vendor/composer/autoload_psr4.php index e8ea2ed78..00a183cc1 100644 --- a/vendor/composer/autoload_psr4.php +++ b/vendor/composer/autoload_psr4.php @@ -7,6 +7,7 @@ $baseDir = dirname($vendorDir); return array( 'Zotlabs\\' => array($baseDir . '/Zotlabs'), + 'Test\\Markdownify\\' => array($vendorDir . '/pixel418/markdownify/test'), 'Sabre\\Xml\\' => array($vendorDir . '/sabre/xml/lib'), 'Sabre\\VObject\\' => array($vendorDir . '/sabre/vobject/lib'), 'Sabre\\Uri\\' => array($vendorDir . '/sabre/uri/lib'), @@ -17,5 +18,6 @@ return array( 'Sabre\\CardDAV\\' => array($vendorDir . '/sabre/dav/lib/CardDAV'), 'Sabre\\CalDAV\\' => array($vendorDir . '/sabre/dav/lib/CalDAV'), 'Psr\\Log\\' => array($vendorDir . '/psr/log/Psr/Log'), + 'Markdownify\\' => array($vendorDir . '/pixel418/markdownify/src'), 'Hubzilla\\' => array($baseDir . '/include'), ); diff --git a/vendor/composer/autoload_static.php b/vendor/composer/autoload_static.php index 070595527..a7c84b7f1 100644 --- a/vendor/composer/autoload_static.php +++ b/vendor/composer/autoload_static.php @@ -21,6 +21,10 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d array ( 'Zotlabs\\' => 8, ), + 'T' => + array ( + 'Test\\Markdownify\\' => 17, + ), 'S' => array ( 'Sabre\\Xml\\' => 10, @@ -37,6 +41,10 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d array ( 'Psr\\Log\\' => 8, ), + 'M' => + array ( + 'Markdownify\\' => 12, + ), 'H' => array ( 'Hubzilla\\' => 9, @@ -48,6 +56,10 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d array ( 0 => __DIR__ . '/../..' . '/Zotlabs', ), + 'Test\\Markdownify\\' => + array ( + 0 => __DIR__ . '/..' . '/pixel418/markdownify/test', + ), 'Sabre\\Xml\\' => array ( 0 => __DIR__ . '/..' . '/sabre/xml/lib', @@ -88,6 +100,10 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d array ( 0 => __DIR__ . '/..' . '/psr/log/Psr/Log', ), + 'Markdownify\\' => + array ( + 0 => __DIR__ . '/..' . '/pixel418/markdownify/src', + ), 'Hubzilla\\' => array ( 0 => __DIR__ . '/../..' . '/include', @@ -106,6 +122,9 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d public static $classMap = array ( 'Hubzilla\\Import\\Import' => __DIR__ . '/../..' . '/include/Import/Importer.php', + 'Markdownify\\Converter' => __DIR__ . '/..' . '/pixel418/markdownify/src/Converter.php', + 'Markdownify\\ConverterExtra' => __DIR__ . '/..' . '/pixel418/markdownify/src/ConverterExtra.php', + 'Markdownify\\Parser' => __DIR__ . '/..' . '/pixel418/markdownify/src/Parser.php', 'Michelf\\Markdown' => __DIR__ . '/..' . '/michelf/php-markdown/Michelf/Markdown.php', 'Michelf\\MarkdownExtra' => __DIR__ . '/..' . '/michelf/php-markdown/Michelf/MarkdownExtra.php', 'Michelf\\MarkdownInterface' => __DIR__ . '/..' . '/michelf/php-markdown/Michelf/MarkdownInterface.php', @@ -117,6 +136,8 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d 'Psr\\Log\\LoggerInterface' => __DIR__ . '/..' . '/psr/log/Psr/Log/LoggerInterface.php', 'Psr\\Log\\LoggerTrait' => __DIR__ . '/..' . '/psr/log/Psr/Log/LoggerTrait.php', 'Psr\\Log\\NullLogger' => __DIR__ . '/..' . '/psr/log/Psr/Log/NullLogger.php', + 'Psr\\Log\\Test\\DummyTest' => __DIR__ . '/..' . '/psr/log/Psr/Log/Test/LoggerInterfaceTest.php', + 'Psr\\Log\\Test\\LoggerInterfaceTest' => __DIR__ . '/..' . '/psr/log/Psr/Log/Test/LoggerInterfaceTest.php', 'Sabre\\CalDAV\\Backend\\AbstractBackend' => __DIR__ . '/..' . '/sabre/dav/lib/CalDAV/Backend/AbstractBackend.php', 'Sabre\\CalDAV\\Backend\\BackendInterface' => __DIR__ . '/..' . '/sabre/dav/lib/CalDAV/Backend/BackendInterface.php', 'Sabre\\CalDAV\\Backend\\NotificationSupport' => __DIR__ . '/..' . '/sabre/dav/lib/CalDAV/Backend/NotificationSupport.php', @@ -367,7 +388,6 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d 'Sabre\\HTTP\\URLUtil' => __DIR__ . '/..' . '/sabre/http/lib/URLUtil.php', 'Sabre\\HTTP\\Util' => __DIR__ . '/..' . '/sabre/http/lib/Util.php', 'Sabre\\HTTP\\Version' => __DIR__ . '/..' . '/sabre/http/lib/Version.php', - 'Sabre\\Uri\\InvalidUriException' => __DIR__ . '/..' . '/sabre/uri/lib/InvalidUriException.php', 'Sabre\\Uri\\Version' => __DIR__ . '/..' . '/sabre/uri/lib/Version.php', 'Sabre\\VObject\\BirthdayCalendarGenerator' => __DIR__ . '/..' . '/sabre/vobject/lib/BirthdayCalendarGenerator.php', 'Sabre\\VObject\\Cli' => __DIR__ . '/..' . '/sabre/vobject/lib/Cli.php', @@ -456,6 +476,9 @@ class ComposerStaticInit7b34d7e50a62201ec5d5e526a5b8b35d 'Sabre\\Xml\\Writer' => __DIR__ . '/..' . '/sabre/xml/lib/Writer.php', 'Sabre\\Xml\\XmlDeserializable' => __DIR__ . '/..' . '/sabre/xml/lib/XmlDeserializable.php', 'Sabre\\Xml\\XmlSerializable' => __DIR__ . '/..' . '/sabre/xml/lib/XmlSerializable.php', + 'Test\\Markdownify\\ConverterExtraTest' => __DIR__ . '/..' . '/pixel418/markdownify/test/ConverterExtraTest.php', + 'Test\\Markdownify\\ConverterTest' => __DIR__ . '/..' . '/pixel418/markdownify/test/ConverterTest.php', + 'Test\\Markdownify\\ConverterTestCase' => __DIR__ . '/..' . '/pixel418/markdownify/test/ConverterTestCase.php', 'Zotlabs\\Access\\AccessList' => __DIR__ . '/../..' . '/Zotlabs/Access/AccessList.php', 'Zotlabs\\Access\\PermissionLimits' => __DIR__ . '/../..' . '/Zotlabs/Access/PermissionLimits.php', 'Zotlabs\\Access\\PermissionRoles' => __DIR__ . '/../..' . '/Zotlabs/Access/PermissionRoles.php', diff --git a/vendor/composer/installed.json b/vendor/composer/installed.json index 73827081f..0fb88e5a1 100644 --- a/vendor/composer/installed.json +++ b/vendor/composer/installed.json @@ -518,5 +518,63 @@ "keywords": [ "markdown" ] + }, + { + "name": "pixel418/markdownify", + "version": "v2.2.1", + "version_normalized": "2.2.1.0", + "source": { + "type": "git", + "url": "https://github.com/Elephant418/Markdownify.git", + "reference": "0160677f04c784550dd10fd72fdf3994967db848" + }, + "dist": { + "type": "zip", + "url": "https://api.github.com/repos/Elephant418/Markdownify/zipball/0160677f04c784550dd10fd72fdf3994967db848", + "reference": "0160677f04c784550dd10fd72fdf3994967db848", + "shasum": "" + }, + "require": { + "php": ">=5.3.0" + }, + "require-dev": { + "phpunit/phpunit": "^4.8" + }, + "time": "2016-09-21T13:01:43+00:00", + "type": "lib", + "installation-source": "dist", + "autoload": { + "psr-4": { + "Markdownify\\": "src", + "Test\\Markdownify\\": "test" + } + }, + "notification-url": "https://packagist.org/downloads/", + "license": [ + "LGPL" + ], + "authors": [ + { + "name": "Peter Kruithof", + "email": "pkruithof@gmail.com", + "homepage": "http://pkruithof.tumblr.com/" + }, + { + "name": "Milian Wolff", + "email": "mail@milianw.de", + "homepage": "http://milianw.de" + }, + { + "name": "Thomas Zilliox", + "email": "hello@tzi.fr", + "homepage": "http://tzi.fr" + } + ], + "description": "The HTML to Markdown converter for PHP ", + "homepage": "https://github.com/elephant418/Markdownify", + "keywords": [ + "markdown", + "markdownify" + ] } ] diff --git a/vendor/pixel418/markdownify/CHANGELOG.md b/vendor/pixel418/markdownify/CHANGELOG.md new file mode 100644 index 000000000..a0900fdf5 --- /dev/null +++ b/vendor/pixel418/markdownify/CHANGELOG.md @@ -0,0 +1,76 @@ +CHANGELOG +============== + + +21/09/2016 v2.2.1 +-------------- + + * Fix: Moving trailing whitespace from inline elements outside of the element + * Feature: Use PSR-4 + * Feature: PHP 7.0 support in continuous integration + * Doc: Update of the README + + +07/09/2016 v2.2.0 +-------------- + + * Fix: Reset state between each parsing + + +19/02/2016 v2.1.11 +-------------- + + * Fix: Empty table cell conversion + + +10/02/2016 v2.1.10 +-------------- + + * Fix: Handle nested table. + + +01/04/2015 v2.1.9 +-------------- + + * Fix: Handle HTML breaks & spaces in a less destructive way. + + +26/03/2015 v2.1.8 +-------------- + + * Fix: Use alternative italic character + * Fix: Handle HTML breaks inside another tag + * Fix: Handle HTML spaces around tags + + +07/11/2014 v2.1.7 +-------------- + + * Change composer name to "elephant418/markdownify" + + +14/07/2014 v2.1.6 +-------------- + + * Fix: Simulate a paragraph for inline text preceding block element + * Fix: Nested lists + * Fix: setKeepHTML method + * Feature: PHP 5.5 & 5.6 support in continuous integration + + +16/03/2014 v2.1.5 +-------------- + +Add display settings + + * Test: Add tests for footnotes after every paragraph or not + * Feature: Allow to display link reference in paragraph, without footnotes + + +27/02/2014 v2.1.4 +-------------- + +Improve how ConverterExtra handle id & class attributes: + + * Feature: Allow id & class attributes on links + * Feature: Allow class attributes on headings
\ No newline at end of file diff --git a/library/markdownify/LICENSE_LGPL.txt b/vendor/pixel418/markdownify/LICENSE index 5ab7695ab..5ab7695ab 100644 --- a/library/markdownify/LICENSE_LGPL.txt +++ b/vendor/pixel418/markdownify/LICENSE diff --git a/vendor/pixel418/markdownify/README.md b/vendor/pixel418/markdownify/README.md new file mode 100644 index 000000000..8855b0d05 --- /dev/null +++ b/vendor/pixel418/markdownify/README.md @@ -0,0 +1,63 @@ +# Markdownify + +[![Build Status](https://travis-ci.org/Elephant418/Markdownify.png?branch=master)](https://travis-ci.org/Elephant418/Markdownify?branch=master) +[![Total Downloads](https://poser.pugx.org/pixel418/markdownify/downloads)](https://packagist.org/packages/pixel418/markdownify) +[![License LGPL](https://poser.pugx.org/pixel418/markdownify/license)](https://opensource.org/licenses/lgpl-2.1.php) + +The HTML to Markdown converter for PHP + +[Code example](#code-example) | [How to Install](#how-to-install) | [How to Contribute](#how-to-contribute) | [Author & Community](#author--community) + + + +Code example +-------- + +### Markdown + +```php +$converter = new Markdownify\Converter; +$converter->parseString('<h1>Heading</h1>'); +// Returns: # Heading +``` + +### Markdown Extra [as defined by @michelf](http://michelf.ca/projects/php-markdown/extra/) + +```php +$converter = new Markdownify\ConverterExtra; +$converter->parseString('<h1 id="md">Heading</h1>'); +// Returns: # Heading {#md} +``` + + + +How to Install +-------- + +This library package requires `PHP 5.3` or later.<br> +Install [Composer](http://getcomposer.org/doc/01-basic-usage.md#installation) and run the following command to get the latest version: + +```sh +composer require pixel418/markdownify +``` + + + +How to Contribute +-------- + +1. Fork the Markdownify repository +2. Create a new branch for each feature or improvement +3. Send a pull request from each feature branch to the **v2.x** branch + +If you don't know much about pull request, you can read [the Github article](https://help.github.com/articles/using-pull-requests) + + + +Author & Community +-------- + +Markdownify is under [LGPL License](http://opensource.org/licenses/LGPL-2.1)<br> +It was created by [Milian Wolff](http://milianw.de)<br> +It was converted to a Symfony Bundle by [Peter Kruithof](https://github.com/pkruithof)<br> +It is maintained by [Thomas ZILLIOX](http://tzi.fr) diff --git a/vendor/pixel418/markdownify/composer.json b/vendor/pixel418/markdownify/composer.json new file mode 100644 index 000000000..71d9f3565 --- /dev/null +++ b/vendor/pixel418/markdownify/composer.json @@ -0,0 +1,38 @@ +{ + "name": "pixel418/markdownify", + "type": "lib", + "description": "The HTML to Markdown converter for PHP ", + "keywords": ["markdown", "markdownify"], + "license": "LGPL", + "homepage": "https://github.com/elephant418/Markdownify", + "authors": [ + { + "name": "Milian Wolff", + "email": "mail@milianw.de", + "homepage": "http://milianw.de" + + }, + { + "name": "Peter Kruithof", + "email": "pkruithof@gmail.com", + "homepage": "http://pkruithof.tumblr.com/" + }, + { + "name": "Thomas Zilliox", + "email": "hello@tzi.fr", + "homepage": "http://tzi.fr" + } + ], + "require": { + "php": ">=5.3.0" + }, + "require-dev": { + "phpunit/phpunit": "^4.8" + }, + "autoload": { + "psr-4": { + "Markdownify\\": "src", + "Test\\Markdownify\\": "test" + } + } +} diff --git a/vendor/pixel418/markdownify/src/Converter.php b/vendor/pixel418/markdownify/src/Converter.php new file mode 100644 index 000000000..77c62dc7e --- /dev/null +++ b/vendor/pixel418/markdownify/src/Converter.php @@ -0,0 +1,1400 @@ +<?php + +/* This file is part of the Markdownify project, which is under LGPL license */ + +namespace Markdownify; + +/** + * default configuration + */ +define('MDFY_BODYWIDTH', false); +define('MDFY_KEEPHTML', true); + +/** + * HTML to Markdown converter class + */ +class Converter +{ + /** + * html parser object + * + * @var parseHTML + */ + protected $parser; + + /** + * markdown output + * + * @var string + */ + protected $output; + + /** + * stack with tags which where not converted to html + * + * @var array<string> + */ + protected $notConverted = array(); + + /** + * skip conversion to markdown + * + * @var bool + */ + protected $skipConversion = false; + + /* options */ + + /** + * keep html tags which cannot be converted to markdown + * + * @var bool + */ + protected $keepHTML = false; + + /** + * wrap output, set to 0 to skip wrapping + * + * @var int + */ + protected $bodyWidth = 0; + + /** + * minimum body width + * + * @var int + */ + protected $minBodyWidth = 25; + + /** + * position where the link reference will be displayed + * + * + * @var int + */ + protected $linkPosition; + const LINK_AFTER_CONTENT = 0; + const LINK_AFTER_PARAGRAPH = 1; + const LINK_IN_PARAGRAPH = 2; + + /** + * stores current buffers + * + * @var array<string> + */ + protected $buffer = array(); + + /** + * stores current buffers + * + * @var array<string> + */ + protected $footnotes = array(); + + /** + * tags with elements which can be handled by markdown + * + * @var array<string> + */ + protected $isMarkdownable = array( + 'p' => array(), + 'ul' => array(), + 'ol' => array(), + 'li' => array(), + 'br' => array(), + 'blockquote' => array(), + 'code' => array(), + 'pre' => array(), + 'a' => array( + 'href' => 'required', + 'title' => 'optional', + ), + 'strong' => array(), + 'b' => array(), + 'em' => array(), + 'i' => array(), + 'img' => array( + 'src' => 'required', + 'alt' => 'optional', + 'title' => 'optional', + ), + 'h1' => array(), + 'h2' => array(), + 'h3' => array(), + 'h4' => array(), + 'h5' => array(), + 'h6' => array(), + 'hr' => array(), + ); + + /** + * html tags to be ignored (contents will be parsed) + * + * @var array<string> + */ + protected $ignore = array( + 'html', + 'body', + ); + + /** + * html tags to be dropped (contents will not be parsed!) + * + * @var array<string> + */ + protected $drop = array( + 'script', + 'head', + 'style', + 'form', + 'area', + 'object', + 'param', + 'iframe', + ); + + /** + * html block tags that allow inline & block children + * + * @var array<string> + */ + protected $allowMixedChildren = array( + 'li' + ); + + /** + * Markdown indents which could be wrapped + * @note: use strings in regex format + * + * @var array<string> + */ + protected $wrappableIndents = array( + '\* ', // ul + '\d. ', // ol + '\d\d. ', // ol + '> ', // blockquote + '', // p + ); + + /** + * list of chars which have to be escaped in normal text + * @note: use strings in regex format + * + * @var array + * + * TODO: what's with block chars / sequences at the beginning of a block? + */ + protected $escapeInText = array( + '\*\*([^*]+)\*\*' => '\*\*$1\*\*', // strong + '\*([^*]+)\*' => '\*$1\*', // em + '__(?! |_)(.+)(?!<_| )__' => '\_\_$1\_\_', // strong + '_(?! |_)(.+)(?!<_| )_' => '\_$1\_', // em + '([-*_])([ ]{0,2}\1){2,}' => '\\\\$0', // hr + '`' => '\`', // code + '\[(.+)\](\s*\()' => '\[$1\]$2', // links: [text] (url) => [text\] (url) + '\[(.+)\](\s*)\[(.*)\]' => '\[$1\]$2\[$3\]', // links: [text][id] => [text\][id\] + '^#(#{0,5}) ' => '\#$1 ', // header + ); + + /** + * wether last processed node was a block tag or not + * + * @var bool + */ + protected $lastWasBlockTag = false; + + /** + * name of last closed tag + * + * @var string + */ + protected $lastClosedTag = ''; + + /** + * number of line breaks before next inline output + */ + protected $lineBreaks = 0; + + /** + * node stack, e.g. for <a> and <abbr> tags + * + * @var array<array> + */ + protected $stack = array(); + + /** + * current indentation + * + * @var string + */ + protected $indent = ''; + + /** + * constructor, set options, setup parser + * + * @param int $linkPosition define the position of links + * @param int $bodyWidth whether or not to wrap the output to the given width + * defaults to false + * @param bool $keepHTML whether to keep non markdownable HTML or to discard it + * defaults to true (HTML will be kept) + * @return void + */ + public function __construct($linkPosition = self::LINK_AFTER_CONTENT, $bodyWidth = MDFY_BODYWIDTH, $keepHTML = MDFY_KEEPHTML) + { + $this->linkPosition = $linkPosition; + $this->keepHTML = $keepHTML; + + if ($bodyWidth > $this->minBodyWidth) { + $this->bodyWidth = intval($bodyWidth); + } else { + $this->bodyWidth = false; + } + + $this->parser = new Parser; + $this->parser->noTagsInCode = true; + + // we don't have to do this every time + $search = array(); + $replace = array(); + foreach ($this->escapeInText as $s => $r) { + array_push($search, '@(?<!\\\)' . $s . '@U'); + array_push($replace, $r); + } + $this->escapeInText = array( + 'search' => $search, + 'replace' => $replace + ); + } + + /** + * parse a HTML string + * + * @param string $html + * @return string markdown formatted + */ + public function parseString($html) + { + $this->resetState(); + + $this->parser->html = $html; + $this->parse(); + + return $this->output; + } + + /** + * set the position where the link reference will be displayed + * + * @param int $linkPosition + * @return void + */ + public function setLinkPosition($linkPosition) + { + $this->linkPosition = $linkPosition; + } + + /** + * set keep HTML tags which cannot be converted to markdown + * + * @param bool $linkPosition + * @return void + */ + public function setKeepHTML($keepHTML) + { + $this->keepHTML = $keepHTML; + } + + /** + * iterate through the nodes and decide what we + * shall do with the current node + * + * @param void + * @return void + */ + protected function parse() + { + $this->output = ''; + // drop tags + $this->parser->html = preg_replace('#<(' . implode('|', $this->drop) . ')[^>]*>.*</\\1>#sU', '', $this->parser->html); + while ($this->parser->nextNode()) { + switch ($this->parser->nodeType) { + case 'doctype': + break; + case 'pi': + case 'comment': + if ($this->keepHTML) { + $this->flushLinebreaks(); + $this->out($this->parser->node); + $this->setLineBreaks(2); + } + // else drop + break; + case 'text': + $this->handleText(); + break; + case 'tag': + if (in_array($this->parser->tagName, $this->ignore)) { + break; + } + // If the previous tag was not a block element, we simulate a paragraph tag + if ($this->parser->isBlockElement && $this->parser->isNextToInlineContext && !in_array($this->parent(), $this->allowMixedChildren)) { + $this->setLineBreaks(2); + } + if ($this->parser->isStartTag) { + $this->flushLinebreaks(); + } + if ($this->skipConversion) { + $this->isMarkdownable(); // update notConverted + $this->handleTagToText(); + continue; + } + + // block elements + if (!$this->parser->keepWhitespace && $this->parser->isBlockElement) { + $this->fixBlockElementSpacing(); + } + + // inline elements + if (!$this->parser->keepWhitespace && $this->parser->isInlineContext) { + $this->fixInlineElementSpacing(); + } + + if ($this->isMarkdownable()) { + if ($this->parser->isBlockElement && $this->parser->isStartTag && !$this->lastWasBlockTag && !empty($this->output)) { + if (!empty($this->buffer)) { + $str =& $this->buffer[count($this->buffer) - 1]; + } else { + $str =& $this->output; + } + if (substr($str, -strlen($this->indent) - 1) != "\n" . $this->indent) { + $str .= "\n" . $this->indent; + } + } + $func = 'handleTag_' . $this->parser->tagName; + $this->$func(); + if ($this->linkPosition == self::LINK_AFTER_PARAGRAPH && $this->parser->isBlockElement && !$this->parser->isStartTag && empty($this->parser->openTags)) { + $this->flushFootnotes(); + } + if (!$this->parser->isStartTag) { + $this->lastClosedTag = $this->parser->tagName; + } + } else { + $this->handleTagToText(); + $this->lastClosedTag = ''; + } + break; + default: + trigger_error('invalid node type', E_USER_ERROR); + break; + } + $this->lastWasBlockTag = $this->parser->nodeType == 'tag' && $this->parser->isStartTag && $this->parser->isBlockElement; + } + if (!empty($this->buffer)) { + // trigger_error('buffer was not flushed, this is a bug. please report!', E_USER_WARNING); + while (!empty($this->buffer)) { + $this->out($this->unbuffer()); + } + } + // cleanup + $this->output = rtrim(str_replace('&', '&', str_replace('<', '<', str_replace('>', '>', $this->output)))); + // end parsing, flush stacked tags + $this->flushFootnotes(); + $this->stack = array(); + } + + /** + * check if current tag can be converted to Markdown + * + * @param void + * @return bool + */ + protected function isMarkdownable() + { + if (!isset($this->isMarkdownable[$this->parser->tagName])) { + // simply not markdownable + + return false; + } + if ($this->parser->isStartTag) { + $return = true; + if ($this->keepHTML) { + $diff = array_diff(array_keys($this->parser->tagAttributes), array_keys($this->isMarkdownable[$this->parser->tagName])); + if (!empty($diff)) { + // non markdownable attributes given + $return = false; + } + } + if ($return) { + foreach ($this->isMarkdownable[$this->parser->tagName] as $attr => $type) { + if ($type == 'required' && !isset($this->parser->tagAttributes[$attr])) { + // required markdown attribute not given + $return = false; + break; + } + } + } + if (!$return) { + array_push($this->notConverted, $this->parser->tagName . '::' . implode('/', $this->parser->openTags)); + } + + return $return; + } else { + if (!empty($this->notConverted) && end($this->notConverted) === $this->parser->tagName . '::' . implode('/', $this->parser->openTags)) { + array_pop($this->notConverted); + + return false; + } + + return true; + } + } + + /** + * output footnotes + * + * @param void + * @return void + */ + protected function flushFootnotes() + { + $out = false; + foreach ($this->footnotes as $k => $tag) { + if (!isset($tag['unstacked'])) { + if (!$out) { + $out = true; + $this->out("\n\n", true); + } else { + $this->out("\n", true); + } + $this->out(' [' . $tag['linkID'] . ']: ' . $this->getLinkReference($tag), true); + $tag['unstacked'] = true; + $this->footnotes[$k] = $tag; + } + } + } + + /** + * return formated link reference + * + * @param array $tag + * @return string link reference + */ + protected function getLinkReference($tag) + { + return $tag['href'] . (isset($tag['title']) ? ' "' . $tag['title'] . '"' : ''); + } + + /** + * flush enqued linebreaks + * + * @param void + * @return void + */ + protected function flushLinebreaks() + { + if ($this->lineBreaks && !empty($this->output)) { + $this->out(str_repeat("\n" . $this->indent, $this->lineBreaks), true); + } + $this->lineBreaks = 0; + } + + /** + * handle non Markdownable tags + * + * @param void + * @return void + */ + protected function handleTagToText() + { + if (!$this->keepHTML) { + if (!$this->parser->isStartTag && $this->parser->isBlockElement) { + $this->setLineBreaks(2); + } + } else { + // dont convert to markdown inside this tag + /** TODO: markdown extra **/ + if (!$this->parser->isEmptyTag) { + if ($this->parser->isStartTag) { + if (!$this->skipConversion) { + $this->skipConversion = $this->parser->tagName . '::' . implode('/', $this->parser->openTags); + } + } else { + if ($this->skipConversion == $this->parser->tagName . '::' . implode('/', $this->parser->openTags)) { + $this->skipConversion = false; + } + } + } + + if ($this->parser->isBlockElement) { + if ($this->parser->isStartTag) { + // looks like ins or del are block elements now + if (in_array($this->parent(), array('ins', 'del'))) { + $this->out("\n", true); + $this->indent(' '); + } + // don't indent inside <pre> tags + if ($this->parser->tagName == 'pre') { + $this->out($this->parser->node); + static $indent; + $indent = $this->indent; + $this->indent = ''; + } else { + $this->out($this->parser->node . "\n" . $this->indent); + if (!$this->parser->isEmptyTag) { + $this->indent(' '); + } else { + $this->setLineBreaks(1); + } + $this->parser->html = ltrim($this->parser->html); + } + } else { + if (!$this->parser->keepWhitespace) { + $this->output = rtrim($this->output); + } + if ($this->parser->tagName != 'pre') { + $this->indent(' '); + $this->out("\n" . $this->indent . $this->parser->node); + } else { + // reset indentation + $this->out($this->parser->node); + static $indent; + $this->indent = $indent; + } + + if (in_array($this->parent(), array('ins', 'del'))) { + // ins or del was block element + $this->out("\n"); + $this->indent(' '); + } + if ($this->parser->tagName == 'li') { + $this->setLineBreaks(1); + } else { + $this->setLineBreaks(2); + } + } + } else { + $this->out($this->parser->node); + } + if (in_array($this->parser->tagName, array('code', 'pre'))) { + if ($this->parser->isStartTag) { + $this->buffer(); + } else { + // add stuff so cleanup just reverses this + $this->out(str_replace('<', '&lt;', str_replace('>', '&gt;', $this->unbuffer()))); + } + } + } + } + + /** + * handle plain text + * + * @param void + * @return void + */ + protected function handleText() + { + if ($this->hasParent('pre') && strpos($this->parser->node, "\n") !== false) { + $this->parser->node = str_replace("\n", "\n" . $this->indent, $this->parser->node); + } + if (!$this->hasParent('code') && !$this->hasParent('pre')) { + // entity decode + $this->parser->node = $this->decode($this->parser->node); + if (!$this->skipConversion) { + // escape some chars in normal Text + $this->parser->node = preg_replace($this->escapeInText['search'], $this->escapeInText['replace'], $this->parser->node); + } + } else { + $this->parser->node = str_replace(array('"', '&apos'), array('"', '\''), $this->parser->node); + } + $this->out($this->parser->node); + $this->lastClosedTag = ''; + } + + /** + * handle <em> and <i> tags + * + * @param void + * @return void + */ + protected function handleTag_em() + { + $this->out('_', true); + } + + protected function handleTag_i() + { + $this->handleTag_em(); + } + + /** + * handle <strong> and <b> tags + * + * @param void + * @return void + */ + protected function handleTag_strong() + { + $this->out('**', true); + } + + protected function handleTag_b() + { + $this->handleTag_strong(); + } + + /** + * handle <h1> tags + * + * @param void + * @return void + */ + protected function handleTag_h1() + { + $this->handleHeader(1); + } + + /** + * handle <h2> tags + * + * @param void + * @return void + */ + protected function handleTag_h2() + { + $this->handleHeader(2); + } + + /** + * handle <h3> tags + * + * @param void + * @return void + */ + protected function handleTag_h3() + { + $this->handleHeader(3); + } + + /** + * handle <h4> tags + * + * @param void + * @return void + */ + protected function handleTag_h4() + { + $this->handleHeader(4); + } + + /** + * handle <h5> tags + * + * @param void + * @return void + */ + protected function handleTag_h5() + { + $this->handleHeader(5); + } + + /** + * handle <h6> tags + * + * @param void + * @return void + */ + protected function handleTag_h6() + { + $this->handleHeader(6); + } + + /** + * handle header tags (<h1> - <h6>) + * + * @param int $level 1-6 + * @return void + */ + protected function handleHeader($level) + { + if ($this->parser->isStartTag) { + $this->out(str_repeat('#', $level) . ' ', true); + } else { + $this->setLineBreaks(2); + } + } + + /** + * handle <p> tags + * + * @param void + * @return void + */ + protected function handleTag_p() + { + if (!$this->parser->isStartTag) { + $this->setLineBreaks(2); + } + } + + /** + * handle <a> tags + * + * @param void + * @return void + */ + protected function handleTag_a() + { + if ($this->parser->isStartTag) { + $this->buffer(); + $this->handleTag_a_parser(); + $this->stack(); + } else { + $tag = $this->unstack(); + $buffer = $this->unbuffer(); + $this->handleTag_a_converter($tag, $buffer); + $this->out($this->handleTag_a_converter($tag, $buffer), true); + } + } + + /** + * handle <a> tags parsing + * + * @param void + * @return void + */ + protected function handleTag_a_parser() + { + if (isset($this->parser->tagAttributes['title'])) { + $this->parser->tagAttributes['title'] = $this->decode($this->parser->tagAttributes['title']); + } else { + $this->parser->tagAttributes['title'] = null; + } + $this->parser->tagAttributes['href'] = $this->decode(trim($this->parser->tagAttributes['href'])); + } + + /** + * handle <a> tags conversion + * + * @param array $tag + * @param string $buffer + * @return string The markdownified link + */ + protected function handleTag_a_converter($tag, $buffer) + { + if (empty($tag['href']) && empty($tag['title'])) { + // empty links... testcase mania, who would possibly do anything like that?! + return '[' . $buffer . ']()'; + } + + if ($buffer == $tag['href'] && empty($tag['title'])) { + // <http://example.com> + return '<' . $buffer . '>'; + } + + $bufferDecoded = $this->decode(trim($buffer)); + if (substr($tag['href'], 0, 7) == 'mailto:' && 'mailto:' . $bufferDecoded == $tag['href']) { + if (is_null($tag['title'])) { + // <mail@example.com> + return '<' . $bufferDecoded . '>'; + } + // [mail@example.com][1] + // ... + // [1]: mailto:mail@example.com Title + $tag['href'] = 'mailto:' . $bufferDecoded; + } + + if ($this->linkPosition == self::LINK_IN_PARAGRAPH) { + return '[' . $buffer . '](' . $this->getLinkReference($tag) . ')'; + } + + // [This link][id] + foreach ($this->footnotes as $tag2) { + if ($tag2['href'] == $tag['href'] && $tag2['title'] === $tag['title']) { + $tag['linkID'] = $tag2['linkID']; + break; + } + } + if (!isset($tag['linkID'])) { + $tag['linkID'] = count($this->footnotes) + 1; + array_push($this->footnotes, $tag); + } + + return '[' . $buffer . '][' . $tag['linkID'] . ']'; + } + + /** + * handle <img /> tags + * + * @param void + * @return void + */ + protected function handleTag_img() + { + if (!$this->parser->isStartTag) { + return; // just to be sure this is really an empty tag... + } + + if (isset($this->parser->tagAttributes['title'])) { + $this->parser->tagAttributes['title'] = $this->decode($this->parser->tagAttributes['title']); + } else { + $this->parser->tagAttributes['title'] = null; + } + if (isset($this->parser->tagAttributes['alt'])) { + $this->parser->tagAttributes['alt'] = $this->decode($this->parser->tagAttributes['alt']); + } else { + $this->parser->tagAttributes['alt'] = null; + } + + if (empty($this->parser->tagAttributes['src'])) { + // support for "empty" images... dunno if this is really needed + // but there are some test cases which do that... + if (!empty($this->parser->tagAttributes['title'])) { + $this->parser->tagAttributes['title'] = ' ' . $this->parser->tagAttributes['title'] . ' '; + } + $this->out('![' . $this->parser->tagAttributes['alt'] . '](' . $this->parser->tagAttributes['title'] . ')', true); + + return; + } else { + $this->parser->tagAttributes['src'] = $this->decode($this->parser->tagAttributes['src']); + } + + $out = '![' . $this->parser->tagAttributes['alt'] . ']'; + if ($this->linkPosition == self::LINK_IN_PARAGRAPH) { + $out .= '(' . $this->parser->tagAttributes['src']; + if ($this->parser->tagAttributes['title']) { + $out .= ' "' . $this->parser->tagAttributes['title'] . '"'; + } + $out .= ')'; + $this->out($out, true); + return; + } + + // ![This image][id] + $link_id = false; + if (!empty($this->footnotes)) { + foreach ($this->footnotes as $tag) { + if ($tag['href'] == $this->parser->tagAttributes['src'] + && $tag['title'] === $this->parser->tagAttributes['title'] + ) { + $link_id = $tag['linkID']; + break; + } + } + } + if (!$link_id) { + $link_id = count($this->footnotes) + 1; + $tag = array( + 'href' => $this->parser->tagAttributes['src'], + 'linkID' => $link_id, + 'title' => $this->parser->tagAttributes['title'] + ); + array_push($this->footnotes, $tag); + } + $out .= '[' . $link_id . ']'; + + $this->out($out, true); + } + + /** + * handle <code> tags + * + * @param void + * @return void + */ + protected function handleTag_code() + { + if ($this->hasParent('pre')) { + // ignore code blocks inside <pre> + + return; + } + if ($this->parser->isStartTag) { + $this->buffer(); + } else { + $buffer = $this->unbuffer(); + // use as many backticks as needed + preg_match_all('#`+#', $buffer, $matches); + if (!empty($matches[0])) { + rsort($matches[0]); + + $ticks = '`'; + while (true) { + if (!in_array($ticks, $matches[0])) { + break; + } + $ticks .= '`'; + } + } else { + $ticks = '`'; + } + if ($buffer[0] == '`' || substr($buffer, -1) == '`') { + $buffer = ' ' . $buffer . ' '; + } + $this->out($ticks . $buffer . $ticks, true); + } + } + + /** + * handle <pre> tags + * + * @param void + * @return void + */ + protected function handleTag_pre() + { + if ($this->keepHTML && $this->parser->isStartTag) { + // check if a simple <code> follows + if (!preg_match('#^\s*<code\s*>#Us', $this->parser->html)) { + // this is no standard markdown code block + $this->handleTagToText(); + + return; + } + } + $this->indent(' '); + if (!$this->parser->isStartTag) { + $this->setLineBreaks(2); + } else { + $this->parser->html = ltrim($this->parser->html); + } + } + + /** + * handle <blockquote> tags + * + * @param void + * @return void + */ + protected function handleTag_blockquote() + { + $this->indent('> '); + } + + /** + * handle <ul> tags + * + * @param void + * @return void + */ + protected function handleTag_ul() + { + if ($this->parser->isStartTag) { + $this->stack(); + if (!$this->keepHTML && $this->lastClosedTag == $this->parser->tagName) { + $this->out("\n" . $this->indent . '<!-- -->' . "\n" . $this->indent . "\n" . $this->indent); + } + } else { + $this->unstack(); + if ($this->parent() != 'li' || preg_match('#^\s*(</li\s*>\s*<li\s*>\s*)?<(p|blockquote)\s*>#sU', $this->parser->html)) { + // dont make Markdown add unneeded paragraphs + $this->setLineBreaks(2); + } + } + } + + /** + * handle <ul> tags + * + * @param void + * @return void + */ + protected function handleTag_ol() + { + // same as above + $this->parser->tagAttributes['num'] = 0; + $this->handleTag_ul(); + } + + /** + * handle <li> tags + * + * @param void + * @return void + */ + protected function handleTag_li() + { + if ($this->parent() == 'ol') { + $parent =& $this->getStacked('ol'); + if ($this->parser->isStartTag) { + $parent['num']++; + $this->out(str_repeat(' ', 3 - strlen($parent['num'])) . $parent['num'] . '. ', true); + } + } else { + if ($this->parser->isStartTag) { + $this->out(' * ', true); + } + } + $this->indent(' ', false); + if (!$this->parser->isStartTag) { + $this->setLineBreaks(1); + } + } + + /** + * handle <hr /> tags + * + * @param void + * @return void + */ + protected function handleTag_hr() + { + if (!$this->parser->isStartTag) { + return; // just to be sure this really is an empty tag + } + $this->out('* * *', true); + $this->setLineBreaks(2); + } + + /** + * handle <br /> tags + * + * @param void + * @return void + */ + protected function handleTag_br() + { + $this->out(" \n" . $this->indent, true); + $this->parser->html = ltrim($this->parser->html); + } + + /** + * add current node to the stack + * this only stores the attributes + * + * @param void + * @return void + */ + protected function stack() + { + if (!isset($this->stack[$this->parser->tagName])) { + $this->stack[$this->parser->tagName] = array(); + } + array_push($this->stack[$this->parser->tagName], $this->parser->tagAttributes); + } + + /** + * remove current tag from stack + * + * @param void + * @return array + */ + protected function unstack() + { + if (!isset($this->stack[$this->parser->tagName]) || !is_array($this->stack[$this->parser->tagName])) { + trigger_error('Trying to unstack from empty stack. This must not happen.', E_USER_ERROR); + } + + return array_pop($this->stack[$this->parser->tagName]); + } + + /** + * get last stacked element of type $tagName + * + * @param string $tagName + * @return array + */ + protected function &getStacked($tagName) + { + // no end() so it can be referenced + return $this->stack[$tagName][count($this->stack[$tagName]) - 1]; + } + + /** + * set number of line breaks before next start tag + * + * @param int $number + * @return void + */ + protected function setLineBreaks($number) + { + if ($this->lineBreaks < $number) { + $this->lineBreaks = $number; + } + } + + /** + * buffer next parser output until unbuffer() is called + * + * @param void + * @return void + */ + protected function buffer() + { + array_push($this->buffer, ''); + } + + /** + * end current buffer and return buffered output + * + * @param void + * @return string + */ + protected function unbuffer() + { + return array_pop($this->buffer); + } + + /** + * append string to the correct var, either + * directly to $this->output or to the current + * buffers + * + * @param string $put + * @param boolean $nowrap + * @return void + */ + protected function out($put, $nowrap = false) + { + if (empty($put)) { + return; + } + if (!empty($this->buffer)) { + $this->buffer[count($this->buffer) - 1] .= $put; + } else { + if ($this->bodyWidth && !$this->parser->keepWhitespace) { // wrap lines + // get last line + $pos = strrpos($this->output, "\n"); + if ($pos === false) { + $line = $this->output; + } else { + $line = substr($this->output, $pos); + } + + if ($nowrap) { + if ($put[0] != "\n" && $this->strlen($line) + $this->strlen($put) > $this->bodyWidth) { + $this->output .= "\n" . $this->indent . $put; + } else { + $this->output .= $put; + } + + return; + } else { + $put .= "\n"; // make sure we get all lines in the while below + $lineLen = $this->strlen($line); + while ($pos = strpos($put, "\n")) { + $putLine = substr($put, 0, $pos + 1); + $put = substr($put, $pos + 1); + $putLen = $this->strlen($putLine); + if ($lineLen + $putLen < $this->bodyWidth) { + $this->output .= $putLine; + $lineLen = $putLen; + } else { + $split = preg_split('#^(.{0,' . ($this->bodyWidth - $lineLen) . '})\b#', $putLine, 2, PREG_SPLIT_OFFSET_CAPTURE | PREG_SPLIT_DELIM_CAPTURE); + $this->output .= rtrim($split[1][0]) . "\n" . $this->indent . $this->wordwrap(ltrim($split[2][0]), $this->bodyWidth, "\n" . $this->indent, false); + } + } + $this->output = substr($this->output, 0, -1); + + return; + } + } else { + $this->output .= $put; + } + } + } + + /** + * indent next output (start tag) or unindent (end tag) + * + * @param string $str indentation + * @param bool $output add indendation to output + * @return void + */ + protected function indent($str, $output = true) + { + if ($this->parser->isStartTag) { + $this->indent .= $str; + if ($output) { + $this->out($str, true); + } + } else { + $this->indent = substr($this->indent, 0, -strlen($str)); + } + } + + /** + * decode email addresses + * + * @author derernst@gmx.ch <http://www.php.net/manual/en/function.html-entity-decode.php#68536> + * @author Milian Wolff <http://milianw.de> + */ + protected function decode($text, $quote_style = ENT_QUOTES) + { + return htmlspecialchars_decode($text, $quote_style); + } + + /** + * callback for decode() which converts a hexadecimal entity to UTF-8 + * + * @param array $matches + * @return string UTF-8 encoded + */ + protected function _decode_hex($matches) + { + return $this->unichr(hexdec($matches[1])); + } + + /** + * callback for decode() which converts a numerical entity to UTF-8 + * + * @param array $matches + * @return string UTF-8 encoded + */ + protected function _decode_numeric($matches) + { + return $this->unichr($matches[1]); + } + + /** + * UTF-8 chr() which supports numeric entities + * + * @author grey - greywyvern - com <http://www.php.net/manual/en/function.chr.php#55978> + * @param array $matches + * @return string UTF-8 encoded + */ + protected function unichr($dec) + { + if ($dec < 128) { + $utf = chr($dec); + } elseif ($dec < 2048) { + $utf = chr(192 + (($dec - ($dec % 64)) / 64)); + $utf .= chr(128 + ($dec % 64)); + } else { + $utf = chr(224 + (($dec - ($dec % 4096)) / 4096)); + $utf .= chr(128 + ((($dec % 4096) - ($dec % 64)) / 64)); + $utf .= chr(128 + ($dec % 64)); + } + + return $utf; + } + + /** + * UTF-8 strlen() + * + * @param string $str + * @return int + * + * @author dtorop 932 at hotmail dot com <http://www.php.net/manual/en/function.strlen.php#37975> + * @author Milian Wolff <http://milianw.de> + */ + protected function strlen($str) + { + if (function_exists('mb_strlen')) { + return mb_strlen($str, 'UTF-8'); + } else { + return preg_match_all('/[\x00-\x7F\xC0-\xFD]/', $str, $var_empty); + } + } + + /** + * wordwrap for utf8 encoded strings + * + * @param string $str + * @param integer $len + * @param string $what + * @return string + */ + protected function wordwrap($str, $width, $break, $cut = false) + { + if (!$cut) { + $regexp = '#^(?:[\x00-\x7F]|[\xC0-\xFF][\x80-\xBF]+){1,' . $width . '}\b#'; + } else { + $regexp = '#^(?:[\x00-\x7F]|[\xC0-\xFF][\x80-\xBF]+){' . $width . '}#'; + } + $return = ''; + while (preg_match($regexp, $str, $matches)) { + $string = $matches[0]; + $str = ltrim(substr($str, strlen($string))); + if (!$cut && isset($str[0]) && in_array($str[0], array('.', '!', ';', ':', '?', ','))) { + $string .= $str[0]; + $str = ltrim(substr($str, 1)); + } + $return .= $string . $break; + } + + return $return . ltrim($str); + } + + /** + * check if current node has a $tagName as parent (somewhere, not only the direct parent) + * + * @param string $tagName + * @return bool + */ + protected function hasParent($tagName) + { + return in_array($tagName, $this->parser->openTags); + } + + /** + * get tagName of direct parent tag + * + * @param void + * @return string $tagName + */ + protected function parent() + { + return end($this->parser->openTags); + } + + /** + * Trims whitespace in block-level elements, on the left side. + */ + protected function fixBlockElementSpacing() + { + if ($this->parser->isStartTag) { + $this->parser->html = ltrim($this->parser->html); + } + } + + /** + * Moves leading/trailing whitespace from inline elements outside of the + * element. This is to fix cases like `<strong> Text</strong>`, which if + * converted to `** strong**` would be incorrect Markdown. + * + * Examples: + * + * * leading: `<strong> Text</strong>` becomes ` <strong>Text</strong>` + * * trailing: `<strong>Text </strong>` becomes `<strong>Text</strong> ` + */ + protected function fixInlineElementSpacing() + { + if ($this->parser->isStartTag) { + // move spaces after the start element to before the element + if (preg_match('~^(\s+)~', $this->parser->html, $matches)) { + $this->out($matches[1]); + $this->parser->html = ltrim($this->parser->html, " \t\0\x0B"); + } + } else { + if (!empty($this->buffer)) { + $str =& $this->buffer[count($this->buffer) - 1]; + } else { + $str =& $this->output; + } + + // move spaces before the end element to after the element + if (preg_match('~(\s+)$~', $str, $matches)) { + $str = rtrim($str, " \t\0\x0B"); + $this->parser->html = $matches[1] . $this->parser->html; + } + } + } + + /** + * Resetting the state forces the instance to behave as a fresh instance. + * Ideal for running within a loop where you want to maintain a single instance. + */ + protected function resetState() + { + $this->notConverted = array(); + $this->skipConversion = false; + $this->buffer = array(); + $this->indent = ''; + $this->stack = array(); + $this->lineBreaks = 0; + $this->lastClosedTag = ''; + $this->lastWasBlockTag = false; + $this->footnotes = array(); + } +} diff --git a/vendor/pixel418/markdownify/src/ConverterExtra.php b/vendor/pixel418/markdownify/src/ConverterExtra.php new file mode 100644 index 000000000..733955448 --- /dev/null +++ b/vendor/pixel418/markdownify/src/ConverterExtra.php @@ -0,0 +1,573 @@ +<?php + +/* This file is part of the Markdownify project, which is under LGPL license */ + +namespace Markdownify; + +class ConverterExtra extends Converter +{ + + /** + * table data, including rows with content and the maximum width of each col + * + * @var array + */ + protected $table = array(); + + /** + * current col + * + * @var int + */ + protected $col = -1; + + /** + * current row + * + * @var int + */ + protected $row = 0; + + /** + * constructor, see Markdownify::Markdownify() for more information + */ + public function __construct($linksAfterEachParagraph = self::LINK_AFTER_CONTENT, $bodyWidth = MDFY_BODYWIDTH, $keepHTML = MDFY_KEEPHTML) + { + parent::__construct($linksAfterEachParagraph, $bodyWidth, $keepHTML); + + // new markdownable tags & attributes + // header ids: # foo {bar} + $this->isMarkdownable['h1']['id'] = 'optional'; + $this->isMarkdownable['h1']['class'] = 'optional'; + $this->isMarkdownable['h2']['id'] = 'optional'; + $this->isMarkdownable['h2']['class'] = 'optional'; + $this->isMarkdownable['h3']['id'] = 'optional'; + $this->isMarkdownable['h3']['class'] = 'optional'; + $this->isMarkdownable['h4']['id'] = 'optional'; + $this->isMarkdownable['h4']['class'] = 'optional'; + $this->isMarkdownable['h5']['id'] = 'optional'; + $this->isMarkdownable['h5']['class'] = 'optional'; + $this->isMarkdownable['h6']['id'] = 'optional'; + $this->isMarkdownable['h6']['class'] = 'optional'; + // tables + $this->isMarkdownable['table'] = array(); + $this->isMarkdownable['th'] = array( + 'align' => 'optional', + ); + $this->isMarkdownable['td'] = array( + 'align' => 'optional', + ); + $this->isMarkdownable['tr'] = array(); + array_push($this->ignore, 'thead'); + array_push($this->ignore, 'tbody'); + array_push($this->ignore, 'tfoot'); + // definition lists + $this->isMarkdownable['dl'] = array(); + $this->isMarkdownable['dd'] = array(); + $this->isMarkdownable['dt'] = array(); + // link class + $this->isMarkdownable['a']['id'] = 'optional'; + $this->isMarkdownable['a']['class'] = 'optional'; + // footnotes + $this->isMarkdownable['fnref'] = array( + 'target' => 'required', + ); + $this->isMarkdownable['footnotes'] = array(); + $this->isMarkdownable['fn'] = array( + 'name' => 'required', + ); + $this->parser->blockElements['fnref'] = false; + $this->parser->blockElements['fn'] = true; + $this->parser->blockElements['footnotes'] = true; + // abbr + $this->isMarkdownable['abbr'] = array( + 'title' => 'required', + ); + // build RegEx lookahead to decide wether table can pe parsed or not + $inlineTags = array_keys($this->parser->blockElements, false); + $colContents = '(?:[^<]|<(?:' . implode('|', $inlineTags) . '|[^a-z]))*'; + $this->tableLookaheadHeader = '{ + ^\s*(?:<thead\s*>)?\s* # open optional thead + <tr\s*>\s*(?: # start required row with headers + <th(?:\s+align=("|\')(?:left|center|right)\1)?\s*> # header with optional align + \s*' . $colContents . '\s* # contents + </th>\s* # close header + )+</tr> # close row with headers + \s*(?:</thead>)? # close optional thead + }sxi'; + $this->tdSubstitute = '\s*' . $colContents . '\s* # contents + </td>\s*'; + $this->tableLookaheadBody = '{ + \s*(?:<tbody\s*>)?\s* # open optional tbody + (?:<tr\s*>\s* # start row + %s # cols to be substituted + </tr>)+ # close row + \s*(?:</tbody>)? # close optional tbody + \s*</table> # close table + }sxi'; + } + + /** + * handle header tags (<h1> - <h6>) + * + * @param int $level 1-6 + * @return void + */ + protected function handleHeader($level) + { + if ($this->parser->isStartTag) { + $this->parser->tagAttributes['cssSelector'] = $this->getCurrentCssSelector(); + $this->stack(); + } else { + $tag = $this->unstack(); + if (!empty($tag['cssSelector'])) { + // {#id.class} + $this->out(' {' . $tag['cssSelector'] . '}'); + } + } + parent::handleHeader($level); + } + + /** + * handle <a> tags parsing + * + * @param void + * @return void + */ + protected function handleTag_a_parser() + { + parent::handleTag_a_parser(); + $this->parser->tagAttributes['cssSelector'] = $this->getCurrentCssSelector(); + } + + /** + * handle <a> tags conversion + * + * @param array $tag + * @param string $buffer + * @return string The markdownified link + */ + protected function handleTag_a_converter($tag, $buffer) + { + $output = parent::handleTag_a_converter($tag, $buffer); + if (!empty($tag['cssSelector'])) { + // [This link][id]{#id.class} + $output .= '{' . $tag['cssSelector'] . '}'; + } + + return $output; + } + + /** + * handle <abbr> tags + * + * @param void + * @return void + */ + protected function handleTag_abbr() + { + if ($this->parser->isStartTag) { + $this->stack(); + $this->buffer(); + } else { + $tag = $this->unstack(); + $tag['text'] = $this->unbuffer(); + $add = true; + foreach ($this->stack['abbr'] as $stacked) { + if ($stacked['text'] == $tag['text']) { + /** TODO: differing abbr definitions, i.e. different titles for same text **/ + $add = false; + break; + } + } + $this->out($tag['text']); + if ($add) { + array_push($this->stack['abbr'], $tag); + } + } + } + + /** + * flush stacked abbr tags + * + * @param void + * @return void + */ + protected function flushStacked_abbr() + { + $out = array(); + foreach ($this->stack['abbr'] as $k => $tag) { + if (!isset($tag['unstacked'])) { + array_push($out, ' *[' . $tag['text'] . ']: ' . $tag['title']); + $tag['unstacked'] = true; + $this->stack['abbr'][$k] = $tag; + } + } + if (!empty($out)) { + $this->out("\n\n" . implode("\n", $out)); + } + } + + /** + * handle <table> tags + * + * @param void + * @return void + */ + protected function handleTag_table() + { + if ($this->parser->isStartTag) { + // check if upcoming table can be converted + if ($this->keepHTML) { + if (preg_match($this->tableLookaheadHeader, $this->parser->html, $matches)) { + // header seems good, now check body + // get align & number of cols + preg_match_all('#<th(?:\s+align=("|\')(left|right|center)\1)?\s*>#si', $matches[0], $cols); + $regEx = ''; + $i = 1; + $aligns = array(); + foreach ($cols[2] as $align) { + $align = strtolower($align); + array_push($aligns, $align); + if (empty($align)) { + $align = 'left'; // default value + } + $td = '\s+align=("|\')' . $align . '\\' . $i; + $i++; + if ($align == 'left') { + // look for empty align or left + $td = '(?:' . $td . ')?'; + } + $td = '<td' . $td . '\s*>'; + $regEx .= $td . $this->tdSubstitute; + } + $regEx = sprintf($this->tableLookaheadBody, $regEx); + if (preg_match($regEx, $this->parser->html, $matches, null, strlen($matches[0]))) { + // this is a markdownable table tag! + $this->table = array( + 'rows' => array(), + 'col_widths' => array(), + 'aligns' => $aligns, + ); + $this->row = 0; + } else { + // non markdownable table + $this->handleTagToText(); + } + } else { + // non markdownable table + $this->handleTagToText(); + } + } else { + $this->table = array( + 'rows' => array(), + 'col_widths' => array(), + 'aligns' => array(), + ); + $this->row = 0; + } + } else { + // finally build the table in Markdown Extra syntax + $separator = array(); + if (!isset($this->table['aligns'])) { + $this->table['aligns'] = array(); + } + // seperator with correct align identifiers + foreach ($this->table['aligns'] as $col => $align) { + if (!$this->keepHTML && !isset($this->table['col_widths'][$col])) { + break; + } + $left = ' '; + $right = ' '; + switch ($align) { + case 'left': + $left = ':'; + break; + case 'center': + $right = ':'; + $left = ':'; + case 'right': + $right = ':'; + break; + } + array_push($separator, $left . str_repeat('-', $this->table['col_widths'][$col]) . $right); + } + $separator = '|' . implode('|', $separator) . '|'; + + $rows = array(); + // add padding + array_walk_recursive($this->table['rows'], array(&$this, 'alignTdContent')); + $header = array_shift($this->table['rows']); + array_push($rows, '| ' . implode(' | ', $header) . ' |'); + array_push($rows, $separator); + foreach ($this->table['rows'] as $row) { + array_push($rows, '| ' . implode(' | ', $row) . ' |'); + } + $this->out(implode("\n" . $this->indent, $rows)); + $this->table = array(); + $this->setLineBreaks(2); + } + } + + /** + * properly pad content so it is aligned as whished + * should be used with array_walk_recursive on $this->table['rows'] + * + * @param string &$content + * @param int $col + * @return void + */ + protected function alignTdContent(&$content, $col) + { + if (!isset($this->table['aligns'][$col])) { + $this->table['aligns'][$col] = 'left'; + } + switch ($this->table['aligns'][$col]) { + default: + case 'left': + $content .= str_repeat(' ', $this->table['col_widths'][$col] - $this->strlen($content)); + break; + case 'right': + $content = str_repeat(' ', $this->table['col_widths'][$col] - $this->strlen($content)) . $content; + break; + case 'center': + $paddingNeeded = $this->table['col_widths'][$col] - $this->strlen($content); + $left = floor($paddingNeeded / 2); + $right = $paddingNeeded - $left; + $content = str_repeat(' ', $left) . $content . str_repeat(' ', $right); + break; + } + } + + /** + * handle <tr> tags + * + * @param void + * @return void + */ + protected function handleTag_tr() + { + if ($this->parser->isStartTag) { + $this->col = -1; + } else { + $this->row++; + } + } + + /** + * handle <td> tags + * + * @param void + * @return void + */ + protected function handleTag_td() + { + if ($this->parser->isStartTag) { + $this->col++; + if (!isset($this->table['col_widths'][$this->col])) { + $this->table['col_widths'][$this->col] = 0; + } + $this->buffer(); + } else { + $buffer = trim($this->unbuffer()); + if (!isset($this->table['col_widths'][$this->col])) { + $this->table['col_widths'][$this->col] = 0; + } + $this->table['col_widths'][$this->col] = max($this->table['col_widths'][$this->col], $this->strlen($buffer)); + $this->table['rows'][$this->row][$this->col] = $buffer; + } + } + + /** + * handle <th> tags + * + * @param void + * @return void + */ + protected function handleTag_th() + { + if (!$this->keepHTML && !isset($this->table['rows'][1]) && !isset($this->table['aligns'][$this->col + 1])) { + if (isset($this->parser->tagAttributes['align'])) { + $this->table['aligns'][$this->col + 1] = $this->parser->tagAttributes['align']; + } else { + $this->table['aligns'][$this->col + 1] = ''; + } + } + $this->handleTag_td(); + } + + /** + * handle <dl> tags + * + * @param void + * @return void + */ + protected function handleTag_dl() + { + if (!$this->parser->isStartTag) { + $this->setLineBreaks(2); + } + } + + /** + * handle <dt> tags + * + * @param void + * @return void + **/ + protected function handleTag_dt() + { + if (!$this->parser->isStartTag) { + $this->setLineBreaks(1); + } + } + + /** + * handle <dd> tags + * + * @param void + * @return void + */ + protected function handleTag_dd() + { + if ($this->parser->isStartTag) { + if (substr(ltrim($this->parser->html), 0, 3) == '<p>') { + // next comes a paragraph, so we'll need an extra line + $this->out("\n" . $this->indent); + } elseif (substr($this->output, -2) == "\n\n") { + $this->output = substr($this->output, 0, -1); + } + $this->out(': '); + $this->indent(' ', false); + } else { + // lookahead for next dt + if (substr(ltrim($this->parser->html), 0, 4) == '<dt>') { + $this->setLineBreaks(2); + } else { + $this->setLineBreaks(1); + } + $this->indent(' '); + } + } + + /** + * handle <fnref /> tags (custom footnote references, see markdownify_extra::parseString()) + * + * @param void + * @return void + */ + protected function handleTag_fnref() + { + $this->out('[^' . $this->parser->tagAttributes['target'] . ']'); + } + + /** + * handle <fn> tags (custom footnotes, see markdownify_extra::parseString() + * and markdownify_extra::_makeFootnotes()) + * + * @param void + * @return void + */ + protected function handleTag_fn() + { + if ($this->parser->isStartTag) { + $this->out('[^' . $this->parser->tagAttributes['name'] . ']:'); + $this->setLineBreaks(1); + } else { + $this->setLineBreaks(2); + } + $this->indent(' '); + } + + /** + * handle <footnotes> tag (custom footnotes, see markdownify_extra::parseString() + * and markdownify_extra::_makeFootnotes()) + * + * @param void + * @return void + */ + protected function handleTag_footnotes() + { + if (!$this->parser->isStartTag) { + $this->setLineBreaks(2); + } + } + + /** + * parse a HTML string, clean up footnotes prior + * + * @param string $HTML input + * @return string Markdown formatted output + */ + public function parseString($html) + { + /** TODO: custom markdown-extra options, e.g. titles & classes **/ + // <sup id="fnref:..."><a href"#fn..." rel="footnote">...</a></sup> + // => <fnref target="..." /> + $html = preg_replace('@<sup id="fnref:([^"]+)">\s*<a href="#fn:\1" rel="footnote">\s*\d+\s*</a>\s*</sup>@Us', '<fnref target="$1" />', $html); + // <div class="footnotes"> + // <hr /> + // <ol> + // + // <li id="fn:...">...</li> + // ... + // + // </ol> + // </div> + // => + // <footnotes> + // <fn name="...">...</fn> + // ... + // </footnotes> + $html = preg_replace_callback('#<div class="footnotes">\s*<hr />\s*<ol>\s*(.+)\s*</ol>\s*</div>#Us', array(&$this, '_makeFootnotes'), $html); + + return parent::parseString($html); + } + + /** + * replace HTML representation of footnotes with something more easily parsable + * + * @note this is a callback to be used in parseString() + * + * @param array $matches + * @return string + */ + protected function _makeFootnotes($matches) + { + // <li id="fn:1"> + // ... + // <a href="#fnref:block" rev="footnote">↩</a></p> + // </li> + // => <fn name="1">...</fn> + // remove footnote link + $fns = preg_replace('@\s*( \s*)?<a href="#fnref:[^"]+" rev="footnote"[^>]*>↩</a>\s*@s', '', $matches[1]); + // remove empty paragraph + $fns = preg_replace('@<p>\s*</p>@s', '', $fns); + // <li id="fn:1">...</li> -> <footnote nr="1">...</footnote> + $fns = str_replace('<li id="fn:', '<fn name="', $fns); + + $fns = '<footnotes>' . $fns . '</footnotes>'; + + return preg_replace('#</li>\s*(?=(?:<fn|</footnotes>))#s', '</fn>$1', $fns); + } + + /** + * handle <a> tags parsing + * + * @param void + * @return void + */ + protected function getCurrentCssSelector() + { + $cssSelector = ''; + if (isset($this->parser->tagAttributes['id'])) { + $cssSelector .= '#' . $this->decode($this->parser->tagAttributes['id']); + } + if (isset($this->parser->tagAttributes['class'])) { + $classes = explode(' ', $this->decode($this->parser->tagAttributes['class'])); + $classes = array_filter($classes); + $cssSelector .= '.' . join('.', $classes); + } + return $cssSelector; + } +} diff --git a/vendor/pixel418/markdownify/src/Parser.php b/vendor/pixel418/markdownify/src/Parser.php new file mode 100644 index 000000000..90fcdf9f8 --- /dev/null +++ b/vendor/pixel418/markdownify/src/Parser.php @@ -0,0 +1,564 @@ +<?php + +/* This file is part of the Markdownify project, which is under LGPL license */ + +namespace Markdownify; + +class Parser +{ + public static $skipWhitespace = true; + public static $a_ord; + public static $z_ord; + public static $special_ords; + + /** + * tags which are always empty (<br /> etc.) + * + * @var array<string> + */ + public $emptyTags = array( + 'br', + 'hr', + 'input', + 'img', + 'area', + 'link', + 'meta', + 'param', + ); + + /** + * tags with preformatted text + * whitespaces wont be touched in them + * + * @var array<string> + */ + public $preformattedTags = array( + 'script', + 'style', + 'pre', + 'code', + ); + + /** + * supress HTML tags inside preformatted tags (see above) + * + * @var bool + */ + public $noTagsInCode = false; + + /** + * html to be parsed + * + * @var string + */ + public $html = ''; + + /** + * node type: + * + * - tag (see isStartTag) + * - text (includes cdata) + * - comment + * - doctype + * - pi (processing instruction) + * + * @var string + */ + public $nodeType = ''; + + /** + * current node content, i.e. either a + * simple string (text node), or something like + * <tag attrib="value"...> + * + * @var string + */ + public $node = ''; + + /** + * wether current node is an opening tag (<a>) or not (</a>) + * set to NULL if current node is not a tag + * NOTE: empty tags (<br />) set this to true as well! + * + * @var bool | null + */ + public $isStartTag = null; + + /** + * wether current node is an empty tag (<br />) or not (<a></a>) + * + * @var bool | null + */ + public $isEmptyTag = null; + + /** + * tag name + * + * @var string | null + */ + public $tagName = ''; + + /** + * attributes of current tag + * + * @var array (attribName=>value) | null + */ + public $tagAttributes = null; + + /** + * whether or not the actual context is a inline context + * + * @var bool | null + */ + public $isInlineContext = null; + + /** + * whether the current tag is a block element + * + * @var bool | null + */ + public $isBlockElement = null; + + /** + * whether the previous tag (browser) is a block element + * + * @var bool | null + */ + public $isNextToInlineContext = null; + + /** + * keep whitespace + * + * @var int + */ + public $keepWhitespace = 0; + + /** + * list of open tags + * count this to get current depth + * + * @var array + */ + public $openTags = array(); + + /** + * list of block elements + * + * @var array + * TODO: what shall we do with <del> and <ins> ?! + */ + public $blockElements = array( + // tag name => <bool> is block + // block elements + 'address' => true, + 'blockquote' => true, + 'center' => true, + 'del' => true, + 'dir' => true, + 'div' => true, + 'dl' => true, + 'fieldset' => true, + 'form' => true, + 'h1' => true, + 'h2' => true, + 'h3' => true, + 'h4' => true, + 'h5' => true, + 'h6' => true, + 'hr' => true, + 'ins' => true, + 'isindex' => true, + 'menu' => true, + 'noframes' => true, + 'noscript' => true, + 'ol' => true, + 'p' => true, + 'pre' => true, + 'table' => true, + 'ul' => true, + // set table elements and list items to block as well + 'thead' => true, + 'tbody' => true, + 'tfoot' => true, + 'td' => true, + 'tr' => true, + 'th' => true, + 'li' => true, + 'dd' => true, + 'dt' => true, + // header items and html / body as well + 'html' => true, + 'body' => true, + 'head' => true, + 'meta' => true, + 'link' => true, + 'style' => true, + 'title' => true, + // unfancy media tags, when indented should be rendered as block + 'map' => true, + 'object' => true, + 'param' => true, + 'embed' => true, + 'area' => true, + // inline elements + 'a' => false, + 'abbr' => false, + 'acronym' => false, + 'applet' => false, + 'b' => false, + 'basefont' => false, + 'bdo' => false, + 'big' => false, + 'br' => false, + 'button' => false, + 'cite' => false, + 'code' => false, + 'del' => false, + 'dfn' => false, + 'em' => false, + 'font' => false, + 'i' => false, + 'img' => false, + 'ins' => false, + 'input' => false, + 'iframe' => false, + 'kbd' => false, + 'label' => false, + 'q' => false, + 'samp' => false, + 'script' => false, + 'select' => false, + 'small' => false, + 'span' => false, + 'strong' => false, + 'sub' => false, + 'sup' => false, + 'textarea' => false, + 'tt' => false, + 'var' => false, + ); + + /** + * get next node, set $this->html prior! + * + * @param void + * @return bool + */ + public function nextNode() + { + if (empty($this->html)) { + // we are done with parsing the html string + + return false; + } + + if ($this->isStartTag && !$this->isEmptyTag) { + array_push($this->openTags, $this->tagName); + if (in_array($this->tagName, $this->preformattedTags)) { + // dont truncate whitespaces for <code> or <pre> contents + $this->keepWhitespace++; + } + } + + if ($this->html[0] == '<') { + $token = substr($this->html, 0, 9); + if (substr($token, 0, 2) == '<?') { + // xml prolog or other pi's + /** TODO **/ + // trigger_error('this might need some work', E_USER_NOTICE); + $pos = strpos($this->html, '>'); + $this->setNode('pi', $pos + 1); + + return true; + } + if (substr($token, 0, 4) == '<!--') { + // comment + $pos = strpos($this->html, '-->'); + if ($pos === false) { + // could not find a closing -->, use next gt instead + // this is firefox' behaviour + $pos = strpos($this->html, '>') + 1; + } else { + $pos += 3; + } + $this->setNode('comment', $pos); + + static::$skipWhitespace = true; + + return true; + } + if ($token == '<!DOCTYPE') { + // doctype + $this->setNode('doctype', strpos($this->html, '>') + 1); + + static::$skipWhitespace = true; + + return true; + } + if ($token == '<![CDATA[') { + // cdata, use text node + + // remove leading <![CDATA[ + $this->html = substr($this->html, 9); + + $this->setNode('text', strpos($this->html, ']]>') + 3); + + // remove trailing ]]> and trim + $this->node = substr($this->node, 0, -3); + $this->handleWhitespaces(); + + static::$skipWhitespace = true; + + return true; + } + if ($this->parseTag()) { + // seems to be a tag + // handle whitespaces + if ($this->isBlockElement) { + static::$skipWhitespace = true; + } else { + static::$skipWhitespace = false; + } + + return true; + } + } + if ($this->keepWhitespace) { + static::$skipWhitespace = false; + } + // when we get here it seems to be a text node + $pos = strpos($this->html, '<'); + if ($pos === false) { + $pos = strlen($this->html); + } + $this->setNode('text', $pos); + $this->handleWhitespaces(); + if (static::$skipWhitespace && $this->node == ' ') { + return $this->nextNode(); + } + $this->isInlineContext = true; + static::$skipWhitespace = false; + + return true; + } + + /** + * parse tag, set tag name and attributes, see if it's a closing tag and so forth... + * + * @param void + * @return bool + */ + protected function parseTag() + { + if (!isset(static::$a_ord)) { + static::$a_ord = ord('a'); + static::$z_ord = ord('z'); + static::$special_ords = array( + ord(':'), // for xml:lang + ord('-'), // for http-equiv + ); + } + + $tagName = ''; + + $pos = 1; + $isStartTag = $this->html[$pos] != '/'; + if (!$isStartTag) { + $pos++; + } + // get tagName + while (isset($this->html[$pos])) { + $pos_ord = ord(strtolower($this->html[$pos])); + if (($pos_ord >= static::$a_ord && $pos_ord <= static::$z_ord) || (!empty($tagName) && is_numeric($this->html[$pos]))) { + $tagName .= $this->html[$pos]; + $pos++; + } else { + $pos--; + break; + } + } + + $tagName = strtolower($tagName); + if (empty($tagName) || !isset($this->blockElements[$tagName])) { + // something went wrong => invalid tag + $this->invalidTag(); + + return false; + } + if ($this->noTagsInCode && end($this->openTags) == 'code' && !($tagName == 'code' && !$isStartTag)) { + // we supress all HTML tags inside code tags + $this->invalidTag(); + + return false; + } + + // get tag attributes + /** TODO: in html 4 attributes do not need to be quoted **/ + $isEmptyTag = false; + $attributes = array(); + $currAttrib = ''; + while (isset($this->html[$pos + 1])) { + $pos++; + // close tag + if ($this->html[$pos] == '>' || $this->html[$pos] . $this->html[$pos + 1] == '/>') { + if ($this->html[$pos] == '/') { + $isEmptyTag = true; + $pos++; + } + break; + } + + $pos_ord = ord(strtolower($this->html[$pos])); + if (($pos_ord >= static::$a_ord && $pos_ord <= static::$z_ord) || in_array($pos_ord, static::$special_ords)) { + // attribute name + $currAttrib .= $this->html[$pos]; + } elseif (in_array($this->html[$pos], array(' ', "\t", "\n"))) { + // drop whitespace + } elseif (in_array($this->html[$pos] . $this->html[$pos + 1], array('="', "='"))) { + // get attribute value + $pos++; + $await = $this->html[$pos]; // single or double quote + $pos++; + $value = ''; + while (isset($this->html[$pos]) && $this->html[$pos] != $await) { + $value .= $this->html[$pos]; + $pos++; + } + $attributes[$currAttrib] = $value; + $currAttrib = ''; + } else { + $this->invalidTag(); + + return false; + } + } + if ($this->html[$pos] != '>') { + $this->invalidTag(); + + return false; + } + + if (!empty($currAttrib)) { + // html 4 allows something like <option selected> instead of <option selected="selected"> + $attributes[$currAttrib] = $currAttrib; + } + if (!$isStartTag) { + if (!empty($attributes) || $tagName != end($this->openTags)) { + // end tags must not contain any attributes + // or maybe we did not expect a different tag to be closed + $this->invalidTag(); + + return false; + } + array_pop($this->openTags); + if (in_array($tagName, $this->preformattedTags)) { + $this->keepWhitespace--; + } + } + $pos++; + $this->node = substr($this->html, 0, $pos); + $this->html = substr($this->html, $pos); + $this->tagName = $tagName; + $this->tagAttributes = $attributes; + $this->isStartTag = $isStartTag; + $this->isEmptyTag = $isEmptyTag || in_array($tagName, $this->emptyTags); + if ($this->isEmptyTag) { + // might be not well formed + $this->node = preg_replace('# */? *>$#', ' />', $this->node); + } + $this->nodeType = 'tag'; + $this->isBlockElement = $this->blockElements[$tagName]; + $this->isNextToInlineContext = $isStartTag && $this->isInlineContext; + $this->isInlineContext = !$this->isBlockElement; + return true; + } + + /** + * handle invalid tags + * + * @param void + * @return void + */ + protected function invalidTag() + { + $this->html = substr_replace($this->html, '<', 0, 1); + } + + /** + * update all vars and make $this->html shorter + * + * @param string $type see description for $this->nodeType + * @param int $pos to which position shall we cut? + * @return void + */ + protected function setNode($type, $pos) + { + if ($this->nodeType == 'tag') { + // set tag specific vars to null + // $type == tag should not be called here + // see this::parseTag() for more + $this->tagName = null; + $this->tagAttributes = null; + $this->isStartTag = null; + $this->isEmptyTag = null; + $this->isBlockElement = null; + + } + $this->nodeType = $type; + $this->node = substr($this->html, 0, $pos); + $this->html = substr($this->html, $pos); + } + + /** + * check if $this->html begins with $str + * + * @param string $str + * @return bool + */ + protected function match($str) + { + return substr($this->html, 0, strlen($str)) == $str; + } + + /** + * truncate whitespaces + * + * @param void + * @return void + */ + protected function handleWhitespaces() + { + if ($this->keepWhitespace) { + // <pre> or <code> before... + + return; + } + // truncate multiple whitespaces to a single one + $this->node = preg_replace('#\s+#s', ' ', $this->node); + } + + /** + * normalize self::node + * + * @param void + * @return void + */ + protected function normalizeNode() + { + $this->node = '<'; + if (!$this->isStartTag) { + $this->node .= '/' . $this->tagName . '>'; + + return; + } + $this->node .= $this->tagName; + foreach ($this->tagAttributes as $name => $value) { + $this->node .= ' ' . $name . '="' . str_replace('"', '"', $value) . '"'; + } + if ($this->isEmptyTag) { + $this->node .= ' /'; + } + $this->node .= '>'; + } +} |