diff --git a/parsers/BBCodeParser.ini b/parsers/BBCodeParser.ini new file mode 100644 index 000000000..83bb44920 --- /dev/null +++ b/parsers/BBCodeParser.ini @@ -0,0 +1,24 @@ +[SSHTMLBBCodeParser] + +; possible values: single|double +; use single or double quotes for attributes +quotestyle = "double" + +; possible values: all|nothing|strings +; quote all attribute values, none, or only the strings +quotewhat = "all" + +; the closing tag character +close = "]" + +; the opening tag character +open = "[" + +; possible values: true|false +; use xml style closing tags for single html tags ( or ) +xmlclose = "true" + +; possible values: a comma seperated list of filters +; comma seperated list of filters to use +filters = "Basic,Extended,EmailLinks,Images,Links,Lists" + diff --git a/parsers/BBCodeParser.php b/parsers/BBCodeParser.php new file mode 100644 index 000000000..f124ba152 --- /dev/null +++ b/parsers/BBCodeParser.php @@ -0,0 +1,112 @@ + "Bold Text", + "Example" => "[b]Bold[/b]" + )), + new ArrayData(array( + "Title" => "Italic Text", + "Example" => "[i]Italics[/i]" + )), + new ArrayData(array( + "Title" => "Underlined Text", + "Example" => "[u]Underlined[/u]" + )), + new ArrayData(array( + "Title" => "Struck-out Text", + "Example" => "[s]Struck-out[/s]" + )), + new ArrayData(array( + "Title" => "Colored text", + "Example" => "[color=blue]blue text[/color]" + )), + new ArrayData(array( + "Title" => "Alignment", + "Example" => "[align=right]right aligned[/align]" + )), + + new ArrayData(array( + "Title" => "Website link", + "Description" => "Link to another website or URL", + "Example" => "[url]http://www.website.com/[/url]" + )), + new ArrayData(array( + "Title" => "Website link", + "Description" => "Link to another website or URL", + "Example" => "[url=http://www.website.com/]Some website[/url]" + )), + new ArrayData(array( + "Title" => "Email link", + "Description" => "Create link to an email address", + "Example" => "[email]you@yoursite.com[/email]" + )), + new ArrayData(array( + "Title" => "Email link", + "Description" => "Create link to an email address", + "Example" => "[email=you@yoursite.com]email me[/email]" + )), + + new ArrayData(array( + "Title" => "Image", + "Description" => "Show an image in your post", + "Example" => "[img]http://www.website.com/image.jpg[/img]" + )), + + new ArrayData(array( + "Title" => "Code Block", + "Description" => "Unformatted code block", + "Example" => "[code]Code block[/code]" + )), + new ArrayData(array( + "Title" => "Unordered list", + "Description" => "Unordered list", + "Example" => "[ulist][*]unordered item 1[*]unordered item 2[/ulist]" + )) + + + ); + } + + function useable_tagsHTML(){ + $useabletags = ""; + } + + function parse() { + $this->content = str_replace(array('&', '<', '>'), array('&', '<', '>'), $this->content); + $this->content = SSHTMLBBCodeParser::staticQparse($this->content); + $this->content = "

".$this->content."

"; + $this->content = str_replace("\n\n", "

", $this->content); + $this->content = str_replace("\n", "
", $this->content); + return $this->content; + } + +} +?> \ No newline at end of file diff --git a/parsers/HTML/BBCodeParser/Filter.php b/parsers/HTML/BBCodeParser/Filter.php new file mode 100644 index 000000000..ca92de584 --- /dev/null +++ b/parsers/HTML/BBCodeParser/Filter.php @@ -0,0 +1,10 @@ + \ No newline at end of file diff --git a/parsers/HTML/BBCodeParser/Filter/Basic.php b/parsers/HTML/BBCodeParser/Filter/Basic.php new file mode 100644 index 000000000..81c58e01a --- /dev/null +++ b/parsers/HTML/BBCodeParser/Filter/Basic.php @@ -0,0 +1,71 @@ + | +// +----------------------------------------------------------------------+ +// +// $Id: Basic.php,v 1.6 2007/07/02 16:54:25 cweiske Exp $ +// + +/** +* @package SSHTMLBBCodeParser +* @author Stijn de Reede +*/ + + +require_once 'HTML/BBCodeParser/Filter.php'; + + + + +class SSHTMLBBCodeParser_Filter_Basic extends SSHTMLBBCodeParser_Filter +{ + + /** + * An array of tags parsed by the engine + * + * @access private + * @var array + */ + var $_definedTags = array( 'b' => array( 'htmlopen' => 'strong', + 'htmlclose' => 'strong', + 'allowed' => 'all', + 'attributes'=> array()), + 'i' => array( 'htmlopen' => 'em', + 'htmlclose' => 'em', + 'allowed' => 'all', + 'attributes'=> array()), + 'u' => array( 'htmlopen' => 'span style="text-decoration:underline;"', + 'htmlclose' => 'span', + 'allowed' => 'all', + 'attributes'=> array()), + 's' => array( 'htmlopen' => 'del', + 'htmlclose' => 'del', + 'allowed' => 'all', + 'attributes'=> array()), + 'sub' => array( 'htmlopen' => 'sub', + 'htmlclose' => 'sub', + 'allowed' => 'all', + 'attributes'=> array()), + 'sup' => array( 'htmlopen' => 'sup', + 'htmlclose' => 'sup', + 'allowed' => 'all', + 'attributes'=> array()) + ); + +} + + +?> \ No newline at end of file diff --git a/parsers/HTML/BBCodeParser/Filter/EmailLinks.php b/parsers/HTML/BBCodeParser/Filter/EmailLinks.php new file mode 100644 index 000000000..1c5f9b6cf --- /dev/null +++ b/parsers/HTML/BBCodeParser/Filter/EmailLinks.php @@ -0,0 +1,85 @@ + | +// +----------------------------------------------------------------------+ +// +// $Id: Email.php,v 1.5 2007/07/02 16:54:25 cweiske Exp $ +// + +/** +* @package SSHTMLBBCodeParser +* @author Stijn de Reede +*/ + + +require_once 'HTML/BBCodeParser/Filter.php'; + + + + +class SSHTMLBBCodeParser_Filter_EmailLinks extends SSHTMLBBCodeParser_Filter +{ + + /** + * An array of tags parsed by the engine + * + * @access private + * @var array + */ + var $_definedTags = array( 'email' => array( 'htmlopen' => 'a', + 'htmlclose' => 'a', + 'allowed' => 'none^img', + 'attributes'=> array('email' =>'href=%2$smailto:%1$s%2$s') + + ) + ); + + + /** + * Executes statements before the actual array building starts + * + * This method should be overwritten in a filter if you want to do + * something before the parsing process starts. This can be useful to + * allow certain short alternative tags which then can be converted into + * proper tags with preg_replace() calls. + * The main class walks through all the filters and and calls this + * method if it exists. The filters should modify their private $_text + * variable. + * + * @return none + * @access private + * @see $_text + * @author Stijn de Reede + */ + function _preparse() + { + $options = SSHTMLBBCodeParser::getStaticProperty('SSHTMLBBCodeParser','_options'); + $o = $options['open']; + $c = $options['close']; + $oe = $options['open_esc']; + $ce = $options['close_esc']; + $pattern = array( "!(^|\s)([-a-z0-9_.]+@[-a-z0-9.]+\.[a-z]{2,4})!i", + "!".$oe."email(".$ce."|\s.*".$ce.")(.*)".$oe."/email".$ce."!Ui"); + $replace = array( "\\1".$o."email=\\2".$c."\\2".$o."/email".$c, + $o."email=\\2\\1\\2".$o."/email".$c); + $this->_preparsed = preg_replace($pattern, $replace, $this->_text); + } + + +} + + +?> diff --git a/parsers/HTML/BBCodeParser/Filter/Extended.php b/parsers/HTML/BBCodeParser/Filter/Extended.php new file mode 100644 index 000000000..7482c7287 --- /dev/null +++ b/parsers/HTML/BBCodeParser/Filter/Extended.php @@ -0,0 +1,97 @@ + | +// +----------------------------------------------------------------------+ +// +// $Id: Extended.php,v 1.3 2007/07/02 16:54:25 cweiske Exp $ +// + +/** +* @package SSHTMLBBCodeParser +* @author Stijn de Reede +*/ + + +require_once 'HTML/BBCodeParser/Filter.php'; + + + + +class SSHTMLBBCodeParser_Filter_Extended extends SSHTMLBBCodeParser_Filter +{ + + /** + * An array of tags parsed by the engine + * + * @access private + * @var array + */ + var $_definedTags = array( + 'color' => array( 'htmlopen' => 'span', + 'htmlclose' => 'span', + 'allowed' => 'all', + 'attributes'=> array('color' =>'style=%2$scolor:%1$s%2$s')), + 'size' => array( 'htmlopen' => 'span', + 'htmlclose' => 'span', + 'allowed' => 'all', + 'attributes'=> array('size' =>'style=%2$sfont-size:%1$spt%2$s')), + 'font' => array( 'htmlopen' => 'span', + 'htmlclose' => 'span', + 'allowed' => 'all', + 'attributes'=> array('font' =>'style=%2$sfont-family:%1$s%2$s')), + 'align' => array( 'htmlopen' => 'div', + 'htmlclose' => 'div', + 'allowed' => 'all', + 'attributes'=> array('align' =>'style=%2$stext-align:%1$s%2$s')), + 'quote' => array('htmlopen' => 'q', + 'htmlclose' => 'q', + 'allowed' => 'all', + 'attributes'=> array('quote' =>'cite=%2$s%1$s%2$s')), + 'code' => array('htmlopen' => 'pre', + 'htmlclose' => 'pre', + 'allowed' => 'all', + 'attributes'=> array()), + 'h1' => array('htmlopen' => 'h1', + 'htmlclose' => 'h1', + 'allowed' => 'all', + 'attributes'=> array()), + 'h2' => array('htmlopen' => 'h2', + 'htmlclose' => 'h2', + 'allowed' => 'all', + 'attributes'=> array()), + 'h3' => array('htmlopen' => 'h3', + 'htmlclose' => 'h3', + 'allowed' => 'all', + 'attributes'=> array()), + 'h4' => array('htmlopen' => 'h4', + 'htmlclose' => 'h4', + 'allowed' => 'all', + 'attributes'=> array()), + 'h5' => array('htmlopen' => 'h5', + 'htmlclose' => 'h5', + 'allowed' => 'all', + 'attributes'=> array()), + 'h6' => array('htmlopen' => 'h6', + 'htmlclose' => 'h6', + 'allowed' => 'all', + 'attributes'=> array()) + + ); + + +} + +?> \ No newline at end of file diff --git a/parsers/HTML/BBCodeParser/Filter/Images.php b/parsers/HTML/BBCodeParser/Filter/Images.php new file mode 100644 index 000000000..0abc82e9d --- /dev/null +++ b/parsers/HTML/BBCodeParser/Filter/Images.php @@ -0,0 +1,79 @@ + | +// +----------------------------------------------------------------------+ +// +// $Id: Images.php,v 1.8 2007/07/02 17:44:47 cweiske Exp $ +// + +/** +* @package SSHTMLBBCodeParser +* @author Stijn de Reede +*/ +require_once 'HTML/BBCodeParser/Filter.php'; + +class SSHTMLBBCodeParser_Filter_Images extends SSHTMLBBCodeParser_Filter +{ + + /** + * An array of tags parsed by the engine + * + * @access private + * @var array + */ + var $_definedTags = array( + 'img' => array( + 'htmlopen' => 'img', + 'htmlclose' => '', + 'allowed' => 'none', + 'attributes'=> array( + 'img' => 'src=%2$s%1$s%2$s', + 'w' => 'width=%2$s%1$d%2$s', + 'h' => 'height=%2$s%1$d%2$s', + 'alt' => 'alt=%2$s%1$s%2$s', + ) + ) + ); + + /** + * Executes statements before the actual array building starts + * + * This method should be overwritten in a filter if you want to do + * something before the parsing process starts. This can be useful to + * allow certain short alternative tags which then can be converted into + * proper tags with preg_replace() calls. + * The main class walks through all the filters and and calls this + * method if it exists. The filters should modify their private $_text + * variable. + * + * @return none + * @access private + * @see $_text + * @author Stijn de Reede + */ + function _preparse() + { + $options = SSHTMLBBCodeParser::getStaticProperty('SSHTMLBBCodeParser','_options'); + $o = $options['open']; + $c = $options['close']; + $oe = $options['open_esc']; + $ce = $options['close_esc']; + $this->_preparsed = preg_replace( + "!".$oe."img(\s?.*)".$ce."(.*)".$oe."/img".$ce."!Ui", + $o."img=\"\$2\"\$1".$c.$o."/img".$c, + $this->_text); + } +} \ No newline at end of file diff --git a/parsers/HTML/BBCodeParser/Filter/Links.php b/parsers/HTML/BBCodeParser/Filter/Links.php new file mode 100644 index 000000000..c40898f03 --- /dev/null +++ b/parsers/HTML/BBCodeParser/Filter/Links.php @@ -0,0 +1,205 @@ + | +// +----------------------------------------------------------------------+ +// +// $Id: Links.php,v 1.12 2007/07/02 18:26:17 cweiske Exp $ +// + +/** +* @package SSHTMLBBCodeParser +* @author Stijn de Reede +*/ +require_once 'HTML/BBCodeParser/Filter.php'; + +/** + * + */ +class SSHTMLBBCodeParser_Filter_Links extends SSHTMLBBCodeParser_Filter +{ + /** + * List of allowed schemes + * + * @access private + * @var array + */ + var $_allowedSchemes = array('http', 'https', 'ftp'); + + /** + * Default scheme + * + * @access private + * @var string + */ + var $_defaultScheme = 'http'; + + /** + * An array of tags parsed by the engine + * + * @access private + * @var array + */ + var $_definedTags = array( + 'url' => array( + 'htmlopen' => 'a', + 'htmlclose' => 'a', + 'allowed' => 'none^img', + 'attributes'=> array('url' => 'href=%2$s%1$s%2$s') + ) + ); + + + /** + * Executes statements before the actual array building starts + * + * This method should be overwritten in a filter if you want to do + * something before the parsing process starts. This can be useful to + * allow certain short alternative tags which then can be converted into + * proper tags with preg_replace() calls. + * The main class walks through all the filters and and calls this + * method if it exists. The filters should modify their private $_text + * variable. + * + * @return none + * @access private + * @see $_text + * @author Stijn de Reede + * @author Seth Price + */ + function _preparse() + { + $options = SSHTMLBBCodeParser::getStaticProperty('SSHTMLBBCodeParser', '_options'); + $o = $options['open']; + $c = $options['close']; + $oe = $options['open_esc']; + $ce = $options['close_esc']; + + $schemes = implode('|', $this->_allowedSchemes); + + $pattern = array( "/(?_text); + $pp = preg_replace($pattern[1], $o."url=\$2\$1\$2".$o."/url".$c, $pp); + $this->_preparsed = preg_replace_callback($pattern[2], array($this, 'smarterPPLink'), $pp); + + } + + /** + * Intelligently expand a URL into a link + * + * @return string + * @access private + * @author Seth Price + * @author Lorenzo Alberton + */ + function smarterPPLinkExpand($matches) + { + //echo '


';var_dump($matches);echo '

'; + $options = SSHTMLBBCodeParser::getStaticProperty('SSHTMLBBCodeParser','_options'); + $o = $options['open']; + $c = $options['close']; + + //If we have an intro tag that is [url], then skip this match + if ($matches[1] == $o.'url'.$c) { + return $matches[0]; + } + + if(!BBCodeParser::autolinkUrls()){ + return $matches[0]; + } + + + $punctuation = '.,;:'; // Links can't end with these chars + $trailing = ''; + // Knock off ending punctuation + $last = substr($matches[2], -1); + while (strpos($punctuation, $last) !== false) { + // Last character is punctuation - remove it from the url + $trailing = $last.$trailing; + $matches[2] = substr($matches[2], 0, -1); + $last = substr($matches[2], -1); + } + + $off = strpos($matches[2], ':'); + + //Is a ":" (therefore a scheme) defined? + if ($off === false) { + /* + * Create a link with the default scheme of http. Notice that the + * text that is viewable to the user is unchanged, but the link + * itself contains the "http://". + */ + return $matches[1].$o.'url='.$this->_defaultScheme.'://'.$matches[2].$c.$matches[2].$o.'/url'.$c.$trailing; + } + + $scheme = substr($matches[2], 0, $off); + + /* + * If protocol is in the approved list than allow it. Note that this + * check isn't really needed, but the created link will just be deleted + * later in smarterPPLink() if we create it now and it isn't on the + * scheme list. + */ + if (in_array($scheme, $this->_allowedSchemes)) { + return $matches[1].$o.'url'.$c.$matches[2].$o.'/url'.$c.$trailing; + } + + return $matches[0]; + } + + /** + * Finish preparsing URL to clean it up + * + * @return string + * @access private + * @author Seth Price + */ + function smarterPPLink($matches) + { + $options = SSHTMLBBCodeParser::getStaticProperty('SSHTMLBBCodeParser','_options'); + $o = $options['open']; + $c = $options['close']; + + $urlServ = $matches[1]; + $path = $matches[5]; + + $off = strpos($urlServ, ':'); + + if ($off === false) { + //Default to http + $urlServ = $this->_defaultScheme.'://'.$urlServ; + $off = strpos($urlServ, ':'); + } + + //Add trailing slash if missing (to create a valid URL) + if (!$path) { + $path = '/'; + } + + $protocol = substr($urlServ, 0, $off); + + if (in_array($protocol, $this->_allowedSchemes)) { + //If protocol is in the approved list than allow it + return $o.'url='.$urlServ.$path.$c.$matches[6].$o.'/url'.$c; + } + + //Else remove url tag + return $matches[6]; + } +} +?> \ No newline at end of file diff --git a/parsers/HTML/BBCodeParser/Filter/Lists.php b/parsers/HTML/BBCodeParser/Filter/Lists.php new file mode 100644 index 000000000..053ddb39c --- /dev/null +++ b/parsers/HTML/BBCodeParser/Filter/Lists.php @@ -0,0 +1,108 @@ + | +// +----------------------------------------------------------------------+ +// +// $Id: Lists.php,v 1.5 2007/07/02 16:54:25 cweiske Exp $ +// + + +/** +* @package SSHTMLBBCodeParser +* @author Stijn de Reede +*/ + +require_once 'HTML/BBCodeParser/Filter.php'; + +/** + * + */ +class SSHTMLBBCodeParser_Filter_Lists extends SSHTMLBBCodeParser_Filter +{ + + /** + * An array of tags parsed by the engine + * + * @access private + * @var array + */ + var $_definedTags = array( 'list' => array( 'htmlopen' => 'ol', + 'htmlclose' => 'ol', + 'allowed' => 'all', + 'child' => 'none^li', + 'attributes'=> array('list' => 'style=%2$slist-style-type:%1$s;%2$s') + ), + 'ulist' => array( 'htmlopen' => 'ul', + 'htmlclose' => 'ul', + 'allowed' => 'all', + 'child' => 'none^li', + 'attributes'=> array('list' => 'style=%2$slist-style-type:%1$s;%2$s') + ), + 'li' => array( 'htmlopen' => 'li', + 'htmlclose' => 'li', + 'allowed' => 'all', + 'parent' => 'none^ulist,list', + 'attributes'=> array() + ) + ); + + + /** + * Executes statements before the actual array building starts + * + * This method should be overwritten in a filter if you want to do + * something before the parsing process starts. This can be useful to + * allow certain short alternative tags which then can be converted into + * proper tags with preg_replace() calls. + * The main class walks through all the filters and and calls this + * method if it exists. The filters should modify their private $_text + * variable. + * + * @return none + * @access private + * @see $_text + * @author Stijn de Reede , Seth Price + */ + function _preparse() + { + $options = SSHTMLBBCodeParser::getStaticProperty('SSHTMLBBCodeParser','_options'); + $o = $options['open']; + $c = $options['close']; + $oe = $options['open_esc']; + $ce = $options['close_esc']; + + $pattern = array( "!".$oe."\*".$ce."!", + "!".$oe."(u?)list=(?-i:A)(\s*[^".$ce."]*)".$ce."!i", + "!".$oe."(u?)list=(?-i:a)(\s*[^".$ce."]*)".$ce."!i", + "!".$oe."(u?)list=(?-i:I)(\s*[^".$ce."]*)".$ce."!i", + "!".$oe."(u?)list=(?-i:i)(\s*[^".$ce."]*)".$ce."!i", + "!".$oe."(u?)list=(?-i:1)(\s*[^".$ce."]*)".$ce."!i", + "!".$oe."(u?)list([^".$ce."]*)".$ce."!i"); + + $replace = array( $o."li".$c, + $o."\$1list=upper-alpha\$2".$c, + $o."\$1list=lower-alpha\$2".$c, + $o."\$1list=upper-roman\$2".$c, + $o."\$1list=lower-roman\$2".$c, + $o."\$1list=decimal\$2".$c, + $o."\$1list\$2".$c ); + + $this->_preparsed = preg_replace($pattern, $replace, $this->_text); + } +} + + +?> \ No newline at end of file diff --git a/parsers/HTML/HTMLBBCodeParser.php b/parsers/HTML/HTMLBBCodeParser.php new file mode 100644 index 000000000..80ee3f126 --- /dev/null +++ b/parsers/HTML/HTMLBBCodeParser.php @@ -0,0 +1,873 @@ + | +// +----------------------------------------------------------------------+ +// +// $Id: BBCodeParser.php,v 1.17 2007/07/02 18:46:30 cweiske Exp $ +// +// Modified by SilverStripe www.silverstripe.com + +/** +* @package SSHTMLBBCodeParser +* @author Stijn de Reede , SilverStripe +* +* +* This is a parser to replace UBB style tags with their html equivalents. It +* does not simply do some regex calls, but is complete stack based +* parse engine. This ensures that all tags are properly nested, if not, +* extra tags are added to maintain the nesting. This parser should only produce +* xhtml 1.0 compliant code. All tags are validated and so are all their attributes. +* It should be easy to extend this parser with your own tags, see the _definedTags +* format description below. +* +* +* Usage: +* $parser = new SSHTMLBBCodeParser(); +* $parser->setText('normal [b]bold[/b] and normal again'); +* $parser->parse(); +* echo $parser->getParsed(); +* or: +* $parser = new SSHTMLBBCodeParser(); +* echo $parser->qparse('normal [b]bold[/b] and normal again'); +* or: +* echo SSHTMLBBCodeParser::staticQparse('normal [b]bold[/b] and normal again'); +* +* +* Setting the options from the ini file: +* $config = parse_ini_file('BBCodeParser.ini', true); +* $options = &PEAR::getStaticProperty('SSHTMLBBCodeParser', '_options'); +* $options = $config['SSHTMLBBCodeParser']; +* unset($options); +* +* +*/ + +class SSHTMLBBCodeParser +{ + /** + * An array of tags parsed by the engine, should be overwritten by filters + * + * @access private + * @var array + */ + var $_definedTags = array(); + + /** + * A string containing the input + * + * @access private + * @var string + */ + var $_text = ''; + + /** + * A string containing the preparsed input + * + * @access private + * @var string + */ + var $_preparsed = ''; + + /** + * An array tags and texts build from the input text + * + * @access private + * @var array + */ + var $_tagArray = array(); + + /** + * A string containing the parsed version of the text + * + * @access private + * @var string + */ + var $_parsed = ''; + + /** + * An array of options, filled by an ini file or through the contructor + * + * @access private + * @var array + */ + var $_options = array( + 'quotestyle' => 'double', + 'quotewhat' => 'all', + 'open' => '[', + 'close' => ']', + 'xmlclose' => true, + 'filters' => 'Basic' + ); + + /** + * An array of filters used for parsing + * + * @access private + * @var array + */ + var $_filters = array(); + + /** + * Constructor, initialises the options and filters + * + * Sets the private variable _options with base options defined with + * &PEAR::getStaticProperty(), overwriting them with (if present) + * the argument to this method. + * Then it sets the extra options to properly escape the tag + * characters in preg_replace() etc. The set options are + * then stored back with &PEAR::getStaticProperty(), so that the filter + * classes can use them. + * All the filters in the options are initialised and their defined tags + * are copied into the private variable _definedTags. + * + * @param array options to use, can be left out + * @return none + * @access public + * @author Stijn de Reede + */ + function SSHTMLBBCodeParser($options = array()) + { + // set the already set options + $baseoptions = &SSHTMLBBCodeParser::getStaticProperty('SSHTMLBBCodeParser', '_options'); + if (is_array($baseoptions)) { + foreach ($baseoptions as $k => $v) { + $this->_options[$k] = $v; + } + } + + // set the options passed as an argument + foreach ($options as $k => $v ) { + $this->_options[$k] = $v; + } + + // add escape open and close chars to the options for preg escaping + $preg_escape = '\^$.[]|()?*+{}'; + if ($this->_options['open'] != '' && strpos($preg_escape, $this->_options['open'])) { + $this->_options['open_esc'] = "\\".$this->_options['open']; + } else { + $this->_options['open_esc'] = $this->_options['open']; + } + if ($this->_options['close'] != '' && strpos($preg_escape, $this->_options['close'])) { + $this->_options['close_esc'] = "\\".$this->_options['close']; + } else { + $this->_options['close_esc'] = $this->_options['close']; + } + + // set the options back so that child classes can use them */ + $baseoptions = $this->_options; + unset($baseoptions); + + // return if this is a subclass + if (is_subclass_of($this, 'SSHTMLBBCodeParser_Filter')) { + return; + } + + // extract the definedTags from subclasses */ + $this->addFilters($this->_options['filters']); + } + + function &getStaticProperty($class, $var) + { + static $properties; + if (!isset($properties[$class])) { + $properties[$class] = array(); + } + if (!array_key_exists($var, $properties[$class])) { + $properties[$class][$var] = null; + } + return $properties[$class][$var]; + } + + /** + * Option setter + * + * @param string option name + * @param mixed option value + * @author Lorenzo Alberton + */ + function setOption($name, $value) + { + $this->_options[$name] = $value; + } + + /** + * Add a new filter + * + * @param string filter + * @author Lorenzo Alberton + */ + function addFilter($filter) + { + + $filter = ucfirst($filter); + if (!array_key_exists($filter, $this->_filters)) { + $class = 'SSHTMLBBCodeParser_Filter_'.$filter; + @include_once 'HTML/BBCodeParser/Filter/'.$filter.'.php'; + if (!class_exists($class)) { + + //PEAR::raiseError("Failed to load filter $filter", null, PEAR_ERROR_DIE); + } + + $this->_filters[$filter] = new $class; + $this->_definedTags = array_merge( + $this->_definedTags, + $this->_filters[$filter]->_definedTags + ); + } + + } + + /** + * Remove an existing filter + * + * @param string $filter + * @author Lorenzo Alberton + */ + function removeFilter($filter) + { + $filter = ucfirst(trim($filter)); + if (!empty($filter) && array_key_exists($filter, $this->_filters)) { + unset($this->_filters[$filter]); + } + // also remove the related $this->_definedTags for this filter, + // preserving the others + $this->_definedTags = array(); + foreach (array_keys($this->_filters) as $filter) { + $this->_definedTags = array_merge( + $this->_definedTags, + $this->_filters[$filter]->_definedTags + ); + } + } + + /** + * Add new filters + * + * @param mixed (array or string) + * @return boolean true if all ok, false if not. + * @author Lorenzo Alberton + */ + function addFilters($filters) + { + if (is_string($filters)) { + //comma-separated list + if (strpos($filters, ',') !== false) { + $filters = explode(',', $filters); + } else { + $filters = array($filters); + } + } + if (!is_array($filters)) { + //invalid format + return false; + } + foreach ($filters as $filter) { + if (trim($filter)){ + $this->addFilter($filter); + } + } + return true; + } + + /** + * Executes statements before the actual array building starts + * + * This method should be overwritten in a filter if you want to do + * something before the parsing process starts. This can be useful to + * allow certain short alternative tags which then can be converted into + * proper tags with preg_replace() calls. + * The main class walks through all the filters and and calls this + * method. The filters should modify their private $_preparsed + * variable, with input from $_text. + * + * @return none + * @access private + * @see $_text + * @author Stijn de Reede + */ + function _preparse() + { + // default: assign _text to _preparsed, to be overwritten by filters + $this->_preparsed = $this->_text; + + // return if this is a subclass + if (is_subclass_of($this, 'SSHTMLBBCodeParser')) { + return; + } + + // walk through the filters and execute _preparse + foreach ($this->_filters as $filter) { + $filter->setText($this->_preparsed); + $filter->_preparse(); + $this->_preparsed = $filter->getPreparsed(); + } + } + + /** + * Builds the tag array from the input string $_text + * + * An array consisting of tag and text elements is contructed from the + * $_preparsed variable. The method uses _buildTag() to check if a tag is + * valid and to build the actual tag to be added to the tag array. + * + * TODO: - rewrite whole method, as this one is old and probably slow + * - see if a recursive method would be better than an iterative one + * + * @return none + * @access private + * @see _buildTag() + * @see $_text + * @see $_tagArray + * @author Stijn de Reede + */ + function _buildTagArray() + { + $this->_tagArray = array(); + $str = $this->_preparsed; + $strPos = 0; + $strLength = strlen($str); + + while (($strPos < $strLength)) { + $tag = array(); + $openPos = strpos($str, $this->_options['open'], $strPos); + if ($openPos === false) { + $openPos = $strLength; + $nextOpenPos = $strLength; + } + if ($openPos + 1 > $strLength) { + $nextOpenPos = $strLength; + } else { + $nextOpenPos = strpos($str, $this->_options['open'], $openPos + 1); + if ($nextOpenPos === false) { + $nextOpenPos = $strLength; + } + } + $closePos = strpos($str, $this->_options['close'], $strPos); + if ($closePos === false) { + $closePos = $strLength + 1; + } + + if ($openPos == $strPos) { + if (($nextOpenPos < $closePos)) { + // new open tag before closing tag: treat as text + $newPos = $nextOpenPos; + $tag['text'] = substr($str, $strPos, $nextOpenPos - $strPos); + $tag['type'] = 0; + } else { + // possible valid tag + $newPos = $closePos + 1; + $newTag = $this->_buildTag(substr($str, $strPos, $closePos - $strPos + 1)); + if (($newTag !== false)) { + $tag = $newTag; + } else { + // no valid tag after all + $tag['text'] = substr($str, $strPos, $closePos - $strPos + 1); + $tag['type'] = 0; + } + } + } else { + // just text + $newPos = $openPos; + $tag['text'] = substr($str, $strPos, $openPos - $strPos); + $tag['type'] = 0; + } + + // join 2 following text elements + if ($tag['type'] === 0 && isset($prev) && $prev['type'] === 0) { + $tag['text'] = $prev['text'].$tag['text']; + array_pop($this->_tagArray); + } + + $this->_tagArray[] = $tag; + $prev = $tag; + $strPos = $newPos; + } + } + + /** + * Builds a tag from the input string + * + * This method builds a tag array based on the string it got as an + * argument. If the tag is invalid, is returned. The tag + * attributes are extracted from the string and stored in the tag + * array as an associative array. + * + * @param string string to build tag from + * @return array tag in array format + * @access private + * @see _buildTagArray() + * @author Stijn de Reede + */ + function _buildTag($str) + { + $tag = array('text' => $str, 'attributes' => array()); + + if (substr($str, 1, 1) == '/') { // closing tag + + $tag['tag'] = strtolower(substr($str, 2, strlen($str) - 3)); + if (!in_array($tag['tag'], array_keys($this->_definedTags))) { + return false; // nope, it's not valid + } else { + $tag['type'] = 2; + return $tag; + } + } else { // opening tag + + $tag['type'] = 1; + if (strpos($str, ' ') && (strpos($str, '=') === false)) { + return false; // nope, it's not valid + } + + // tnx to Onno for the regex + // split the tag with arguments and all + $oe = $this->_options['open_esc']; + $ce = $this->_options['close_esc']; + $tagArray = array(); + if (preg_match("!$oe([a-z0-9]+)[^$ce]*$ce!i", $str, $tagArray) == 0) { + return false; + } + $tag['tag'] = strtolower($tagArray[1]); + if (!in_array($tag['tag'], array_keys($this->_definedTags))) { + return false; // nope, it's not valid + } + + // tnx to Onno for the regex + // validate the arguments + $attributeArray = array(); + $regex = "![\s$oe]([a-z0-9]+)=(\"[^\s$ce]+\"|[^\s$ce]"; + if ($tag['tag'] != 'url') { + $regex .= "[^=]"; + } + $regex .= "+)(?=[\s$ce])!i"; + preg_match_all($regex, $str, $attributeArray, PREG_SET_ORDER); + foreach ($attributeArray as $attribute) { + $attNam = strtolower($attribute[1]); + if (in_array($attNam, array_keys($this->_definedTags[$tag['tag']]['attributes']))) { + if ($attribute[2][0] == '"' && $attribute[2][strlen($attribute[2])-1] == '"') { + $tag['attributes'][$attNam] = substr($attribute[2], 1, -1); + } else { + $tag['attributes'][$attNam] = $attribute[2]; + } + } + } + return $tag; + } + } + + /** + * Validates the tag array, regarding the allowed tags + * + * While looping through the tag array, two following text tags are + * joined, and it is checked that the tag is allowed inside the + * last opened tag. + * By remembering what tags have been opened it is checked that + * there is correct (xml compliant) nesting. + * In the end all still opened tags are closed. + * + * @return none + * @access private + * @see _isAllowed() + * @see $_tagArray + * @author Stijn de Reede , Seth Price + */ + function _validateTagArray() + { + $newTagArray = array(); + $openTags = array(); + foreach ($this->_tagArray as $tag) { + $prevTag = end($newTagArray); + switch ($tag['type']) { + case 0: + if (($child = $this->_childNeeded(end($openTags), 'text')) && + $child !== false && + /* + * No idea what to do in this case: A child is needed, but + * no valid one is returned. We'll ignore it here and live + * with it until someone reports a valid bug. + */ + $child !== true ) + { + if (trim($tag['text']) == '') { + //just an empty indentation or newline without value? + continue; + } + $newTagArray[] = $child; + $openTags[] = $child['tag']; + } + if ($prevTag['type'] === 0) { + $tag['text'] = $prevTag['text'].$tag['text']; + array_pop($newTagArray); + } + $newTagArray[] = $tag; + break; + + case 1: + if (!$this->_isAllowed(end($openTags), $tag['tag']) || + ($parent = $this->_parentNeeded(end($openTags), $tag['tag'])) === true || + ($child = $this->_childNeeded(end($openTags), $tag['tag'])) === true) { + $tag['type'] = 0; + if ($prevTag['type'] === 0) { + $tag['text'] = $prevTag['text'].$tag['text']; + array_pop($newTagArray); + } + } else { + if ($parent) { + /* + * Avoid use of parent if we can help it. If we are + * trying to insert a new parent, but the current tag is + * the same as the previous tag, then assume that the + * previous tag structure is valid, and add this tag as + * a sibling. To add as a sibling, we need to close the + * current tag. + */ + if ($tag['tag'] == end($openTags)){ + $newTagArray[] = $this->_buildTag('[/'.$tag['tag'].']'); + array_pop($openTags); + } else { + $newTagArray[] = $parent; + $openTags[] = $parent['tag']; + } + } + if ($child) { + $newTagArray[] = $child; + $openTags[] = $child['tag']; + } + $openTags[] = $tag['tag']; + } + $newTagArray[] = $tag; + break; + + case 2: + if (($tag['tag'] == end($openTags) || $this->_isAllowed(end($openTags), $tag['tag']))) { + if (in_array($tag['tag'], $openTags)) { + $tmpOpenTags = array(); + while (end($openTags) != $tag['tag']) { + $newTagArray[] = $this->_buildTag('[/'.end($openTags).']'); + $tmpOpenTags[] = end($openTags); + array_pop($openTags); + } + $newTagArray[] = $tag; + array_pop($openTags); + /* why is this here? it just seems to break things + * (nested lists where closing tags need to be + * generated) + while (end($tmpOpenTags)) { + $tmpTag = $this->_buildTag('['.end($tmpOpenTags).']'); + $newTagArray[] = $tmpTag; + $openTags[] = $tmpTag['tag']; + array_pop($tmpOpenTags); + }*/ + } + } else { + $tag['type'] = 0; + if ($prevTag['type'] === 0) { + $tag['text'] = $prevTag['text'].$tag['text']; + array_pop($newTagArray); + } + $newTagArray[] = $tag; + } + break; + } + } + while (end($openTags)) { + $newTagArray[] = $this->_buildTag('[/'.end($openTags).']'); + array_pop($openTags); + } + $this->_tagArray = $newTagArray; + } + + /** + * Checks to see if a parent is needed + * + * Checks to see if the current $in tag has an appropriate parent. If it + * does, then it returns false. If a parent is needed, then it returns the + * first tag in the list to add to the stack. + * + * @param array tag that is on the outside + * @param array tag that is on the inside + * @return boolean false if not needed, tag if needed, true if out + * of our minds + * @access private + * @see _validateTagArray() + * @author Seth Price + */ + function _parentNeeded($out, $in) + { + if (!isset($this->_definedTags[$in]['parent']) || + ($this->_definedTags[$in]['parent'] == 'all') + ) { + return false; + } + + $ar = explode('^', $this->_definedTags[$in]['parent']); + $tags = explode(',', $ar[1]); + if ($ar[0] == 'none'){ + if ($out && in_array($out, $tags)) { + return false; + } + //Create a tag from the first one on the list + return $this->_buildTag('['.$tags[0].']'); + } + if ($ar[0] == 'all' && $out && !in_array($out, $tags)) { + return false; + } + // Tag is needed, we don't know which one. We could make something up, + // but it would be so random, I think that it would be worthless. + return true; + } + + /** + * Checks to see if a child is needed + * + * Checks to see if the current $out tag has an appropriate child. If it + * does, then it returns false. If a child is needed, then it returns the + * first tag in the list to add to the stack. + * + * @param array tag that is on the outside + * @param array tag that is on the inside + * @return boolean false if not needed, tag if needed, true if out + * of our minds + * @access private + * @see _validateTagArray() + * @author Seth Price + */ + function _childNeeded($out, $in) + { + if (!isset($this->_definedTags[$out]['child']) || + ($this->_definedTags[$out]['child'] == 'all') + ) { + return false; + } + + $ar = explode('^', $this->_definedTags[$out]['child']); + $tags = explode(',', $ar[1]); + if ($ar[0] == 'none'){ + if ($in && in_array($in, $tags)) { + return false; + } + //Create a tag from the first one on the list + return $this->_buildTag('['.$tags[0].']'); + } + if ($ar[0] == 'all' && $in && !in_array($in, $tags)) { + return false; + } + // Tag is needed, we don't know which one. We could make something up, + // but it would be so random, I think that it would be worthless. + return true; + } + + /** + * Checks to see if a tag is allowed inside another tag + * + * The allowed tags are extracted from the private _definedTags array. + * + * @param array tag that is on the outside + * @param array tag that is on the inside + * @return boolean return true if the tag is allowed, false + * otherwise + * @access private + * @see _validateTagArray() + * @author Stijn de Reede + */ + function _isAllowed($out, $in) + { + if (!$out || ($this->_definedTags[$out]['allowed'] == 'all')) { + return true; + } + if ($this->_definedTags[$out]['allowed'] == 'none') { + return false; + } + + $ar = explode('^', $this->_definedTags[$out]['allowed']); + $tags = explode(',', $ar[1]); + if ($ar[0] == 'none' && in_array($in, $tags)) { + return true; + } + if ($ar[0] == 'all' && in_array($in, $tags)) { + return false; + } + return false; + } + + /** + * Builds a parsed string based on the tag array + * + * The correct html and attribute values are extracted from the private + * _definedTags array. + * + * @return none + * @access private + * @see $_tagArray + * @see $_parsed + * @author Stijn de Reede + */ + function _buildParsedString() + { + $this->_parsed = ''; + foreach ($this->_tagArray as $tag) { + switch ($tag['type']) { + + // just text + case 0: + $this->_parsed .= $tag['text']; + break; + + // opening tag + case 1: + $this->_parsed .= '<'.$this->_definedTags[$tag['tag']]['htmlopen']; + if ($this->_options['quotestyle'] == 'single') $q = "'"; + if ($this->_options['quotestyle'] == 'double') $q = '"'; + foreach ($tag['attributes'] as $a => $v) { + //prevent XSS attacks. IMHO this is not enough, though... + //@see http://pear.php.net/bugs/bug.php?id=5609 + $v = preg_replace('#(script|about|applet|activex|chrome):#is', "\\1:", $v); + $v = htmlspecialchars($v); + $v = str_replace('&amp;', '&', $v); + + if (($this->_options['quotewhat'] == 'nothing') || + (($this->_options['quotewhat'] == 'strings') && is_numeric($v)) + ) { + $this->_parsed .= ' '.sprintf($this->_definedTags[$tag['tag']]['attributes'][$a], $v, ''); + } else { + $this->_parsed .= ' '.sprintf($this->_definedTags[$tag['tag']]['attributes'][$a], $v, $q); + } + } + if ($this->_definedTags[$tag['tag']]['htmlclose'] == '' && $this->_options['xmlclose']) { + $this->_parsed .= ' /'; + } + $this->_parsed .= '>'; + break; + + // closing tag + case 2: + if ($this->_definedTags[$tag['tag']]['htmlclose'] != '') { + $this->_parsed .= '_definedTags[$tag['tag']]['htmlclose'].'>'; + } + break; + } + } + } + + /** + * Sets text in the object to be parsed + * + * @param string the text to set in the object + * @return none + * @access public + * @see getText() + * @see $_text + * @author Stijn de Reede + */ + function setText($str) + { + $this->_text = $str; + } + + /** + * Gets the unparsed text from the object + * + * @return string the text set in the object + * @access public + * @see setText() + * @see $_text + * @author Stijn de Reede + */ + function getText() + { + return $this->_text; + } + + /** + * Gets the preparsed text from the object + * + * @return string the text set in the object + * @access public + * @see _preparse() + * @see $_preparsed + * @author Stijn de Reede + */ + function getPreparsed() + { + return $this->_preparsed; + } + + /** + * Gets the parsed text from the object + * + * @return string the parsed text set in the object + * @access public + * @see parse() + * @see $_parsed + * @author Stijn de Reede + */ + function getParsed() + { + return $this->_parsed; + } + + /** + * Parses the text set in the object + * + * @return none + * @access public + * @see _preparse() + * @see _buildTagArray() + * @see _validateTagArray() + * @see _buildParsedString() + * @author Stijn de Reede + */ + function parse() + { + $this->_preparse(); + $this->_buildTagArray(); + $this->_validateTagArray(); + $this->_buildParsedString(); + } + + /** + * Quick method to do setText(), parse() and getParsed at once + * + * @return none + * @access public + * @see parse() + * @see $_text + * @author Stijn de Reede + */ + function qparse($str) + { + $this->_text = $str; + $this->parse(); + return $this->_parsed; + } + + /** + * Quick static method to do setText(), parse() and getParsed at once + * + * @return none + * @access public + * @see parse() + * @see $_text + * @author Stijn de Reede + */ + function staticQparse($str) + { + $p = new SSHTMLBBCodeParser(); + $str = $p->qparse($str); + unset($p); + return $str; + } +} +?> \ No newline at end of file diff --git a/parsers/TextParser.php b/parsers/TextParser.php new file mode 100644 index 000000000..f18ee002c --- /dev/null +++ b/parsers/TextParser.php @@ -0,0 +1,51 @@ +content and parse it however you want. For an example + * of the implementation, @see BBCodeParser. + * + * Your sub-class will be initialized with a string of text, then parse() will be called. + * parse() should (after processing) return the formatted string. + * + * Note: $this->content will have NO conversions applied to it. + * You should run Covert::raw2xml or whatever is appropriate before using it. + * + * Optionally (but recommended), is creating a static usable_tags method, + * which will return a DataObjectSet of all the usable tags that can be parsed. + * This will (mostly) be used to create helper blocks - telling users what things will be parsed. + * Again, @see BBCodeParser for an example of the syntax + * + * TODO Define a proper syntax for (or refactor) usable_tags that can be extended as needed. + */ +abstract class TextParser extends Object { + protected $content; + + /** + * Creates a new TextParser object. + * + * @param string $content The contents of the dbfield + */ + function __construct($content = "") { + $this->content = $content; + } + + /** + * Convenience method, shouldn't really be used, but it's here if you want it + */ + function setContent($content = "") { + $this->content = $content; + } + + /** + * Define your own parse method to parse $this->content appropriately. + * See the class doc-block for more implementation details. + */ + abstract function parse(); +} +?> \ No newline at end of file