tags var $block_elements = 'address|blockquote|div|dl|fieldset|form|h\d|hr|noscript|object|ol|p|pre|script|table|ul'; // Elements that should not have

and
tags within them. var $skip_elements = 'p|pre|ol|ul|dl|object|table'; // Tags we want the parser to completely ignore when splitting the string. var $inline_elements = 'a|abbr|acronym|b|bdo|big|br|button|cite|code|del|dfn|em|i|img|ins|input|label|map|kbd|q|samp|select|small|span|strong|sub|sup|textarea|tt|var'; // whether or not to protect quotes within { curly braces } var $protect_braced_quotes = FALSE; /** * Nothing to do here... * */ function CI_Typography() { } /** * Auto Typography * * This function converts text, making it typographically correct: * - Converts double spaces into paragraphs. * - Converts single line breaks into
tags * - Converts single and double quotes into correctly facing curly quote entities. * - Converts three dots into ellipsis. * - Converts double dashes into em-dashes. * - Converts two spaces into entities * * @access public * @param string * @param bool whether to reduce more then two consecutive newlines to two * @return string */ function auto_typography($str, $reduce_linebreaks = FALSE) { if ($str == '') { return ''; } // Standardize Newlines to make matching easier if (strpos($str, "\r") !== FALSE) { $str = str_replace(array("\r\n", "\r"), "\n", $str); } // Reduce line breaks. If there are more than two consecutive linebreaks // we'll compress them down to a maximum of two since there's no benefit to more. if ($reduce_linebreaks === TRUE) { $str = preg_replace("/\n\n+/", "\n\n", $str); } // Convert quotes within tags to temporary markers. We don't want quotes converted // within tags so we'll temporarily convert them to {@DQ} and {@SQ} // and we don't want double dashes converted to emdash entities, so they are marked with {@DD} // likewise double spaces are converted to {@NBS} to prevent entity conversion if (preg_match_all("#\<.+?>#si", $str, $matches)) { for ($i = 0, $total = count($matches[0]); $i < $total; $i++) { $str = str_replace($matches[0][$i], str_replace(array("'",'"','--',' '), array('{@SQ}', '{@DQ}', '{@DD}', '{@NBS}'), $matches[0][$i]), $str); } } if ($this->protect_braced_quotes === TRUE) { if (preg_match_all("#\{.+?}#si", $str, $matches)) { for ($i = 0, $total = count($matches[0]); $i < $total; $i++) { $str = str_replace($matches[0][$i], str_replace(array("'",'"'), array('{@SQ}', '{@DQ}'), $matches[0][$i]), $str); } } } // Convert "ignore" tags to temporary marker. The parser splits out the string at every tag // it encounters. Certain inline tags, like image tags, links, span tags, etc. will be // adversely affected if they are split out so we'll convert the opening bracket < temporarily to: {@TAG} $str = preg_replace("#<(/*)(".$this->inline_elements.")([ >])#i", "{@TAG}\\1\\2\\3", $str); // Split the string at every tag. This expression creates an array with this prototype: // // [array] // { // [0] = // [1] = Content... // [2] = // Etc... // } $chunks = preg_split('/(<(?:[^<>]+(?:"[^"]*"|\'[^\']*\')?)+>)/', $str, -1, PREG_SPLIT_DELIM_CAPTURE|PREG_SPLIT_NO_EMPTY); // Build our finalized string. We cycle through the array, skipping tags, and processing the contained text $str = ''; $process = TRUE; $paragraph = FALSE; foreach ($chunks as $chunk) { // Are we dealing with a tag? If so, we'll skip the processing for this cycle. // Well also set the "process" flag which allows us to skip

 tags and a few other things.
			if (preg_match("#<(/*)(".$this->block_elements.").*?>#", $chunk, $match))
			{
				if (preg_match("#".$this->skip_elements."#", $match[2]))
				{
					$process =  ($match[1] == '/') ? TRUE : FALSE;
				}
				
				$str .= $chunk;
				continue;
			}
			elseif (preg_match('/<(\/?)([a-z]*).*?>/s', $chunk, $tagmatch))
			{
				if ($tagmatch[1] == '/' && $tagmatch[2] == $this->last_tag)
				{
					$process = FALSE;
				}
				else
				{
					$process = TRUE;
					$this->last_tag = $tagmatch[2];					
				}
			}

			if ($process == FALSE)
			{
				$str .= $chunk;
				continue;
			}
			
			//  Convert Newlines into 

and
tags $str .= $this->format_characters($this->_format_newlines($chunk)); } // is the whole of the content inside a block level element? if ( ! preg_match("/^<(?:".$this->block_elements.")/i", $str, $match)) { $str = "

{$str}

"; } // some special linebreak cleanup $str = preg_replace_callback('#<(?!/|'.$this->block_elements.')([^>]*)>

(.*?)

<(\w*)#si', array($this, '_linebreak_cleanup'), $str); // and cleanup empty paragraph tags sitting between two closing tags $str = preg_replace('#()

(\s*)

()#si', '$1$2$3', $str); // Final clean up $table = array( // If the user submitted their own paragraph tags within the text // we will retain them instead of using our tags. '/(*?]>)

/' => '$1', // )+#' => '

', '/(

\W*

)+/' => '

', // Clean up stray paragraph tags that appear before block level elements '#

<('.$this->block_elements.')#' => '<$1', // Clean up open paragraph tags that appear before block level elements '#

(\W)<('.$this->block_elements.')#' => '

$1<$2', // Clean up stray non-breaking spaces preceeding block elements '#[  ]+<('.$this->block_elements.')#' => ' <$1', // Replace the temporary markers we added earlier '/\{@TAG\}/' => '<', '/\{@DQ\}/' => '"', '/\{@SQ\}/' => "'", '/\{@DD\}/' => '--', '/\{@NBS\}/' => ' ' ); // Do we need to reduce empty lines? if ($reduce_linebreaks === TRUE) { $table['#

\n*

#'] = ''; } else { // If we have empty paragraph tags we add a non-breaking space // otherwise most browsers won't treat them as true paragraphs $table['#

#'] = '

 

'; } return preg_replace(array_keys($table), $table, $str); } // -------------------------------------------------------------------- /** * Linebreak Cleanup * * Removes paragraph and line break tags inserted inbetween * inline content and a new opening block level element * * @access private * @param array * @return string */ function _linebreak_cleanup($match) { if (in_array($match[3], explode('|', $this->block_elements))) { return "<{$match[1]}>".str_replace('
', '', $match[2])."<{$match[3]}"; } else { return $match[0]; } } // -------------------------------------------------------------------- /** * Format Characters * * This function mainly converts double and single quotes * to curly entities, but it also converts em-dashes, * double spaces, and ampersands * * @access public * @param string * @return string */ function format_characters($str) { static $table; if ( ! isset($table)) { $table = array( // nested smart quotes, opening and closing // note that rules for grammar (English) allow only for two levels deep // and that single quotes are _supposed_ to always be on the outside // but we'll accommodate both '/(^|\W|\s)\'"/' => '$1‘“', '/\'"(\s|\W|$)/' => '’”$1', '/(^|\W|\s)"\'/' => '$1“‘', '/"\'(\s|\W|$)/' => '”’$1', // single quote smart quotes '/\'(\s|\W|$)/' => '’$1', '/(^|\W|\s)\'/' => '$1‘', // double quote smart quotes '/"(\s|\W|$)/' => '”$1', '/(^|\W|\s)"/' => '$1“', // apostrophes "/(\w)'(\w)/" => '$1’$2', // Em dash and ellipses dots '/\s?\-\-\s?/' => '—', '/(\w)\.{3}/' => '$1…', // double space after sentences '/(\W) /' => '$1  ', // ampersands, if not a character entity '/&(?!#?[a-zA-Z0-9]{2,};)/' => '&' ); } return preg_replace(array_keys($table), $table, $str); } // -------------------------------------------------------------------- /** * Format Newlines * * Converts newline characters into either

tags or
* * @access public * @param string * @return string */ function _format_newlines($str) { if ($str == '') { return $str; } if (strpos($str, "\n") === FALSE) { return $str; } // Convert two consecutive newlines to paragraphs $str = str_replace("\n\n", "

\n\n

", $str); // Convert single spaces to
tags $str = preg_replace("/([^\n])(\n)([^\n])/", "\\1
\\2\\3", $str); // Wrap the whole enchilada in enclosing paragraphs if ($str != "\n") { $str = '

'.$str.'

'; } // Remove empty paragraphs if they are on the first line, as this // is a potential unintended consequence of the previous code $str = preg_replace("/

<\/p>(.*)/", "\\1", $str, 1); return $str; } // ------------------------------------------------------------------------ /** * Convert newlines to HTML line breaks except within PRE tags * * @access public * @param string * @return string */ function nl2br_except_pre($str) { $ex = explode("pre>",$str); $ct = count($ex); $newstr = ""; for ($i = 0; $i < $ct; $i++) { if (($i % 2) == 0) { $newstr .= nl2br($ex[$i]); } else { $newstr .= $ex[$i]; } if ($ct - 1 != $i) $newstr .= "pre>"; } return $newstr; } } // END Typography Class /* End of file Typography.php */ /* Location: ./system/libraries/Typography.php */