• Don't indent paragraphs
• Words wrap and fill as needed
• Use blank lines as separators
• Four or more minus signs make a horizontal rule
• %%% or \\\\ makes a linebreak (in headings and lists too)

Lists

• asterisk for first level
  • asterisk-asterisk for second level, etc.
• Use * for bullet lists, # for numbered lists (mix at will)
• semicolon-term-colon-definition for definition lists:

Term Definition

term here

definition here, as in the <DL><DT><DD> list

• One line for each item
• Other leading whitespace signals preformatted text, changes font.

Headings

• '!' at the start of a line makes a small heading
• '!!' at the start of a line makes a medium heading
• '!!!' at the start of a line makes a large heading

Fonts

• Indent with one or more spaces to use a monospace font;
• Or enclose within triple braces {{{ }}}, which may cross multiple lines:

This is in monospace

This is not

Indented Paragraphs

• semicolon-colon -- works like <BLOCKQUOTE>, but is actually implemented as Term Definition above without the term

this is an indented block of text

Emphasis

• Use doubled single-quotes ('') for emphasis (usually italics)
• Use doubled underscores (__) for strong emphasis (usually bold)
• Mix them at will: bold italics
• Emphasis can be used multiple times within a line, but cannot cross line boundaries:

''this
will not work''

References

• Hyperlinks to other pages within the Wiki are made by placing the page name in square brackets: [this is a page link]
• Hyperlinks to external pages are done like this: [http://www.wcsb.org/]
• You can name the links by providing a name, a bar (|) and then the hyperlink or pagename:
  [PhpWiki home page | http://phpwiki.sourceforge.net/]
• You can suppress linking to old-style references and URIs by preceding the word with a '!', e.g. NotLinkedAsWikiName, http://not.linked.to/
• You can create footnotes by using [1], [2], [3], ... like this here ^[1]. See footnote for counterpart. (If the [ is in the first column, it is a footnote definition rather than a footnote reference ^[1].)
• Also, the old way of linking URL's is still supported: precede URLs with "http:", "ftp:" or "mailto:" to create links automatically as in: http://c2.com/
• URLs ending with .png, .gif, or .jpg are inlined if in square brackets, by themselves: []

HTML Mark-Up Language

• Don't bother
• < and > are themselves
• The & characters will not work EOS1; switch (variable_get("wikitext_filter_html", FILTER_HTML_DISABLE)) { case FILTER_HTML_ENABLE: $output .= <<If you really must use HTML, the system administrator has enabled this feature. Start each line with a bar-greater-than (|>). EOS1b; break; } $output .= <<

  Content protection

  • To protect content as is (i.e., in a <pre></pre> block) without interpretation, enclose within triple braces {{{ }}}, which may cross multiple lines. Note that the content within the triple braces may still be interpreted by the browser in the same way <pre></pre> enclosed content is.

  Table is supported

  • Start the line with single-bar (|) or double-bars (||) to create a table
  • One line per row, do not insert blank line, or the table ends
  • Double-bar separates header (<th></th>) cells
  • Single-bar separates normal table (<td></td>) cells
  • Don't end a line with single or double bar; otherwise, a blank cell results
  • Example:
    

    || Heading 1 || Heading 2
    | Cell a1 | Cell a2
    | Cell b1 | Cell b2

    Result:
    

    Heading 1	Heading 2
    Cell a1	Cell a2
    Cell b1	Cell b2

  More detail than you want to know

  Footnotes:

  
  [1]+ By using [1] a second time (in the first column) the footnote itself is defined. You may refer to a footnote as many times as you want, but you may only define it once on the page. Note the the [1] in the footnote links back to the first reference, if there are multiple references there will be +'s after the [1] which will link to the other references. (References which come after the footnote definition will not be linked to.)

  EOS2; return t($output); } else { return t('WikiText is converted to HTML (supported WikiText formatting will show in the long tip format).'); } } /** * @name WikiText filters * @{ * Filters implemented by the filter.module. */ /** * Implementation of hook_filter(). Contains a basic set of essential filters. * - WikiText filter: * Processes wiki text and convert to HTML */ function wikitext_filter($op, $delta = 0, $format = -1, $text = '') { switch ($op) { case 'list': return array(t('WikiText filter')); case 'description': return t('Converts WikiText to HTML.'); case 'process': return _wikitext_filter($text); case 'settings': return _wikitext_filter_settings($format); default: return $text; } } /** * Settings for the WikiText filter. */ function _wikitext_filter_settings($format) { $group = form_radios(t('Support Escaped HTML'), "wikitext_filter_html", variable_get("wikitext_filter_html", FILTER_HTML_DISABLE), array(FILTER_HTML_ENABLE => t('Enable HTML'), FILTER_HTML_DISABLE => t('Disable HTML')), t('Whether to allow |>-escaped HTML. If set to "enable", then any HTML goes on each line after the |> escape marker. If set to "disable", then any HTML tags will be shown as is, unless the HTML Filter has been separately enabled. It is recommended that this be disabled. Turn on the HTML Filter to support HTML, but make sure the HTML filter applies after the WikiText filter.')); $output .= form_group(t('WikiText filter'), $group); return $output; } define("ZERO_LEVEL", 0); define("NESTED_LEVEL", 1); define("FIELD_SEPARATOR", "\263"); // allowed protocols for links - be careful not to allow "javascript:" // within a named link [name|uri] one more protocol is defined: phpwiki //define("ALLOWED_PROTOCOLS", "http|https|mailto|ftp|news|gopher"); define("ALLOWED_PROTOCOLS", "http|https|javascript|mailto|ftp|news|gopher"); // URLs ending with the following extension should be inlined as images define("INLINE_IMAGES", "png|jpg|gif"); class _wikitext_stack { var $items = array(); var $size = 0; function push($item) { $this->items[$this->size] = $item; $this->size++; return true; } function pop() { if ($this->size == 0) { return false; // stack is empty } $this->size--; return $this->items[$this->size]; } function cnt() { return $this->size; } function top() { if ($this->size) { return $this->items[$this->size - 1]; } else { return ''; } } } // end class definition /** * Find any strings in $str that match $pattern and * store them in $orig, replacing them with tokens * starting at number $ntokens - returns tokenized string */ function _wikitext_tokenize($str, $pattern, &$orig, &$ntokens) { $new = ''; while (preg_match("/^(.*?)($pattern)/", $str, $matches)) { $linktoken = FIELD_SEPARATOR . FIELD_SEPARATOR . ($ntokens++) . FIELD_SEPARATOR; $new .= $matches[1] . $linktoken; $orig[] = $matches[2]; $str = substr($str, strlen($matches[0])); } $new .= $str; return $new; } function _wikitext_find_node($title) { $result = db_query("SELECT n.*, u.name, u.uid FROM node n LEFT JOIN users u ON n.uid = u.uid WHERE n.title = '%s' AND n.status = 1 ORDER BY created DESC", $title); if (db_num_rows($result) == 0) { // No node with exact title found, try substring. $result = db_query("SELECT n.* FROM node n WHERE n.title LIKE '%". check_query($title). "%' AND n.status = 1 ORDER BY created DESC"); } else if (db_num_rows($result) == 1) { return db_fetch_object($result); } else { return FALSE; } } function _wikitext_link_unknown($wikiword, $linktext='') { $enc_word = rawurlencode($wikiword); if (empty($linktext)) { $linktext = htmlspecialchars($wikiword); } // this assumed blogger //return "$linktext?"; // this assumes book AND not clean URL //return "$linktext?"; // this assumes book AND clean URL return "$linktext?"; } function _wikitext_link_URL($url, $linktext='') { if (ereg("[<>\"]", $url)) { return "BAD URL -- remove all of <, >, ""; } if (empty($linktext)) { $linktext = htmlspecialchars($url); } return "$linktext"; } function _wikitext_link_image($url, $alt='[External Image]') { if (ereg('[<>"]', $url)) { return "BAD URL -- remove all of <, >, ""; } return ""; } function _wikitext_link_node($node, $linktext='') { if (empty($linktext)) { $linktext = htmlspecialchars($node->title); } // this assumed clean URL //return "nid}\">$linktext"; // this assumes book AND not clean URL //return "nid}\">$linktext"; // this assumes book AND clean URL return "nid}\">$linktext"; } /** * $bracketlink will start and end with brackets; in between * will be either a page name, a URL or both separated by a pipe. */ function _wikitext_parse_and_link($bracketlink) { // strip brackets and leading space preg_match("/(\[\s*)(.+?)(\s*\])/", $bracketlink, $match); // match the contents preg_match("/([^|]+)(\|)?([^|]+)?/", $match[2], $matches); if (isset($matches[3])) { // named link of the form "[some link name | http://blippy.com/]" $URL = trim($matches[3]); $linkname = htmlspecialchars(trim($matches[1])); $linktype = 'named'; } else { // unnamed link of the form "[http://blippy.com/] or [wiki page]" $URL = trim($matches[1]); $linkname = ''; $linktype = 'simple'; } if ($node = _wikitext_find_node($URL)) { $link['type'] = "wiki-$linktype"; $link['link'] = _wikitext_link_node($node, $linkname); } else if (preg_match("/".INLINE_IMAGES."$/i", $URL)) { // if it's an image, embed it; otherwise, it's a regular link $link['type'] = "image-$linktype"; $link['link'] = _wikitext_link_image($URL, $linkname); } else if (preg_match("#^".ALLOWED_PROTOCOLS.":#", $URL)) { $link['type'] = "url-$linktype"; $link['link'] = _wikitext_link_URL($URL, $linkname); } else if (preg_match("#^phpwiki:(.*)#", $URL, $match)) { $link['type'] = "url-wiki-$linktype"; if (empty($linkname)) { $linkname = htmlspecialchars($URL); } $link['link'] = "$linkname"; } else if (preg_match("#^\d+$#", $URL)) { $link['type'] = "reference-$linktype"; $link['link'] = $URL; } else { $link['type'] = "wiki-unknown-$linktype"; $link['link'] = _wikitext_link_unknown($URL, $linkname); } return $link; } /** * Wiki HTML output can, at any given time, be in only one mode. * It will be something like Unordered List, Preformatted Text, * plain text etc. When we change modes we have to issue close tags * for one mode and start tags for another. * * @param $tag HTML tag to insert * @param $tagtype ZERO_LEVEL - close all open tags before inserting $tag * NESTED_LEVEL - close tags until depths match * @param $level nesting level (depth) of $tag * nesting is arbitrary limited to 10 levels */ function _wikitext_set_output_mode($tag, $tagtype, $level) { static $stack; if (!$stack) { $stack = new _wikitext_stack; } $retvar = ''; if ($level > 10) { /* * arbitrarily limit tag nesting * We just clamp the the maximum depth. */ $level = 10; } if ($tagtype == ZERO_LEVEL) { // empty the stack until $level == 0; if ($tag == $stack->top()) { return; // same tag? -> nothing to do } while ($stack->cnt() > 0) { $closetag = $stack->pop(); $retvar .= "\n"; } if ($tag) { $retvar .= "<$tag>\n"; $stack->push($tag); } } else if ($tagtype == NESTED_LEVEL) { if ($level < $stack->cnt()) { // $tag has fewer nestings (old: tabs) than stack, // reduce stack to that tab count while ($stack->cnt() > $level) { $closetag = $stack->pop(); if ($closetag == FALSE) { //echo "bounds error in tag stack"; break; } $retvar .= "\n"; } // if list type isn't the same, // back up one more and push new tag if ($tag != $stack->top()) { $closetag = $stack->pop(); $retvar .= "<$tag>\n"; $stack->push($tag); } } else if ($level > $stack->cnt()) { // Test for and close top level elements which are not allowed to contain // other block-level elements. if ($stack->cnt() == 1 and preg_match('/^(p|pre|h\d)$/i', $stack->top())) { $closetag = $stack->pop(); $retvar .= ""; } // we add the diff to the stack // stack might be zero if ($stack->cnt() < $level) { while ($stack->cnt() < $level - 1) { /* * This is a bit of a hack: * * We're not nested deep enough, and have to make up * some kind of block element to nest within. * * Currently, this can only happen for nested list * element (either
  or

  ). What we used * to do here is to open extra lists of whatever * type was requested. This would result in invalid * HTML, since and list is not allowed to contain * another list without first containing a list * item. ("

  • Item

  " is invalid.) * * So now, when we need extra list elements, we use * a

  , and open it with an empty

  . */ $retvar .= "

  "; $stack->push('dl'); } $retvar .= "<$tag>\n"; $stack->push($tag); } } else { // $level == $stack->cnt() if ($tag == $stack->top()) { return; // same tag? -> nothing to do } else { // different tag - close old one, add new one $closetag = $stack->pop(); $retvar .= "\n"; $retvar .= "<$tag>\n"; $stack->push($tag); } } } else { // unknown $tagtype ExitWiki ("Passed bad tag type value in _wikitext_set_output_mode"); } return $retvar; } // end _wikitext_set_output_mode() /** * WikiText filter. Provides filtering of input into accepted HTML. */ function _wikitext_filter($text) { $html = ''; $isprotect = FALSE; // whether we're within a {{{ }}} protection block $istable = FALSE; // whether we're currently in a table // process text line by line $lines = preg_split('/\n/', $text); $numlines = count($lines); for ($index = 0; $index < $numlines; $index++) { unset($tokens); unset($replacements); $ntokens = 0; $replacements = array(); $tmpline = $lines[$index]; /* protect everthing within "{{{" and "}}}" as is */ // multi line case: if (preg_match("/^{{{(.*)/", $tmpline, $matches) && ! preg_match("/}}}/", $tmpline, $matches)) { $isprotect = TRUE; $html .= _wikitext_set_output_mode("pre", ZERO_LEVEL, 0); $html .= $matches[1]; continue; } else if ($isprotect) { if (preg_match("/^}}}/", $tmpline, $matches)) { $isprotect = FALSE; $html .= _wikitext_set_output_mode("", ZERO_LEVEL, 0); } else { $html .= $tmpline; } continue; } // single line case: $oldn = $ntokens; $tmpline = _wikitext_tokenize($tmpline, '{{{.+}}}', $replacements, $ntokens); while ($oldn < $ntokens) { if (preg_match("/{{{(.+)}}}/", $replacements[$oldn], $matches)) { // this, of course, goes through, but we want the matched piece $matches[1] = str_replace('\\\\', '
  ', $matches[1]); $replacements[$oldn] = preg_replace("{{{.+}}}", "

  $matches[1]

  ", $replacements[$oldn]); } $oldn++; } if (!strlen($tmpline) || $tmpline == "\r") { // this is a blank line, send

  $html .= _wikitext_set_output_mode('', ZERO_LEVEL, 0); continue; } else /* phpwiki-way of protecting HTML */ // if (preg_match("/(^\|)(.*)/", $tmpline, $matches)) { // this change is to prevent wiki table from being skipped if (preg_match("/(^\|>)(.*)/", $tmpline, $matches)) { // HTML mode $html .= _wikitext_set_output_mode("", ZERO_LEVEL, 0); $html .= $matches[2]; continue; } /**/ $tmpline = _wikitext_tokenize($tmpline, '<[^>]+>', $replacements, $ntokens); /* * New linking scheme: links are in brackets. This will * emulate typical HTML linking as well as Wiki linking. */ // First need to protect [[. $oldn = $ntokens; $tmpline = _wikitext_tokenize($tmpline, '\[\[', $replacements, $ntokens); while ($oldn < $ntokens) { $replacements[$oldn++] = '['; } // Now process the [\d+] links which are numeric references $oldn = $ntokens; $tmpline = _wikitext_tokenize($tmpline, '\[\s*\d+\s*\]', $replacements, $ntokens); while ($oldn < $ntokens) { $num = (int) substr($replacements[$oldn], 1); if (! empty($embedded[$num])) { $replacements[$oldn] = $embedded[$num]; } $oldn++; } // match anything else between brackets $oldn = $ntokens; $tmpline = _wikitext_tokenize($tmpline, '\[.+?\]', $replacements, $ntokens); while ($oldn < $ntokens) { $link = _wikitext_parse_and_link($replacements[$oldn]); $replacements[$oldn] = $link['link']; $oldn++; } /* * replace all URL's with tokens, so we don't confuse them * with Wiki words later. Wiki words in URL's break things. * URLs preceeded by a '!' are not linked */ $tmpline = _wikitext_tokenize($tmpline, "!?\b(".ALLOWED_PROTOCOLS."):[^\s<>\[\]\"'()]*[^\s<>\[\]\"'(),.?]", $replacements, $ntokens); while ($oldn < $ntokens) { if($replacements[$oldn][0] == '!') { $replacements[$oldn] = substr($replacements[$oldn], 1); } else { $replacements[$oldn] = _wikitext_link_URL($replacements[$oldn]); } $oldn++; } //escape HTML metachars // $tmpline = str_replace('&', '&', $tmpline); $tmpline = str_replace('>', '>', $tmpline); $tmpline = str_replace('<', '<', $tmpline); // %%% are linebreaks $tmpline = str_replace('%%%', '
  ', $tmpline); // \\ are also linebreaks $tmpline = str_replace('\\\\', '
  ', $tmpline); // bold italics (old way) $tmpline = preg_replace("|(''''')(.*?)(''''')|", "\\2", $tmpline); // bold (old way) $tmpline = preg_replace("|(''')(.*?)(''')|", "\\2", $tmpline); // bold $tmpline = preg_replace("|(__)(.*?)(__)|", "\\2", $tmpline); // italics $tmpline = preg_replace("|('')(.*?)('')|", "\\2", $tmpline); /* * unordered, ordered, and dictionary list (using TAB) */ if (preg_match("/(^\t+)(.*?)(:\t)(.*$)/", $tmpline, $matches)) { // this is a dictionary list (

  ) item $numtabs = strlen($matches[1]); $html .= _wikitext_set_output_mode('dl', NESTED_LEVEL, $numtabs); $tmpline = ''; if (trim($matches[2])) { $tmpline = '

  ' . $matches[2]; } $tmpline .= '

  ' . $matches[4]; } else if (preg_match("/(^\t+)(\*|\d+|#)/", $tmpline, $matches)) { // this is part of a list (

  ,
  ) $numtabs = strlen($matches[1]); if ($matches[2] == '*') { $listtag = 'ul'; } else { $listtag = 'ol'; // a rather tacit assumption. oh well. } $tmpline = preg_replace("/^(\t+)(\*|\d+|#)/", "", $tmpline); $html .= _wikitext_set_output_mode($listtag, NESTED_LEVEL, $numtabs); $html .= '
  1. '; /* * tabless markup for unordered, ordered, and dictionary lists * ul/ol list types can be mixed, so we only look at the last * character. Changes e.g. from "**#*" to "###*" go unnoticed. * and wouldn\'t make a difference to the HTML layout anyway. */ // unordered lists
     : "*" } else if (preg_match("/^([#*]*\*)[^#]/", $tmpline, $matches)) { // this is part of an unordered list $numtabs = strlen($matches[1]); $tmpline = preg_replace("/^([#*]*\*)/", '', $tmpline); $html .= _wikitext_set_output_mode('ul', NESTED_LEVEL, $numtabs); $html .= '
     • '; // ordered lists
       : "#" } else if (preg_match("/^([#*]*\#)/", $tmpline, $matches)) { // this is part of an ordered list $numtabs = strlen($matches[1]); $tmpline = preg_replace("/^([#*]*\#)/", "", $tmpline); $html .= _wikitext_set_output_mode('ol', NESTED_LEVEL, $numtabs); $html .= '
       1. '; // definition lists

          : ";text:text" } else if (preg_match("/(^;+)(.*?):(.*$)/", $tmpline, $matches)) { // this is a dictionary list item $numtabs = strlen($matches[1]); $html .= _wikitext_set_output_mode('dl', NESTED_LEVEL, $numtabs); $tmpline = ''; if (trim($matches[2])) { $tmpline = '

          ' . $matches[2]; } $tmpline .= '

          ' . $matches[3]; // remaining modes: preformatted text, headings, normal text } else if (preg_match("/^\s+/", $tmpline)) { // this is preformatted text, i.e.

                $html .= _wikitext_set_output_mode('pre', ZERO_LEVEL, 0);
          
              } else if (preg_match("/^(!{1,3})[^!]/", $tmpline, $whichheading)) {
                // lines starting with !,!!,!!! are headings
                if ($whichheading[1] == '!') {
          	 $heading = 'h3';
                } else if ($whichheading[1] == '!!') {
          	 $heading = 'h2';
                } else if ($whichheading[1] == '!!!') {
          	 $heading = 'h1';
                }
                $tmpline = preg_replace("/^!+/", '', $tmpline);
                $html .= _wikitext_set_output_mode($heading, ZERO_LEVEL, 0);
              } else if (preg_match('/^-{4,}\s*(.*?)\s*$/', $tmpline, $matches)) {
                // four or more dashes to 
          
                // 
           can not be contained in a
                $html .= _wikitext_set_output_mode('', ZERO_LEVEL, 0) . "
          \n";
                if ( ($tmpline = $matches[1]) != '' ) {
          	 $html .= _wikitext_set_output_mode('p', ZERO_LEVEL, 0);
                }
              } else if (preg_match('/^\|/', $tmpline, $matches)) {
                // table!
                if (!$istable) {
          	 $istable = TRUE;
          	 $html .= _wikitext_set_output_mode('', ZERO_LEVEL, 0);
          	 // unfortunately, can't use _wikitext_set_output_mode due to add'l params
          	 $html .= "\n";
                }
                if (preg_match('/^\|\|/', $tmpline, $matches)) {
          	 // first do heading row
          	 $tmpline = preg_replace("/^\|\|\s*/", "" . $tmpline . "";
                } else {
          	 // do normal row
          	 $tmpline = preg_replace("/^\|\s*/", "" . $tmpline . "";
                }
              } else {
                // it's ordinary output if nothing else
                $html .= _wikitext_set_output_mode('p', ZERO_LEVEL, 0);
              }
          
              /*
               * Replace tokens
               */
              for ($i = 0; $i < $ntokens; $i++) {
                $tmpline = str_replace(FIELD_SEPARATOR.FIELD_SEPARATOR.$i.FIELD_SEPARATOR, $replacements[$i], $tmpline);
              }                           
              $html .= $tmpline . "\n";
          
              /* look ahead one row to determine if we should close off the table
               * doing it this way prevents having to disperse table closing
               * in several different parts of this function
               */
              if ($istable && !preg_match('/^\|/', $lines[$index+1], $matches)) {
                // turn table off
                $istable = FALSE;
                $html .= "
          ", $tmpline);
          	 $tmpline = preg_replace("/\s*\|\|\s*/", "
          ", $tmpline);
          	 $tmpline = "
          ", $tmpline);
          	 $tmpline = preg_replace("/\s*\|\s*/", "
          ", $tmpline);
          	 $tmpline = "
          
          \n";
              }
            }
            $html .= _wikitext_set_output_mode('', ZERO_LEVEL, 0);
                 
            return $html;
          }
          
          /**
           * @} End of "WikiText filters".
           */
          
          ?>