Go to indexmenu for the search algorithm(find pages) and to page plugin for page inclusion(combine all to one). I combined both plugins, added some extras and got rid of some smaller bugs.

{{compose>a:b:c#3|nons absolute print numbers numbers_depth}}

{{compose>doku:controller}}

will print the content of all pages in this tree of the DokuWiki, without the namespacenodes (doku:controller , doku:controller:c, doku:controller:d)

doku:controller:a
doku:controller:b
  doku:controller:c:a
  doku:controller:c:b
  doku:controller:d:a
    doku:controller:d:c:b
  doku:controller:d:e

{{compose>doku:controller#3 | print numbers 3}}

will print contents of all pages in doku:controller up till level 3, optimized for printing, and with numbering from headings level 5 to 3

1   doku:controller:a
2   doku:controller:b
3.1   doku:controller:c:a
3.2   doku:controller:c:b
4.1   doku:controller:d:a
        doku:controller:d:c:b - not included since > 3 level
4.2   doku:controller:d:e

It would be very useful to have a downloadable “zip” file which includes a “composer” folder and needed plugin files to conform to other well written plugins and would make it possible to manage with Plugin Manager.

Create syntax.php, containing the php code below, in new plugins/composer/ folder

Should work right after install

Should stop working after remove+cleaned cache(on the pages it was used on)

Please add instructions re manual install here. Exactly where (path) to you put the syntax.php? What effect does it have on cache? Does it work immediately after install or do we have to clear cache? does it stop working right after uninstall/removed or do we have to clear cache? Links to cache instructions would be useful. Thank you.

Doesn't work immediately after install. You have to browse every single page for it to be included.
When cache expires on the browsed pages, it seems you have to browse every single page again for it to be included.
This is probably a bug, unintentional to the design of the plugin.

Move above comment to “bugs & problems” heading if appropriate.

<?php
 
if(!defined('DOKU_INC')) define('DOKU_INC',realpath(dirname(__FILE__).'/../../').'/');
if(!defined('DOKU_PLUGIN')) define('DOKU_PLUGIN',DOKU_INC.'lib/plugins/');
require_once(DOKU_PLUGIN.'syntax.php');
require_once(DOKU_INC.'inc/search.php');
 
/*
 * {{compose> namespace#depth | print numbers how_deep absolute }}
 *
 * print:
 * 		print optimization
 * 	numbers + how_deep:
 * 		numbering for headers(1.1.3 hallo...)
 * 	absolute:
 * 		absolute indentation of headers, corresponding to file-depth in namespace
 * 		or (default) relative indentation
 */
class syntax_plugin_composer extends DokuWiki_Syntax_Plugin {
 
    /**
     * return some info
     */
    function getInfo(){
        return array(
            'author' => 'Michael Grosser',
            'email'  => 'grosser.michael[ät]gmail.com',
            'date'   => '07/09/06',
            'name'   => 'Composer Plugin',
            'desc'   => 'Composes a directory to a single page',
            'url'    => 'http://www.dokuwiki.org/plugin:composer',
            'version'=> '1.1',
        );
    }
 
	/**
	 * how to handle <p>
	 */
  	function getPType(){
  		return 'block';
  	}
 
  	/**
  	 * what other types are allowed iside
  	 */
  	function getAllowedTypes() {
		return array('substition');
  	}
 
    /**
     * What kind of syntax are we?
     */
    function getType(){
        return 'substition';
    }
 
    /**
     * Where to sort in?
     */
    function getSort(){
        return 311;
    }
 
    /**
     * Connect pattern to lexer
     */
    function connectTo($mode) {
      $this->Lexer->addSpecialPattern("{{compose>.*?}}",$mode,'plugin_composer');
    }
 
    /**
     * Handle the match
     */
	function handle($match, $state, $pos, &$handler){
		//defaults
		$level = 0;							//how deep ?
		$nons = true;						//namespacenodes ?
 
		//only match options, no syntax
		$match = substr($match,10,-2);
 
		//split namespace+level | options
		$match = preg_split('/\|/u', $match, 2);
 
		//--------NAMESPACE+LEVEL
		//find namespace & level in namespace+level
		if ( preg_match('/(.*)#(\S*)/u',$match[0],$ns_opt)) {
			$ns=$ns_opt[1];				//inside first brackets
			if (is_numeric($ns_opt[2])){
				$level=$ns_opt[2];			//inside 2nd brackets
			}
		}
		//nothin matched -> original string is only namespace
		else{
			$ns=$match[0];
	   	}
	   	//--------
 
		//--------OPTIONS
	   	//options are sperated by whitespace
		$opts=preg_split('/ /u',$match[1]);
 
		//include namespace nodes ?
		$nons = in_array('nons',$opts);
 
		//absolute indentation ?
		$relative = false;
		if(!in_array('absolute',$opts)){
			$relative = substr_count($ns,":")+1;
		}
 
		//--------NUMBERING ?
		$numbering_depth = 	2;
		$numbers = 			false;
 
		if(in_array('numbers',$opts)) {
			$numbers = true;
 
			$temp = array_search('numbers',$opts)+1;
			//if value after numbers is int
 
			if($temp !== false && isset($opts[$temp])) {
				//depth is value after numbers(if its a int lower 6)
				if(intval($opts[$temp]) < 6) {
					$numbering_depth = intval($opts[$temp]);
				}
			}
		}
		//--------
 
		//--------PRINT OPTIMIZATION ?
		$split_too_long = false;
		if(in_array('print',$opts)) {
			$split_too_long = true;	//split too long lines
		}
		//--------
 
		return array(
			'namespace'=>		$ns,
			'search_options'=>	array('level' => $level,'nons' => $nons),
			'output_options'=>	array(
				'relative'=>		$relative,
				'depth'=>			$level,
				'start_ns'=>		$ns,
				'numbers'=>			$numbers,
				'numbering_depth'=>	$numbering_depth,
				'split_too_long'=>	$split_too_long,
				),
			);
	}
 
    /**
     * Create output
     */
	function render($mode, &$renderer, $data) {
		if($mode == 'xhtml'){
			//for header numbering
			$header_count = array(	1=>0,
									2=>0,
									3=>0,
									4=>0,
									5=>0,
									);
 
			$data['output_options']['header_count']= &$header_count;
 
			//get all files in the supplied namespace, with their level
			$file_data=$this->_get_file_data($data,$renderer);
 
			foreach($file_data as $file) {
				//build the output
				$this->_render_output($renderer,$file,$data['output_options']);
			}
			return true;
		}
		return false;
	}
 
	/*
	 * build the output
	 *
	 *
	 * $data
	 * 	0:
	 * 		namespace (a:b:c)
	 * 	1:
	 * 		level:
	 * 			how deep shall we build the tree ?
	 *
	 * 		nons:
	 * 			no namespace-nodes (aka directories) ?
	 *
	 */
	function _get_file_data($data,&$renderer) {
		global $conf;
		global $ID;
 
		//clean input
		$default = 	array(
			'level'=>0,
			'nons'=>true,
			);
		$options =	array_merge($default,$data['search_options']);
		$ns = 		$data['namespace'];
 
 
		//--------BUILD THE NAMESPACE
		//namespace options . or ..
		if($ns == '.') {
			$ns = dirname(
				//replace / with : in $ID string(current location)
				str_replace(':','/',$ID)
				);
			//still . ?
			if ($ns == '.') $ns = '';
		}
		//normal namespace entered
		else {
			$ns = cleanID($ns);		//Dokuwiki cleaner
		}
 
		//replace : with /
		$ns  = utf8_encodeFN(str_replace(':','/',$ns));
		//--------
 
		//--------FIND THE FILES
		$file_data = array(
						'start_lvl'=>substr_count($ns,"/"),	//start for indentation
						'blocked'=>array(),					//blocked files
						);
 
		//find all matching files
		search($file_data,$conf['datadir'],'composer_search_index',$options,"/".$ns);
		//--------
 
		//clean the file_data from non-files (for safety)
		unset($file_data['start_lvl']);
		unset($file_data['blocked']);
 
		return $file_data;
	}
 
 
	/*
	 * Build includes
	 * options $o:
	 * 	relative:
	 *			output file-tree indentation will be absolute according to file level
	 *			OR relative to level of namespace-level, value of absolute ist level of namespace
	 *			==> false OR 1-999
	 *	start_ns:
	 *			the namespace we are including (ns entered by the user)
	 *	search_depth:
	 *			how deep are we searching ?
	 *
	 */
	function _render_output(&$renderer,$file_data,$o){
		//clean input options
		$default = array(
					'relative'=>		false,
					'start_ns'=>		"",
					'depth'=>			0,
					'header_count'=>	array(),
					'numbers'=>			false,
					'numbering_depth'=>	3,
					'split_too_long'=>	false,
					);
		$o = array_merge($default,$o);
 
		//file attributes
		$id = 		$file_data['id'];
		$open =		$file_data['open'];
 
		//build with relative indentation
		if($o['relative']){
			$clevel = $file_data['level'] - $o['relative'];
		}
		//absolute indentation
		else {
			$clevel = 	$file_data['level'];
		}
 
 
		//dont open the file
		if(!$open) {
			//only in case of directories and nons on
			return false;
		}
 
		//convert a:b -> home/data/pages/a/b
		$file    = wikiFN($id);
 
		//get data(in instructions format) from $file (dont use cache: false)
		$instr   = p_cached_instructions($file, true);
 
		//page was not empty
		if (!empty($instr)) {
			//fix relative links and lower headers of included pages
			unset($o['relative']);	//we dont need this option for conversion
			$instr   = $this->_convertInstructions($instr, $id, $renderer, $clevel,$o, $ok);
 
			//--------RENDER
			//renderer information(TOC build / Cache used)
			$info = array();
 
			//render the instructions to outpu data
			$content = p_render('xhtml', $instr, $info);
			//--------
 
			//Remove TOC`s, section edit buttons and tags
			$content = $this->_cleanXHTML($content);
		}
 
		// embed the included page
		$renderer->doc .= '<div class="include">';
			//add an anchor to find start of a inserted page
			$id = str_replace(":","_",$file_data['id']);	//not possible to use a:b:c for id
			$renderer->doc .= "<a name='$id' id='$id'>";
			$renderer->doc .= $content;
		$renderer->doc .= '</div>';
	}
 
 
 
	/*
	 * Corrects relative internal links and media and
	 * converts headers of included pages to subheaders of the current page
	 */
	function _convertInstructions($instr, $incl, &$renderer, $clevel,$o, &$ok){
		global $ID;
		global $conf;
 
		//clean input options
		$default = array(
					'start_ns'=>		"",
					'depth'=>			0,
					'header_count'=>	array(),
					'numbers'=>			false,
					'numbering_depth'=>	3,
					'split_too_long'=>	false,
					);
		$o = array_merge($default,$o);
 
		// check if included page is already in output namespace
		$iNS =	getNS($incl);
		$id =	getNS($ID);
 
		//the content belongs to the original page
		if ($id == $iNS)
			$convert = false;	//just leave as it is
		//the contetn was newly included, and is not original page content
		else {
			$convert = true;	//convert content
		}
 
		$n = count($instr);
		for ($i = 0; $i < $n; $i++){
			// convert it
			if ($convert) {
				//internal links(links inside this wiki) an relative links
				if((substr($instr[$i][0], 0, 8) == 'internal')){
					$this->_convert_link($renderer,$instr[$i],$iNS,$id,$o);
				}
				// set header level to current section level + header level
				elseif ($instr[$i][0] == 'header'){
					$this->_convert_header($renderer,$instr[$i],$clevel,$o);
				}
				// the same for sections
				elseif ($instr[$i][0] == 'section_open'){
					$level = $instr[$i][1][0]+$clevel;
					if ($level > 5)
						$level = 5;
					$instr[$i][1][0] = $level;
				}
			}//end of convert
 
			//--------SPLIT TOO LONG LINES ?
			if($o['split_too_long'] && $instr[$i][0]=='code') {
				$instr[$i][1][0] = wordwrap($instr[$i][1][0],70,"\n");
			}
			//--------
		}//end of for loop
 
		//if its the document start, cut off the first element(document information)
		if ($instr[0][0] == 'document_start')
			return array_slice($instr, 1, -1);
		else
			return $instr;
 
	}
 
	/*
	 * convert header of given instruction
	 */
	function _convert_header(&$renderer,&$instr,$clevel,$o) {
		global $conf;
 
		$level = $instr[1][1]+$clevel;
		//if a header level gets "lower" than 5
		if ($level > 5)
			$level = 5;
 
		$instr[1][1] = $level;
 
		//--------HEADER NUMBER
		if($o['numbers']) {
			//number infornt of header
			$number = "";
 
			//reset lower levels
			for($x=$level+1;$x<=5;$x++){
				$o['header_count'][$x]=0;
			}
 
			//raise this level by 1
			$o['header_count'][$level]++;
 
 
			//if the level is high (1=high)
			if($level <= $o['numbering_depth']) {
				//build the number
				for($x=1;$x<=$level;$x++) {
					$number .= $o['header_count'][$x] . ".";
				}
			}
 
			//save
			$instr[1][0] = $number . " " . $instr[1][0];
		}
		//--------
 
		// add TOC items
		if ($level >= $conf['toptoclevel'] && $level <= $conf['maxtoclevel']){
			$text = $instr[1][0];
			$header_id  = $renderer->_headerToLink($text, 'true');	//with unique id(true)
			//add item to TOC
			$renderer->toc[] = array(
				'hid'   => $header_id,
				'title' => $text,
				'type'  => 'ul',
				'level' => $level-$conf['toptoclevel']+1,
			);
 
		$ok = true;
		}
	}
 
	/*
	 * Convert link of given instruction
	 */
	function _convert_link(&$renderer,&$instr,$iNS,$id,$o) {
		// relative subnamespace
		if ($instr[1][0]{0} == '.'){
			//build complete address
			$instr[1][0] = $iNS.':'.substr($instr[1][0], 1);
		}
		// relative link
		elseif (strpos($instr[1][0],':') === false) {
			//build complete address
			$instr[1][0] = $iNS.':'.$instr[1][0];
		}
 
		//link to another page, but within our included namespace
		//if the link starts with our namespace
		if(strpos($instr[1][0],$o['start_ns']) === 0) {
			//if the link is inside the included depth of pages
			if($o['depth']===0 ||
				substr_count($o['start_ns'],":")+$o['depth'] >=
				substr_count($instr[1][0],":")) {
 
				//convert to internal link
				if(strpos($instr[1][0],"#") !== false) {
					//get last part a:b#c -> c
					$levels = split("#",$instr[1][0]);
 
					//$id we are in atm + # + id of anchor
					$instr[1][0] = $id . "#" . $levels[sizeof($levels)-1];
				}
				//a:b:c
				else {
					//$id we are in atm + # + full id of page to find(-> unique id)
					//not possible to use a:b:c for id --> : -> _
					$instr[1][0] = $id . "#" . str_replace(":","_",$instr[1][0]);
 
					//if page-link has no name
					if(empty($instr[1][1])) {
						//TODO: get the real name of the page :(
						$name = split("_",$instr[1][0]);
						//last part of id -> name
						$instr[1][1] = $name[sizeof($name)-1];
					}
				}
			}
		}
	}
 
 
  /**
   * Remove TOC, section edit buttons and tags
   */
  function _cleanXHTML($xhtml){
    $replace  = array(
      '!<div class="toc">.*?(</div>\n</div>)!s' => '', // remove TOCs
      '#<!-- SECTION \[(\d*-\d*)\] -->#e'       => '', // remove section edit buttons
      '!<div id="tags">.*?(</div>)!s'           => ''  // remove category tags
    );
    $xhtml  = preg_replace(array_keys($replace), array_values($replace), $xhtml);
    return $xhtml;
  }
}
 
 
/*
 * evaluate the files presented by search
 *
 * modify $data according to our wishes
 */
function composer_search_index(&$data,$base,$file,$type,$lvl,$opts){
	//include in result ?
	$return = true;
 
	//directories
	if($type == 'd'){
		//if max-level is reached, don't return them
		if ($opts['level'] == $lvl) $return=false;
 
		//add the directory name to list of blocked files
		//so that file with same name will not be included
		$data['blocked'][] = $file . ".txt";
 
		//if we don't want namespace-nodes in the resultset
		if ($opts['nons']) return $return;
	}
	//files that are no .txt
	elseif($type == 'f' && !preg_match('#\.txt$#',$file)){
		//don't add
		return false;
	}
 
	$id = pathID($file);
 
	//check hiddens
	if($type=='f' && isHiddenPage($id)){
		return false;
	}
 
	//check ACL (for namespaces too)
	if(auth_quickaclcheck($id) < AUTH_READ){
		//we are not allowed to read
		return false;
	}
	//check if this files was blocked by us
	if($type=='f' && in_array($file,$data['blocked'])){
		return false;
	}
 
	//add the start level(ns we build) to the current level
	$lvl += $data['start_lvl'];
 
	//if we don't want the namespace-nodes ?
	if ($opts['nons']) {
		//lower the level 1
		$lvl-=1;
	}
 
	//pack it up
	$data[]=array( 'id'    => $id,
		'type'  => $type,	//which type is it ?
		'level' => $lvl,	//show on which level ?
		'open'  => $return,	//open it ?
		);
 
	return $return;
}
 
?>
 

1.2 - new option “print” for print optimization(too long code lines will be split) - new option “numbers” and “numbers_depth” to build a numbering in front of the headings, “numbers” = for all , “numbers_depth” = whow deep ? (e.g. no numbers at level 5)

1.1 - links to pages inside the included tree are now included as direct links (#xxx)

1.0

Still need the real name of pages, if no name is supplied

 [[a:b:c]] 

normally DokuWiki would get the name, but it doesn't work because i converted it to #xxx

(see TODO: in code)

if you know an answer please message me :)

This plugin doesn't seem to support images. colons(:) are being replaced with underscores (_) so images are showing up as dead links. — Jonathan McBride 2007-10-31 16:24

Try changing line 361 from

if((substr($instr[$i][0], 0, 8) == 'internal')){

to

if((substr($instr[$i][0], 0, 12) == 'internallink')){

— Menno Dekker 2007-11-14 20:18

if you want to email me, try grosser.michael[Ä|t]gmÄil.com (watch out for the Ä`s)