Blikiの入力画面には特に何も記載されていないので、Templateが扱えるかどうかを確認してみた。
結論から言えば、このコードはそこまで考慮して書かれていないので、Templateを導入することは出来なかった。
そこまでの対応をするくらいなら１から書いた方が早いかもしれない。そんな時間はないので、今回はここで諦めよう。
しかし、最初からHeaderと末尾にTag位は入れたいなぁ。
そうそう、コードを読んでたらあまりにも整形されていなかったので、手で整形し直した。ついでに、無駄な判断分があったので、一部修正。
まぁ、あんまりにも小さな修正なので、影響は皆無だろう。ま、いいか。取りあえず貼り付けてみる。
$ cat ./syntax.php
<?php
/**
* bliki Plugin: Adds a simple blogging engine to your wiki
*
* @license GPL 2 (http://www.gnu.org/licenses/gpl.html)
* @author Beau Lebens <beau@dentedreality.com.au>
* @author Anthony Caetano <Anthony.Caetano@Sanlam.co.za>
* @author HEO SeonMeyong <seirios.seirios.org>
*
*/
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');
/**
* All DokuWiki plugins to extend the parser/rendering mechanism
* need to inherit from this class
*/
class syntax_plugin_bliki extends DokuWiki_Syntax_Plugin {
/**
* return some info
*/
function getInfo(){
return array(
'author' => 'Beau Lebens',
'email' => 'beau@dentedreality.com.au',
'date' => '2005-09-06',
'name' => 'Bliki: The Wiki Blog',
'desc' => 'Adds basic blogging functionality to any page of your wiki.',
'url' => 'http://www.dokuwiki.org/plugin:bliki',
);
}
/**
* What kind of syntax are we?
*/
function getType(){
return 'substition';
}
/**
* What kind of syntax do we allow (optional)
*/
function getAllowedTypes() {
return array('container', 'formatting', 'substition', 'protected', 'disabled', 'paragraphs');
}
/**
* What about paragraphs? (optional)
*/
function getPType(){
return 'block';
}
/**
* Where to sort in?
*/
function getSort(){
return 400;
}
/**
* Connect pattern to lexer
*/
function connectTo($mode) {
$this->Lexer->addSpecialPattern('~~BLIKI~~', $mode, 'plugin_bliki');
}
/**
* Handle the match
*/
function handle($match, $state, $pos, &$handler){
return array();
}
/**
* @return Array
* @param String $ID
* @param Int $num
* @param Int $offset
* @desc Finds the full pathnames of the most recent $num posts, starting at the optional within the $ID blog/namespace.
*/
function getPosts($ID, $num, $offset = 0) {
global $conf;
$recents = array();
$counter = 0;
// fully-qaulify the ID that we're working with (to dig into the namespace)
$fp = wikiFN($ID);
$ID = substr($fp, 0, strrpos($fp, '.'));
// Only do it if the namespace exists
if (is_dir($ID . '/')) {
if ($conf['bliki']['structure'] == 'flat') {
$posts = $this->read_dir_to_array($ID . '/', 'file', '/^[0-9]{4}-[0-9]{2}-[0-9]{2}-[0-9]{6}\.txt$/');
sort($posts);
while (sizeof($recents) < $num && sizeof($posts)) {
$post = array_pop($posts);
$counter++;
if ($counter > $offset) {
$recents[] = $ID . '/' . $post;
}
}
return $recents;
} else { // $conf['bliki']['structure'] == 'deep'
$years = $this->read_dir_to_array($ID . '/', 'dir');
sort($years);
// Now start working backwards through it all and get the most recent posts
while (sizeof($recents) < $num && sizeof($years)) {
$year = array_pop($years);
$months = $this->read_dir_to_array($ID . '/' . $year . '/', 'dir');
sort($months);
while (sizeof($recents) < $num && sizeof($months)) {
$month = array_pop($months);
$days = $this->read_dir_to_array($ID . '/' . $year . '/' . $month . '/', 'dir');
sort($days);
while (sizeof($recents) < $num && sizeof($days)) {
$day = array_pop($days);
$posts = $this->read_dir_to_array($ID . '/' . $year . '/' . $month . '/' . $day . '/',
'file',
'/^[0-9]{6}\.txt$/');
sort($posts);
while (sizeof($recents) < $num && sizeof($posts)) {
$post = array_pop($posts);
$counter++;
if ($counter > $offset) {
$recents[] = $ID . '/' . $year . '/' . $month . '/' . $day . '/' . $post;
}
}
}
}
}
return $recents;
}
}
}
/**
* @return String
* @param Array $list
* @desc Compiles the contents of all the files listed (as fully-qualified paths)
* in $list into a single string. Compiles in the order listed. Adds date headers
* where required and a footer after each post.
*/
function compilePosts($list) {
global $ID, $conf;
if (sizeof($list)) {
$last_date = false;
$str = '';
foreach ($list as $file) {
// Decide if we need to add a date divider
$file = str_replace('\\', '/', $file);
$ts = $this->getTimestampFromFile($file);
$date = date('Y-m-d', $ts);
if ($date != $last_date) {
$str .= $this->getDateHeader($ts);
$last_date = $date;
}
// Add this file's contents to the output
$str .= file_get_contents($file);
// And add a wiki-formatted footer of meta data as well, accounting for rewrites
$post_url = $this->getUrlPartFromTimestamp($ID, $ts);
$edit_url = $this->getRewriteUrl($post_url, 'do=edit', false);
$timestamp = date($conf['bliki']['datefooter'], $ts);
$str .= str_replace(array('{timestamp}', '{permalink}', '{edit}'), array($timestamp, $post_url, "this>$edit_url"), $conf['bliki']['footer']);
}
return $str;
} else {
return '';
}
}
/**
* @return String
* @param timestamp $ts
* @desc Returns a wiki-formatted date header for between posts.
*/
function getDateHeader($ts) {
global $conf;
$date = date($conf['bliki']['dateheader'], $ts);
return $date . "\n";
}
/**
* @return timestamp
* @param String $filename
* @desc Returns a timestamp based on the filename/namespace structure
*/
function getTimestampFromFile($file) {
global $conf;
if ($conf['bliki']['structure'] == 'flat') {
$parts = explode('-', basename($file));
$ts = mktime(substr($parts[3], 0, 2), substr($parts[3], 2, 2), substr($parts[3], 4, 2), $parts[1], $parts[2], $parts[0]);
} else { // $conf['bliki']['structure'] == 'deep'
$parts = explode('/', dirname($file));
$s = sizeof($parts);
$date = $parts[$s-3] . '-' . $parts[$s-2] . '-' . $parts[$s-1];
$filename = basename($file);
$ts = mktime(substr($filename, 0, 2), substr($filename, 2, 2), substr($filename, 4, 2), $parts[$s-2], $parts[$s-1], $parts[$s-3]);
}
return $ts;
}
/**
* @return String
* @param String $ID
* @param timestamp $ts
* @desc Returns a post url for a post based on the post's timestamp and the base ID
*/
function getUrlPartFromTimestamp($ID, $ts=0) {
global $conf;
if ($ts == 0) {
$ts = time();
}
if ($conf['userewrite'] > 0) {
$sep = ($conf['useslash'] == true ? '/' : ':');
} else {
$sep = ':';
}
if ($conf['bliki']['structure'] == 'flat') {
return $ID . $sep . date('Y-m-d-His', $ts);
} else { // $conf['bliki']['structure'] == 'deep'
return $ID . $sep . date('Y' . $sep . 'm' . $sep . 'd' . $sep . 'His', $ts);
}
}
/**
* @return String
* @param String $url
* @param String $query
* @desc Returns a url properly prefixed according to the $conf['rewrite'] option
* A $page is an appropriately constructed (namespace inclusive and considering $conf['useslash']) page reference
* A $query contains a query string parameters to append
*/
function getRewriteUrl($page, $query, $base = true) {
global $conf;
if ($base) {
$str = DOKU_BASE;
}
if ($conf['userewrite'] == 0) {
$str .= DOKU_SCRIPT . '?id=' . $page;
if ($query != '') {
$str .= '&' . $query;
}
} else if ($conf['userewrite'] == 1) {
$str .= idfilter($page, false);
if ($query != '') {
$str .= '?' . $query;
}
} else {
$str .= DOKU_SCRIPT . '/' . idfilter($page, false);
if ($query != '') {
$str .= '?' . $query;
}
}
return $str;
}
/**
* @return String
* @param String $label
* @desc Creates the HTML required for the "New Post" link, using the lable provided.
*/
function newPostLink($label) {
global $conf, $ID;
$sep = ($conf['useslash'] == true ? '/' : ':');
$html = '<div id="blognew"><a href="';
$html .= $this->getRewriteUrl($this->getUrlPartFromTimestamp($ID, time() + (isset($conf['bliki']['offset']) ? ($conf['bliki']['offset'] * 3600) : 0)), 'do=edit');
$html .= '">' . $label . '</a></div>';
return $html;
}
/**
* @return String
* @param Int $page
* @param String $label
* @desc Creates the HTML required for a link to an older/newer page of posts.
*/
function pagingLink($page, $label) {
global $conf, $ID;
$html = '<a href="';
$html .= $this->getRewriteUrl($ID, "page=$page");
$html .= '">' . $label . '</a>';
return $html;
}
/**
* @return Array
* @param String $dir
* @param String $select
* @param String $match
* @desc Reads all entries in a directory into an array, sorted alpha, dirs then files.
* $select is used to selects either only dir(ectories), file(s) or both
* If $match is supplied, it should be a / delimited regex to match the filename against
*/
function read_dir_to_array($dir, $select = 'both', $match = false) {
$files = array();
$dirs = array();
// Read all the entries in the directory specified
$handle = @opendir($dir);
if (!$handle) {
return false;
}
while ($file = @readdir($handle)) {
// Ignore self and parent references
if ($file != '.' && $file != '..') {
if (($select == 'both' || $select == 'dir') && is_dir($dir . $file)) {
$dirs[] = $file;
} else if (($select == 'both' || $select == 'file') && !is_dir($dir . $file)) {
if (is_string($match)) {
if (!preg_match($match, $file)) {
continue;
}
}
$files[] = $file;
}
}
}
@closedir($handle);
// Sort anything found alphabetically and combine the results (dirs->files)
if (sizeof($dirs) > 0) {
sort($dirs, SORT_STRING);
}
if (sizeof($files) > 0) {
sort($files, SORT_STRING);
}
// Put the directories and files back together and return them
return array_merge($dirs, $files);
}
/**
* Create output
*/
function render($mode, &$renderer, $data) {
global $ID, $conf;
// Set the page number (determines which posts we display)
if (isset($_REQUEST['page'])) {
$page = $_REQUEST['page'];
} else {
$page = 0;
}
if ($mode == 'xhtml') {
// Addlink for creating a new post
$renderer->doc .= $this->newPostLink($conf['bliki']['newlabel']);
// Go and get the required blog posts and compile them into one wikitext string
// FIXME $config var for how many? or inline directive?
$recents = $this->getPosts($ID, $conf['bliki']['numposts'], ($page * $conf['bliki']['numposts']));
$compiled = $this->compilePosts($recents);
// Disable section editing to avoid weird links
$conf['maxseclevel'] = 0;
// Disbale caching because we need to get new files always
$renderer->info['cache'] = false;
// Add the compiled blog posts after rendering them.
$renderer->doc .= p_render('xhtml', p_get_instructions($compiled), $info);
// Add a link to older entries if we filled the number per page (assuming there's more)
if (sizeof($recents) == $conf['bliki']['numposts']) {
$renderer->doc .= '<div id="blogolder">' . $this->pagingLink($page+1, $conf['bliki']['olderlabel']) . '</div>';
}
// And also a link to newer posts if we're not on page 0
if ($page != 0) {
$renderer->doc .= '<div id="blognewer">' . $this->pagingLink($page-1, $conf['bliki']['newerlabel']) . '</div>';
}
return true;
}
return false;
}
}
?>
$
ああ、やっぱりTagをうち忘れた。後で困るって。