xref: /plugin/sphinxsearch-was/functions.php (revision 33:f707cbb17007) 
1<?php
2/*
3 * To change this template, choose Tools | Templates
4 * and open the template in the editor.
5 */
6
7function formatXml($data)
8{
9    $xmlFormat = '
10<sphinx:document id="{id}">
11<title><![CDATA[[{title}]]></title>
12<body><![CDATA[[{body}]]></body>
13<categories><![CDATA[[{categories}]]></categories>
14<level>{level}</level>
15<modified><![CDATA[[{modified}]]></modified>
16</sphinx:document>
17
18';
19
20    return str_replace( array('{id}', '{title}', '{body}', '{categories}', '{level}', '{modified}'),
21                        array($data['id'], html_strip($data['title']), html_strip($data['body']), html_strip($data['categories']),
22                             $data['level'], $data['modified']),
23                $xmlFormat
24            );
25}
26
27function getDocumentsByHeadings($id, $metadata)
28{
29    if (empty($metadata) || empty($metadata['description']['tableofcontents'])) return false;
30
31    $sections = array();
32    $level = 1;
33    $previouse_title = '';
34    foreach($metadata['description']['tableofcontents'] as $row){
35        $sections[$row['hid']] = array(
36                                    'section' => getSectionByTitleLevel($id, $row['title']),
37                                    'level' => $row['level'],
38                                    'title' => $row['title']
39                                    );
40        if ($row['level'] > $level){
41            $sections[$row['hid']]['title_text'] = $previouse_title . " &raquo; ".$row['title'];
42        } else {
43            $sections[$row['hid']]['title_text'] = $row['title'];
44            $previouse_title = $row['title'];
45        }
46    }
47    return $sections;
48}
49
50function getSectionByTitleLevel($id, $header, $extended=false)
51{
52    $headerReg = preg_quote($header, '/');
53    $doc = io_readFile(wikiFN($id));
54    $regex = "(={1,6})\s*({$headerReg})\s*(={1,6})";
55    $section = '';
56    if (preg_match("/$regex/i",$doc,$matches)) {
57        $startHeader = $matches[0];
58        $startHeaderPos = strpos($doc, $startHeader) + strlen($startHeader);
59        $endDoc = substr($doc, $startHeaderPos);
60
61        $regex = '(={1,6})(.*?)(={1,6})';
62        if (preg_match("/$regex/i",$endDoc,$matches)) {
63            $endHeader = $matches[0];
64            $endHeaderPos = strpos($doc, $endHeader);
65        } else {
66            $endHeaderPos = 0;
67        }
68        if ($endHeaderPos){
69            $section = substr($doc, $startHeaderPos, $endHeaderPos - $startHeaderPos);
70        } else {
71            $section = substr($doc, $startHeaderPos);
72        }
73    }
74    $section = trim($section);
75    //trying to get next section content if body for first section is empty
76    //working only for extended mode
77    if ($extended && empty($section)){
78        $startHeaderPos = $endHeaderPos + strlen($endHeader);
79        $endDoc = substr($endDoc, $startHeaderPos);
80        $regex = '(={1,6})(.*?)(={1,6})';
81        if (preg_match("/$regex/i",$endDoc,$matches)) {
82            $endHeader = $matches[0];
83            $endHeaderPos = strpos($doc, $endHeader);
84        } else {
85            $endHeaderPos = 0;
86        }
87        if ($endHeaderPos){
88            $section = substr($doc, $startHeaderPos, $endHeaderPos - $startHeaderPos);
89        } else {
90            $section = substr($doc, $startHeaderPos);
91        }
92    }
93    $section = trim($section);
94    return $section;
95}
96
97function getSection($id, $header)
98{
99    static $cacheInstructions = null;
100    static $cacheDoc = null;
101
102    if (empty($cacheDoc[$id])){
103        // Create the parser
104        $Parser = & new Doku_Parser();
105
106        // Add the Handler
107        $Parser->Handler = & new Doku_Handler();
108
109        // Load the header mode to find headers
110        $Parser->addMode('header',new Doku_Parser_Mode_Header());
111        $Parser->addMode('listblock',new Doku_Parser_Mode_ListBlock());
112
113        // Loads the raw wiki document
114        $doc = io_readFile(wikiFN($id));
115
116        // Get a list of instructions
117        $instructions = $Parser->parse($doc);
118
119        unset($Parser->Handler);
120        unset($Parser);
121
122        //free old cache
123        $cacheInstructions = null;
124        $cacheDoc = null;
125
126        //initialize new cache
127        $cacheInstructions[$id] = $instructions;
128        $cacheDoc[$id] = $doc;
129    } else {
130        $instructions = $cacheInstructions[$id];
131        $doc = $cacheDoc[$id];
132    }
133
134
135
136    // Use this to watch when we're inside the section we want
137    $inSection = FALSE;
138    $startPos = 0;
139    $endPos = 0;
140
141    // Loop through the instructions
142    foreach ( $instructions as $instruction ) {
143
144        if ( !$inSection ) {
145
146            // Look for the header for the "Lists" heading
147            if ( $instruction[0] == 'header' &&
148                    trim($instruction[1][0]) == $header ) {
149
150                $startPos = $instruction[2];
151                $inSection = TRUE;
152            }
153        } else {
154
155            // Look for the end of the section
156            if ( $instruction[0] == 'section_close' ) {
157                $endPos = $instruction[2];
158                break;
159            }
160        }
161    }
162
163    // Normalize and pad the document in the same way the parse does
164    // so that byte indexes with match
165    $doc = "\n".str_replace("\r\n","\n",$doc)."\n";
166    $section = substr($doc, $startPos, ($endPos-$startPos));
167
168    return $section;
169}
170
171function getCategories($id)
172{
173    if (empty($id)) return '';
174
175    if (false === strpos($id, ":")){
176        return $id;
177    }
178
179    $ns = explode(":", $id);
180    $nsCount = count($ns);
181
182    $result = '';
183    do{
184        for($i = 0; $i < $nsCount; $i++){
185            $name = $ns[$i];
186            $result .= $name;
187            if ($i < $nsCount - 1){
188                 $result .= ':';
189            }
190        }
191        $result .= ' ';
192    }while($nsCount--);
193    return $result;
194}
195
196
197 /**
198  * Method return all wiki page names
199  * @global array $conf
200  * @return array
201  */
202 function getPagesList()
203 {
204    global $conf;
205
206    $data = array();
207    sort($data);
208    search($data,$conf['datadir'],'search_allpages','','');
209
210    return $data;
211}
212
213function getNsLinks($id, $keywords, $search)
214{
215    global $conf;
216    $parts = explode(':', $id);
217    $count = count($parts);
218
219    // print intermediate namespace links
220    $part = '';
221    $data = array();
222    $titles = array();
223    for($i=0; $i<$count; $i++){
224        $part .= $parts[$i].':';
225        $page = $part;
226        resolve_pageid('',$page,$exists);
227
228        if (preg_match("#:start$#", $page) && !preg_match("#:start:$#", $part)) {
229            $page = substr($page, 0, strpos($page, ":start"));
230        };
231
232        // output
233        if ($exists){
234            $titles[wl($page)] = $parts[$i];
235        } else {
236            $titles[wl($page)] = $parts[$i];
237        }
238        $data[] = array('link' => "?do=search&id={$keywords}".urlencode(" @categories $page"));
239    }
240    $titleExcerpt = $search->getExcerpt($titles, $search->starQuery($keywords));
241    $i = 0;
242    foreach ($data as $key => $notused){
243        $data[$key]['title'] = $titleExcerpt[$i++];
244    }
245    return $data;
246}