Merge "Improve docs for Title::getInternalURL/getCanonicalURL"
[lhc/web/wiklou.git] / includes / export / BaseDump.php
1 <?php
2 /**
3 * Helper class for the --prefetch option of dumpTextPass.php
4 *
5 * Copyright © 2005 Brion Vibber <brion@pobox.com>
6 * https://www.mediawiki.org/
7 *
8 * This program is free software; you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation; either version 2 of the License, or
11 * (at your option) any later version.
12 *
13 * This program is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 * GNU General Public License for more details.
17 *
18 * You should have received a copy of the GNU General Public License along
19 * with this program; if not, write to the Free Software Foundation, Inc.,
20 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
21 * http://www.gnu.org/copyleft/gpl.html
22 *
23 * @file
24 * @ingroup Maintenance
25 */
26
27 /**
28 * Readahead helper for making large MediaWiki data dumps;
29 * reads in a previous XML dump to sequentially prefetch text
30 * records already normalized and decompressed.
31 *
32 * This can save load on the external database servers, hopefully.
33 *
34 * Assumes that dumps will be recorded in the canonical order:
35 * - ascending by page_id
36 * - ascending by rev_id within each page
37 * - text contents are immutable and should not change once
38 * recorded, so the previous dump is a reliable source
39 *
40 * @ingroup Maintenance
41 */
42 class BaseDump {
43 /** @var XMLReader */
44 protected $reader = null;
45 protected $atEnd = false;
46 protected $atPageEnd = false;
47 protected $lastPage = 0;
48 protected $lastRev = 0;
49 protected $infiles = null;
50
51 public function __construct( $infile ) {
52 $this->infiles = explode( ';', $infile );
53 $this->reader = new XMLReader();
54 $infile = array_shift( $this->infiles );
55 $this->reader->open( $infile, null, LIBXML_PARSEHUGE );
56 }
57
58 /**
59 * Attempts to fetch the text of a particular page revision
60 * from the dump stream. May return null if the page is
61 * unavailable.
62 *
63 * @param int $page ID number of page to read
64 * @param int $rev ID number of revision to read
65 * @return string|null
66 */
67 function prefetch( $page, $rev ) {
68 $page = intval( $page );
69 $rev = intval( $rev );
70 while ( $this->lastPage < $page && !$this->atEnd ) {
71 $this->debug( "BaseDump::prefetch at page $this->lastPage, looking for $page" );
72 $this->nextPage();
73 }
74 if ( $this->lastPage > $page || $this->atEnd ) {
75 $this->debug( "BaseDump::prefetch already past page $page "
76 . "looking for rev $rev [$this->lastPage, $this->lastRev]" );
77
78 return null;
79 }
80 while ( $this->lastRev < $rev && !$this->atEnd && !$this->atPageEnd ) {
81 $this->debug( "BaseDump::prefetch at page $this->lastPage, rev $this->lastRev, "
82 . "looking for $page, $rev" );
83 $this->nextRev();
84 }
85 if ( $this->lastRev == $rev && !$this->atEnd ) {
86 $this->debug( "BaseDump::prefetch hit on $page, $rev [$this->lastPage, $this->lastRev]" );
87
88 return $this->nextText();
89 } else {
90 $this->debug( "BaseDump::prefetch already past rev $rev on page $page "
91 . "[$this->lastPage, $this->lastRev]" );
92
93 return null;
94 }
95 }
96
97 function debug( $str ) {
98 wfDebug( $str . "\n" );
99 // global $dumper;
100 // $dumper->progress( $str );
101 }
102
103 /**
104 * @private
105 */
106 function nextPage() {
107 if ( $this->skipTo( 'page', 'mediawiki' ) ) {
108 if ( $this->skipTo( 'id' ) ) {
109 $this->lastPage = intval( $this->nodeContents() );
110 $this->lastRev = 0;
111 $this->atPageEnd = false;
112 }
113 } else {
114 $this->close();
115 if ( count( $this->infiles ) ) {
116 $infile = array_shift( $this->infiles );
117 $this->reader->open( $infile );
118 $this->atEnd = false;
119 }
120 }
121 }
122
123 /**
124 * @private
125 */
126 function nextRev() {
127 if ( $this->skipTo( 'revision' ) ) {
128 if ( $this->skipTo( 'id' ) ) {
129 $this->lastRev = intval( $this->nodeContents() );
130 }
131 } else {
132 $this->atPageEnd = true;
133 }
134 }
135
136 /**
137 * @private
138 * @return string
139 */
140 function nextText() {
141 $this->skipTo( 'text' );
142
143 return strval( $this->nodeContents() );
144 }
145
146 /**
147 * @private
148 * @param string $name
149 * @param string $parent
150 * @return bool|null
151 */
152 function skipTo( $name, $parent = 'page' ) {
153 if ( $this->atEnd ) {
154 return false;
155 }
156 while ( $this->reader->read() ) {
157 if ( $this->reader->nodeType == XMLReader::ELEMENT
158 && $this->reader->name == $name
159 ) {
160 return true;
161 }
162 if ( $this->reader->nodeType == XMLReader::END_ELEMENT
163 && $this->reader->name == $parent
164 ) {
165 $this->debug( "BaseDump::skipTo found </$parent> searching for <$name>" );
166
167 return false;
168 }
169 }
170
171 return $this->close();
172 }
173
174 /**
175 * Shouldn't something like this be built-in to XMLReader?
176 * Fetches text contents of the current element, assuming
177 * no sub-elements or such scary things.
178 *
179 * @return string
180 * @private
181 */
182 function nodeContents() {
183 if ( $this->atEnd ) {
184 return null;
185 }
186 if ( $this->reader->isEmptyElement ) {
187 return "";
188 }
189 $buffer = "";
190 while ( $this->reader->read() ) {
191 switch ( $this->reader->nodeType ) {
192 case XMLReader::TEXT:
193 // case XMLReader::WHITESPACE:
194 case XMLReader::SIGNIFICANT_WHITESPACE:
195 $buffer .= $this->reader->value;
196 break;
197 case XMLReader::END_ELEMENT:
198 return $buffer;
199 }
200 }
201
202 return $this->close();
203 }
204
205 /**
206 * @private
207 * @return null
208 */
209 function close() {
210 $this->reader->close();
211 $this->atEnd = true;
212
213 return null;
214 }
215 }