| Index: trunk/phase3/maintenance/importUseModWiki.php |
| — | — | @@ -40,40 +40,49 @@ |
| 41 | 41 | * @ingroup Maintenance |
| 42 | 42 | */ |
| 43 | 43 | |
| 44 | | -if ( php_sapi_name() != 'cli' ) { |
| 45 | | - echo "Please customize the settings and run me from the command line."; |
| 46 | | - die( -1 ); |
| 47 | | -} |
| | 44 | +require_once( "Maintenance.php" ); |
| 48 | 45 | |
| 49 | | -/** Set these correctly! */ |
| 50 | | -$wgImportEncoding = "CP1252"; /* We convert all to UTF-8 */ |
| 51 | | -$wgRootDirectory = "/kalman/Projects/wiki2002/wiki/lib-http/db/wiki"; |
| | 46 | +class ImportUseModWiki extends Maintenance { |
| 52 | 47 | |
| 53 | | -/* On a large wiki, you might run out of memory */ |
| 54 | | -@ini_set( 'memory_limit', '40M' ); |
| | 48 | + private $encoding, $rootDirectory = ''; |
| 55 | 49 | |
| 56 | | -/* globals */ |
| 57 | | -$wgFieldSeparator = "\xb3"; # Some wikis may use different char |
| 58 | | - $FS = $wgFieldSeparator ; |
| 59 | | - $FS1 = $FS . "1" ; |
| 60 | | - $FS2 = $FS . "2" ; |
| 61 | | - $FS3 = $FS . "3" ; |
| | 50 | + /** |
| | 51 | + * Field separators |
| | 52 | + * @var String |
| | 53 | + */ |
| | 54 | + private $FS1, $FS2, $FS3 = ''; |
| 62 | 55 | |
| 63 | | -# Unicode sanitization tools |
| 64 | | -require_once( dirname( dirname( __FILE__ ) ) . '/includes/normal/UtfNormal.php' ); |
| | 56 | + /** |
| | 57 | + * @var Array |
| | 58 | + */ |
| | 59 | + private $usercache, $nowiki = array(); |
| 65 | 60 | |
| 66 | | -$usercache = array(); |
| | 61 | + public function __construct() { |
| | 62 | + parent::__construct(); |
| | 63 | + $this->mDescription = "Import pages from UseMod wikis"; |
| | 64 | + $this->addOption( 'encoding', 'Encoding of the imported text, default CP1252', false, true ); |
| | 65 | + /** |
| | 66 | + * If UseModWiki's New File System is used: |
| | 67 | + * $NewFS = 1; # 1 = new multibyte $FS, 0 = old $FS |
| | 68 | + * Use "\xb3"; for the Old File System |
| | 69 | + * Changed with UTF-8 UseModWiki |
| | 70 | + * http://www.usemod.com/cgi-bin/wiki.pl?SupportForUtf8 |
| | 71 | + * http://www.usemod.com/cgi-bin/wiki.pl?WikiBugs/NewFieldSeparatorWronglyTreated |
| | 72 | + * http://www.meatballwiki.org/wiki/WikiEngine#Q_amp_A |
| | 73 | + */ |
| | 74 | + $this->addOption( 'separator', 'Field separator to use, default \x1E\xFF\xFE\x1E', false, true ); |
| | 75 | + $this->addArg( 'path', 'Path to your UseMod wiki' ); |
| | 76 | + } |
| 67 | 77 | |
| 68 | | -importPages(); |
| | 78 | + public function execute() { |
| | 79 | + $this->rootDirectory = $this->getArg(); |
| | 80 | + $this->encoding = $this->getOption( 'encoding', 'CP1252' ); |
| | 81 | + $sep = $this->getOption( 'separator', "\x1E\xFF\xFE\x1E" ); |
| | 82 | + $this->FS1 = "{$sep}1"; |
| | 83 | + $this->FS2 = "{$sep}2"; |
| | 84 | + $this->FS3 = "{$sep}3"; |
| 69 | 85 | |
| 70 | | -# ------------------------------------------------------------------------------ |
| 71 | | - |
| 72 | | -function importPages() |
| 73 | | -{ |
| 74 | | - global $wgRootDirectory; |
| 75 | | - |
| 76 | | - $gt = '>'; |
| 77 | | - echo <<<XML |
| | 86 | + echo <<<XML |
| 78 | 87 | <?xml version="1.0" encoding="UTF-8" ?> |
| 79 | 88 | <mediawiki xmlns="http://www.mediawiki.org/xml/export-0.1/" |
| 80 | 89 | xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" |
| — | — | @@ -89,290 +98,278 @@ |
| 90 | 99 | 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', |
| 91 | 100 | 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z', 'other' ); |
| 92 | 101 | foreach ( $letters as $letter ) { |
| 93 | | - $dir = "$wgRootDirectory/page/$letter"; |
| | 102 | + $dir = "{$this->rootDirectory}/page/$letter"; |
| 94 | 103 | if ( is_dir( $dir ) ) |
| 95 | | - importPageDirectory( $dir ); |
| | 104 | + $this->importPageDirectory( $dir ); |
| 96 | 105 | } |
| 97 | 106 | echo <<<XML |
| 98 | 107 | </mediawiki> |
| 99 | 108 | |
| 100 | 109 | XML; |
| 101 | | -} |
| | 110 | + } |
| 102 | 111 | |
| 103 | | -function importPageDirectory( $dir, $prefix = "" ) |
| 104 | | -{ |
| 105 | | - echo "\n<!-- Checking page directory " . xmlCommentSafe( $dir ) . " -->\n"; |
| 106 | | - $mydir = opendir( $dir ); |
| 107 | | - while ( $entry = readdir( $mydir ) ) { |
| 108 | | - $m = array(); |
| 109 | | - if ( preg_match( '/^(.+)\.db$/', $entry, $m ) ) { |
| 110 | | - echo importPage( $prefix . $m[1] ); |
| 111 | | - } else { |
| 112 | | - if ( is_dir( "$dir/$entry" ) ) { |
| 113 | | - if ( $entry != '.' && $entry != '..' ) { |
| 114 | | - importPageDirectory( "$dir/$entry", "$entry/" ); |
| 115 | | - } |
| | 112 | + private function importPageDirectory( $dir, $prefix = "" ) { |
| | 113 | + echo "\n<!-- Checking page directory " . $this->xmlCommentSafe( $dir ) . " -->\n"; |
| | 114 | + $mydir = opendir( $dir ); |
| | 115 | + while ( $entry = readdir( $mydir ) ) { |
| | 116 | + $m = array(); |
| | 117 | + if ( preg_match( '/^(.+)\.db$/', $entry, $m ) ) { |
| | 118 | + echo $this->importPage( $prefix . $m[1] ); |
| 116 | 119 | } else { |
| 117 | | - echo "<!-- File '" . xmlCommentSafe( $entry ) . "' doesn't seem to contain an article. Skipping. -->\n"; |
| | 120 | + if ( is_dir( "$dir/$entry" ) ) { |
| | 121 | + if ( $entry != '.' && $entry != '..' ) { |
| | 122 | + $this->importPageDirectory( "$dir/$entry", "$entry/" ); |
| | 123 | + } |
| | 124 | + } else { |
| | 125 | + echo "<!-- File '" . $this->xmlCommentSafe( $entry ) . "' doesn't seem to contain an article. Skipping. -->\n"; |
| | 126 | + } |
| 118 | 127 | } |
| 119 | 128 | } |
| 120 | 129 | } |
| 121 | | -} |
| 122 | 130 | |
| | 131 | + private function useModFilename( $title ) { |
| | 132 | + $c = substr( $title, 0, 1 ); |
| | 133 | + if ( preg_match( '/[A-Z]/i', $c ) ) { |
| | 134 | + return strtoupper( $c ) . "/$title"; |
| | 135 | + } |
| | 136 | + return "other/$title"; |
| | 137 | + } |
| 123 | 138 | |
| 124 | | -# ------------------------------------------------------------------------------ |
| | 139 | + private function fetchPage( $title ) { |
| | 140 | + $fname = $this->rootDirectory . "/page/" . $this->useModFilename( $title ) . ".db"; |
| | 141 | + if ( !file_exists( $fname ) ) { |
| | 142 | + echo "Couldn't open file '$fname' for page '$title'.\n"; |
| | 143 | + die( -1 ); |
| | 144 | + } |
| 125 | 145 | |
| 126 | | -/* fetch_ functions |
| 127 | | - Grab a given item from the database |
| 128 | | - */ |
| | 146 | + $page = $this->splitHash( $this->FS1, file_get_contents( $fname ) ); |
| | 147 | + $section = $this->splitHash( $this->FS2, $page["text_default"] ); |
| | 148 | + $text = $this->splitHash( $this->FS3, $section["data"] ); |
| 129 | 149 | |
| 130 | | -function useModFilename( $title ) { |
| 131 | | - $c = substr( $title, 0, 1 ); |
| 132 | | - if ( preg_match( '/[A-Z]/i', $c ) ) { |
| 133 | | - return strtoupper( $c ) . "/$title"; |
| | 150 | + return $this->array2object( array( "text" => $text["text"] , "summary" => $text["summary"] , |
| | 151 | + "minor" => $text["minor"] , "ts" => $section["ts"] , |
| | 152 | + "username" => $section["username"] , "host" => $section["host"] ) ); |
| 134 | 153 | } |
| 135 | | - return "other/$title"; |
| 136 | | -} |
| 137 | 154 | |
| 138 | | -function fetchPage( $title ) |
| 139 | | -{ |
| 140 | | - global $FS1, $FS2, $FS3, $wgRootDirectory; |
| | 155 | + private function fetchKeptPages( $title ) { |
| | 156 | + $fname = $this->rootDirectory . "/keep/" . $this->useModFilename( $title ) . ".kp"; |
| | 157 | + if ( !file_exists( $fname ) ) return array(); |
| 141 | 158 | |
| 142 | | - $fname = $wgRootDirectory . "/page/" . useModFilename( $title ) . ".db"; |
| 143 | | - if ( !file_exists( $fname ) ) { |
| 144 | | - echo "Couldn't open file '$fname' for page '$title'.\n"; |
| 145 | | - die( -1 ); |
| 146 | | - } |
| | 159 | + $keptlist = explode( $this->FS1, file_get_contents( $fname ) ); |
| | 160 | + array_shift( $keptlist ); # Drop the junk at beginning of file |
| 147 | 161 | |
| 148 | | - $page = splitHash( $FS1, file_get_contents( $fname ) ); |
| 149 | | - $section = splitHash( $FS2, $page["text_default"] ); |
| 150 | | - $text = splitHash( $FS3, $section["data"] ); |
| 151 | | - |
| 152 | | - return array2object( array( "text" => $text["text"] , "summary" => $text["summary"] , |
| 153 | | - "minor" => $text["minor"] , "ts" => $section["ts"] , |
| 154 | | - "username" => $section["username"] , "host" => $section["host"] ) ); |
| 155 | | -} |
| 156 | | - |
| 157 | | -function fetchKeptPages( $title ) |
| 158 | | -{ |
| 159 | | - global $FS1, $FS2, $FS3, $wgRootDirectory; |
| 160 | | - |
| 161 | | - $fname = $wgRootDirectory . "/keep/" . useModFilename( $title ) . ".kp"; |
| 162 | | - if ( !file_exists( $fname ) ) return array(); |
| 163 | | - |
| 164 | | - $keptlist = explode( $FS1, file_get_contents( $fname ) ); |
| 165 | | - array_shift( $keptlist ); # Drop the junk at beginning of file |
| 166 | | - |
| 167 | | - $revisions = array(); |
| 168 | | - foreach ( $keptlist as $rev ) { |
| 169 | | - $section = splitHash( $FS2, $rev ); |
| 170 | | - $text = splitHash( $FS3, $section["data"] ); |
| 171 | | - if ( $text["text"] && $text["minor"] != "" && ( $section["ts"] * 1 > 0 ) ) { |
| 172 | | - array_push( $revisions, array2object( array ( "text" => $text["text"] , "summary" => $text["summary"] , |
| 173 | | - "minor" => $text["minor"] , "ts" => $section["ts"] , |
| 174 | | - "username" => $section["username"] , "host" => $section["host"] ) ) ); |
| 175 | | - } else { |
| 176 | | - echo "<!-- skipped a bad old revision -->\n"; |
| | 162 | + $revisions = array(); |
| | 163 | + foreach ( $keptlist as $rev ) { |
| | 164 | + $section = $this->splitHash( $this->FS2, $rev ); |
| | 165 | + $text = $this->splitHash( $this->FS3, $section["data"] ); |
| | 166 | + if ( $text["text"] && $text["minor"] != "" && ( $section["ts"] * 1 > 0 ) ) { |
| | 167 | + array_push( $revisions, $this->array2object( array ( "text" => $text["text"] , "summary" => $text["summary"] , |
| | 168 | + "minor" => $text["minor"] , "ts" => $section["ts"] , |
| | 169 | + "username" => $section["username"] , "host" => $section["host"] ) ) ); |
| | 170 | + } else { |
| | 171 | + echo "<!-- skipped a bad old revision -->\n"; |
| | 172 | + } |
| 177 | 173 | } |
| | 174 | + return $revisions; |
| 178 | 175 | } |
| 179 | | - return $revisions; |
| 180 | | -} |
| 181 | 176 | |
| 182 | | -function splitHash ( $sep , $str ) { |
| 183 | | - $temp = explode ( $sep , $str ) ; |
| 184 | | - $ret = array () ; |
| 185 | | - for ( $i = 0; $i + 1 < count ( $temp ) ; $i++ ) { |
| 186 | | - $ret[$temp[$i]] = $temp[++$i] ; |
| 187 | | - } |
| 188 | | - return $ret ; |
| | 177 | + private function splitHash( $sep , $str ) { |
| | 178 | + $temp = explode ( $sep , $str ) ; |
| | 179 | + $ret = array () ; |
| | 180 | + for ( $i = 0; $i + 1 < count ( $temp ) ; $i++ ) { |
| | 181 | + $ret[$temp[$i]] = $temp[++$i] ; |
| | 182 | + } |
| | 183 | + return $ret ; |
| 189 | 184 | } |
| 190 | 185 | |
| 191 | | - |
| 192 | | -/* import_ functions |
| 193 | | - Take a fetched item and produce SQL |
| 194 | | - */ |
| 195 | | - |
| 196 | | -function checkUserCache( $name, $host ) |
| 197 | | -{ |
| 198 | | - global $usercache; |
| 199 | | - |
| 200 | | - if ( $name ) { |
| 201 | | - if ( in_array( $name, $usercache ) ) { |
| 202 | | - $userid = $usercache[$name]; |
| | 186 | + private function checkUserCache( $name, $host ) { |
| | 187 | + if ( $name ) { |
| | 188 | + if ( in_array( $name, $this->usercache ) ) { |
| | 189 | + $userid = $this->usercache[$name]; |
| | 190 | + } else { |
| | 191 | + # If we haven't imported user accounts |
| | 192 | + $userid = 0; |
| | 193 | + } |
| | 194 | + $username = str_replace( '_', ' ', $name ); |
| 203 | 195 | } else { |
| 204 | | - # If we haven't imported user accounts |
| 205 | 196 | $userid = 0; |
| | 197 | + $username = $host; |
| 206 | 198 | } |
| 207 | | - $username = str_replace( '_', ' ', $name ); |
| 208 | | - } else { |
| 209 | | - $userid = 0; |
| 210 | | - $username = $host; |
| | 199 | + return array( $userid, $username ); |
| 211 | 200 | } |
| 212 | | - return array( $userid, $username ); |
| 213 | | -} |
| 214 | 201 | |
| 215 | | -function importPage( $title ) |
| 216 | | -{ |
| 217 | | - echo "\n<!-- Importing page " . xmlCommentSafe( $title ) . " -->\n"; |
| 218 | | - $page = fetchPage( $title ); |
| | 202 | + private function importPage( $title ) { |
| | 203 | + echo "\n<!-- Importing page " . $this->xmlCommentSafe( $title ) . " -->\n"; |
| | 204 | + $page = $this->fetchPage( $title ); |
| 219 | 205 | |
| 220 | | - $newtitle = xmlsafe( str_replace( '_', ' ', recodeText( $title ) ) ); |
| | 206 | + $newtitle = $this->xmlsafe( str_replace( '_', ' ', $this->recodeText( $title ) ) ); |
| 221 | 207 | |
| 222 | | - $munged = mungeFormat( $page->text ); |
| 223 | | - if ( $munged != $page->text ) { |
| 224 | | - /** |
| 225 | | - * Save a *new* revision with the conversion, and put the |
| 226 | | - * previous last version into the history. |
| 227 | | - */ |
| 228 | | - $next = array2object( array( |
| 229 | | - 'text' => $munged, |
| 230 | | - 'minor' => 1, |
| 231 | | - 'username' => 'Conversion script', |
| 232 | | - 'host' => '127.0.0.1', |
| 233 | | - 'ts' => time(), |
| 234 | | - 'summary' => 'link fix', |
| 235 | | - ) ); |
| 236 | | - $revisions = array( $page, $next ); |
| 237 | | - } else { |
| 238 | | - /** |
| 239 | | - * Current revision: |
| 240 | | - */ |
| 241 | | - $revisions = array( $page ); |
| 242 | | - } |
| 243 | | - $xml = <<<XML |
| 244 | | - <page> |
| 245 | | - <title>$newtitle</title> |
| | 208 | + $munged = $this->mungeFormat( $page->text ); |
| | 209 | + if ( $munged != $page->text ) { |
| | 210 | + /** |
| | 211 | + * Save a *new* revision with the conversion, and put the |
| | 212 | + * previous last version into the history. |
| | 213 | + */ |
| | 214 | + $next = $this->array2object( array( |
| | 215 | + 'text' => $munged, |
| | 216 | + 'minor' => 1, |
| | 217 | + 'username' => 'Conversion script', |
| | 218 | + 'host' => '127.0.0.1', |
| | 219 | + 'ts' => time(), |
| | 220 | + 'summary' => 'link fix', |
| | 221 | + ) ); |
| | 222 | + $revisions = array( $page, $next ); |
| | 223 | + } else { |
| | 224 | + /** |
| | 225 | + * Current revision: |
| | 226 | + */ |
| | 227 | + $revisions = array( $page ); |
| | 228 | + } |
| | 229 | + $xml = <<<XML |
| | 230 | + <page> |
| | 231 | + <title>$newtitle</title> |
| 246 | 232 | |
| 247 | 233 | XML; |
| 248 | 234 | |
| 249 | | - # History |
| 250 | | - $revisions = array_merge( $revisions, fetchKeptPages( $title ) ); |
| 251 | | - if ( count( $revisions ) == 0 ) { |
| 252 | | - return NULL; // Was "$sql", which does not appear to be defined. |
| 253 | | - } |
| | 235 | + # History |
| | 236 | + $revisions = array_merge( $revisions, $this->fetchKeptPages( $title ) ); |
| | 237 | + if ( count( $revisions ) == 0 ) { |
| | 238 | + return NULL; // Was "$sql", which does not appear to be defined. |
| | 239 | + } |
| 254 | 240 | |
| 255 | | - foreach ( $revisions as $rev ) { |
| 256 | | - $text = xmlsafe( recodeText( $rev->text ) ); |
| 257 | | - $minor = ( $rev->minor ? '<minor/>' : '' ); |
| 258 | | - list( /* $userid */ , $username ) = checkUserCache( $rev->username, $rev->host ); |
| 259 | | - $username = xmlsafe( recodeText( $username ) ); |
| 260 | | - $timestamp = xmlsafe( timestamp2ISO8601( $rev->ts ) ); |
| 261 | | - $comment = xmlsafe( recodeText( $rev->summary ) ); |
| | 241 | + foreach ( $revisions as $rev ) { |
| | 242 | + $text = $this->xmlsafe( $this->recodeText( $rev->text ) ); |
| | 243 | + $minor = ( $rev->minor ? '<minor/>' : '' ); |
| | 244 | + list( /* $userid */ , $username ) = $this->checkUserCache( $rev->username, $rev->host ); |
| | 245 | + $username = $this->xmlsafe( $this->recodeText( $username ) ); |
| | 246 | + $timestamp = $this->xmlsafe( $this->timestamp2ISO8601( $rev->ts ) ); |
| | 247 | + $comment = $this->xmlsafe( $this->recodeText( $rev->summary ) ); |
| 262 | 248 | |
| 263 | | - $xml .= <<<XML |
| 264 | | - <revision> |
| 265 | | - <timestamp>$timestamp</timestamp> |
| 266 | | - <contributor><username>$username</username></contributor> |
| 267 | | - $minor |
| 268 | | - <comment>$comment</comment> |
| 269 | | - <text>$text</text> |
| 270 | | - </revision> |
| | 249 | + $xml .= <<<XML |
| | 250 | + <revision> |
| | 251 | + <timestamp>$timestamp</timestamp> |
| | 252 | + <contributor><username>$username</username></contributor> |
| | 253 | + $minor |
| | 254 | + <comment>$comment</comment> |
| | 255 | + <text>$text</text> |
| | 256 | + </revision> |
| 271 | 257 | |
| 272 | 258 | XML; |
| | 259 | + } |
| | 260 | + $xml .= "</page>\n\n"; |
| | 261 | + return $xml; |
| 273 | 262 | } |
| 274 | | - $xml .= "</page>\n\n"; |
| 275 | | - return $xml; |
| 276 | | -} |
| 277 | 263 | |
| 278 | | -# Whee! |
| 279 | | -function recodeText( $string ) { |
| 280 | | - global $wgImportEncoding; |
| 281 | | - # For currently latin-1 wikis |
| 282 | | - $string = str_replace( "\r\n", "\n", $string ); |
| 283 | | - $string = @iconv( $wgImportEncoding, "UTF-8", $string ); |
| 284 | | - $string = wfMungeToUtf8( $string ); # Any old Ӓ stuff |
| 285 | | - return $string; |
| 286 | | -} |
| | 264 | + private function recodeText( $string ) { |
| | 265 | + # For currently latin-1 wikis |
| | 266 | + $string = str_replace( "\r\n", "\n", $string ); |
| | 267 | + $string = @iconv( $this->encoding, "UTF-8", $string ); |
| | 268 | + $string = $this->mungeToUtf8( $string ); # Any old Ӓ stuff |
| | 269 | + return $string; |
| | 270 | + } |
| 287 | 271 | |
| 288 | | -function wfUtf8Sequence( $codepoint ) { |
| 289 | | - if ( $codepoint < 0x80 ) return chr( $codepoint ); |
| 290 | | - if ( $codepoint < 0x800 ) return chr( $codepoint >> 6 & 0x3f | 0xc0 ) . |
| 291 | | - chr( $codepoint & 0x3f | 0x80 ); |
| 292 | | - if ( $codepoint < 0x10000 ) return chr( $codepoint >> 12 & 0x0f | 0xe0 ) . |
| 293 | | - chr( $codepoint >> 6 & 0x3f | 0x80 ) . |
| 294 | | - chr( $codepoint & 0x3f | 0x80 ); |
| 295 | | - if ( $codepoint < 0x100000 ) return chr( $codepoint >> 18 & 0x07 | 0xf0 ) . # Double-check this |
| 296 | | - chr( $codepoint >> 12 & 0x3f | 0x80 ) . |
| 297 | | - chr( $codepoint >> 6 & 0x3f | 0x80 ) . |
| 298 | | - chr( $codepoint & 0x3f | 0x80 ); |
| 299 | | - # Doesn't yet handle outside the BMP |
| 300 | | - return "&#$codepoint;"; |
| 301 | | -} |
| | 272 | + /** |
| | 273 | + * @fixme don't use /e |
| | 274 | + */ |
| | 275 | + private function mungeToUtf8( $string ) { |
| | 276 | + $string = preg_replace ( '/&#([0-9]+);/e', 'wfUtf8Sequence($1)', $string ); |
| | 277 | + $string = preg_replace ( '/&#x([0-9a-f]+);/ie', 'wfUtf8Sequence(0x$1)', $string ); |
| | 278 | + # Should also do named entities here |
| | 279 | + return $string; |
| | 280 | + } |
| 302 | 281 | |
| 303 | | -function wfMungeToUtf8( $string ) { |
| 304 | | - $string = preg_replace ( '/&#([0-9]+);/e', 'wfUtf8Sequence($1)', $string ); |
| 305 | | - $string = preg_replace ( '/&#x([0-9a-f]+);/ie', 'wfUtf8Sequence(0x$1)', $string ); |
| 306 | | - # Should also do named entities here |
| 307 | | - return $string; |
| 308 | | -} |
| | 282 | + private function timestamp2ISO8601( $ts ) { |
| | 283 | + # 2003-08-05T18:30:02Z |
| | 284 | + return gmdate( 'Y-m-d', $ts ) . 'T' . gmdate( 'H:i:s', $ts ) . 'Z'; |
| | 285 | + } |
| 309 | 286 | |
| 310 | | -function timestamp2ISO8601( $ts ) { |
| 311 | | - # 2003-08-05T18:30:02Z |
| 312 | | - return gmdate( 'Y-m-d', $ts ) . 'T' . gmdate( 'H:i:s', $ts ) . 'Z'; |
| 313 | | -} |
| 314 | | - |
| 315 | | -function xmlsafe( $string ) { |
| 316 | 287 | /** |
| 317 | 288 | * The page may contain old data which has not been properly normalized. |
| 318 | 289 | * Invalid UTF-8 sequences or forbidden control characters will make our |
| 319 | 290 | * XML output invalid, so be sure to strip them out. |
| | 291 | + * @param String $string Text to clean up |
| | 292 | + * @return String |
| 320 | 293 | */ |
| 321 | | - $string = UtfNormal::cleanUp( $string ); |
| | 294 | + private function xmlsafe( $string ) { |
| | 295 | + $string = UtfNormal::cleanUp( $string ); |
| | 296 | + $string = htmlspecialchars( $string ); |
| | 297 | + return $string; |
| | 298 | + } |
| 322 | 299 | |
| 323 | | - $string = htmlspecialchars( $string ); |
| 324 | | - return $string; |
| 325 | | -} |
| | 300 | + private function xmlCommentSafe( $text ) { |
| | 301 | + return str_replace( '--', '\\-\\-', $this->xmlsafe( $this->recodeText( $text ) ) ); |
| | 302 | + } |
| 326 | 303 | |
| 327 | | -function xmlCommentSafe( $text ) { |
| 328 | | - return str_replace( '--', '\\-\\-', xmlsafe( recodeText( $text ) ) ); |
| 329 | | -} |
| 330 | | - |
| 331 | | - |
| 332 | | -function array2object( $arr ) { |
| 333 | | - $o = (object)0; |
| 334 | | - foreach ( $arr as $x => $y ) { |
| 335 | | - $o->$x = $y; |
| | 304 | + private function array2object( $arr ) { |
| | 305 | + $o = (object)0; |
| | 306 | + foreach ( $arr as $x => $y ) { |
| | 307 | + $o->$x = $y; |
| | 308 | + } |
| | 309 | + return $o; |
| 336 | 310 | } |
| 337 | | - return $o; |
| 338 | | -} |
| 339 | 311 | |
| | 312 | + /** |
| | 313 | + * Make CamelCase and /Talk links work |
| | 314 | + */ |
| | 315 | + private function mungeFormat( $text ) { |
| | 316 | + $this->nowiki = array(); |
| | 317 | + $staged = preg_replace_callback( |
| | 318 | + '/(<nowiki>.*?<\\/nowiki>|(?:http|https|ftp):\\S+|\[\[[^]\\n]+]])/s', |
| | 319 | + array( $this, 'nowikiPlaceholder' ), $text ); |
| 340 | 320 | |
| 341 | | -/** |
| 342 | | - * Make CamelCase and /Talk links work |
| 343 | | - */ |
| 344 | | -function mungeFormat( $text ) { |
| 345 | | - global $nowiki; |
| 346 | | - $nowiki = array(); |
| 347 | | - $staged = preg_replace_callback( |
| 348 | | - '/(<nowiki>.*?<\\/nowiki>|(?:http|https|ftp):\\S+|\[\[[^]\\n]+]])/s', |
| 349 | | - 'nowikiPlaceholder', $text ); |
| | 321 | + # This is probably not 100% correct, I'm just |
| | 322 | + # glancing at the UseModWiki code. |
| | 323 | + $upper = "[A-Z]"; |
| | 324 | + $lower = "[a-z_0-9]"; |
| | 325 | + $any = "[A-Za-z_0-9]"; |
| | 326 | + $camel = "(?:$upper+$lower+$upper+$any*)"; |
| | 327 | + $subpage = "(?:\\/$any+)"; |
| | 328 | + $substart = "(?:\\/$upper$any*)"; |
| 350 | 329 | |
| 351 | | - # This is probably not 100% correct, I'm just |
| 352 | | - # glancing at the UseModWiki code. |
| 353 | | - $upper = "[A-Z]"; |
| 354 | | - $lower = "[a-z_0-9]"; |
| 355 | | - $any = "[A-Za-z_0-9]"; |
| 356 | | - $camel = "(?:$upper+$lower+$upper+$any*)"; |
| 357 | | - $subpage = "(?:\\/$any+)"; |
| 358 | | - $substart = "(?:\\/$upper$any*)"; |
| | 330 | + $munged = preg_replace( "/(?!\\[\\[)($camel$subpage*|$substart$subpage*)\\b(?!\\]\\]|>)/", |
| | 331 | + '[[$1]]', $staged ); |
| 359 | 332 | |
| 360 | | - $munged = preg_replace( "/(?!\\[\\[)($camel$subpage*|$substart$subpage*)\\b(?!\\]\\]|>)/", |
| 361 | | - '[[$1]]', $staged ); |
| | 333 | + $final = preg_replace( '/' . preg_quote( $this->placeholder() ) . '/s', |
| | 334 | + array( $this, 'nowikiShift' ), $munged ); |
| | 335 | + return $final; |
| | 336 | + } |
| 362 | 337 | |
| 363 | | - $final = preg_replace( '/' . preg_quote( placeholder() ) . '/es', |
| 364 | | - 'array_shift( $nowiki )', $munged ); |
| 365 | | - return $final; |
| 366 | | -} |
| | 338 | + private function placeholder( $x = null ) { |
| | 339 | + return '\xffplaceholder\xff'; |
| | 340 | + } |
| 367 | 341 | |
| | 342 | + public function nowikiPlaceholder( $matches ) { |
| | 343 | + $this->nowiki[] = $matches[1]; |
| | 344 | + return $this->placeholder(); |
| | 345 | + } |
| 368 | 346 | |
| 369 | | -function placeholder( $x = null ) { |
| 370 | | - return '\xffplaceholder\xff'; |
| | 347 | + public function nowikiShift() { |
| | 348 | + return array_shift( $this->nowiki ); |
| | 349 | + } |
| 371 | 350 | } |
| 372 | 351 | |
| 373 | | -function nowikiPlaceholder( $matches ) { |
| 374 | | - global $nowiki; |
| 375 | | - $nowiki[] = $matches[1]; |
| 376 | | - return placeholder(); |
| | 352 | +function wfUtf8Sequence( $codepoint ) { |
| | 353 | + if ( $codepoint < 0x80 ) { |
| | 354 | + return chr( $codepoint ); |
| | 355 | + } |
| | 356 | + if ( $codepoint < 0x800 ) { |
| | 357 | + return chr( $codepoint >> 6 & 0x3f | 0xc0 ) . |
| | 358 | + chr( $codepoint & 0x3f | 0x80 ); |
| | 359 | + } |
| | 360 | + if ( $codepoint < 0x10000 ) { |
| | 361 | + return chr( $codepoint >> 12 & 0x0f | 0xe0 ) . |
| | 362 | + chr( $codepoint >> 6 & 0x3f | 0x80 ) . |
| | 363 | + chr( $codepoint & 0x3f | 0x80 ); |
| | 364 | + } |
| | 365 | + if ( $codepoint < 0x100000 ) { |
| | 366 | + return chr( $codepoint >> 18 & 0x07 | 0xf0 ) . # Double-check this |
| | 367 | + chr( $codepoint >> 12 & 0x3f | 0x80 ) . |
| | 368 | + chr( $codepoint >> 6 & 0x3f | 0x80 ) . |
| | 369 | + chr( $codepoint & 0x3f | 0x80 ); |
| | 370 | + } |
| | 371 | + # Doesn't yet handle outside the BMP |
| | 372 | + return "&#$codepoint;"; |
| 377 | 373 | } |
| 378 | 374 | |
| 379 | | - |
| | 375 | +$maintClass = 'ImportUseModWiki'; |
| | 376 | +require_once( RUN_MAINTENANCE_IF_MAIN ); |