Revision 442
Added by Matthias over 18 years ago
| functions.php | ||
|---|---|---|
| 338 | 338 |
return $subs; |
| 339 | 339 |
} |
| 340 | 340 |
|
| 341 |
// Function as replecement for php's htmlspecialchars() |
|
| 342 |
function my_htmlspecialchars($string) {
|
|
| 343 |
$string = umlauts_to_entities($string); |
|
| 344 |
$string = entities_to_umlauts($string); |
|
| 345 |
return($string); |
|
| 346 |
} |
|
| 347 |
|
|
| 348 |
// Function to get the DEFAULT_CHARSET |
|
| 349 |
function get_wbcharset() {
|
|
| 350 |
$charset=strtoupper(DEFAULT_CHARSET); |
|
| 351 |
if(strcmp($charset,"BIG5") == 0) {
|
|
| 352 |
$charset="BIG-5"; |
|
| 353 |
} |
|
| 354 |
return($charset); |
|
| 355 |
} |
|
| 356 |
|
|
| 357 |
// Function to convert a string from $from- to $to-encoding, using mysql |
|
| 358 |
function my_mysql_iconv($string, $from, $to) {
|
|
| 359 |
// keep current character set values: |
|
| 360 |
$character_set_database = mysql_result(mysql_query("SELECT @@character_set_client"),0,0);
|
|
| 361 |
$character_set_results = mysql_result(mysql_query("SELECT @@character_set_results"),0,0);
|
|
| 362 |
$collation_results = mysql_result(mysql_query("SELECT @@collation_connection"),0,0);
|
|
| 363 |
mysql_query("SET character_set_client=$from");
|
|
| 364 |
mysql_query("SET character_set_results=$to");
|
|
| 365 |
mysql_query("SET collation_connection=utf8_unicode_ci");
|
|
| 366 |
$string_escaped = mysql_real_escape_string($string); |
|
| 367 |
$converted_string = mysql_result(mysql_query("SELECT '$string_escaped'"),0,0);
|
|
| 368 |
// restore previous character set values: |
|
| 369 |
mysql_query("SET character_set_client=$character_set_database");
|
|
| 370 |
mysql_query("SET character_set_results=$character_set_results");
|
|
| 371 |
mysql_query("SET collation_connection=$collation_results");
|
|
| 372 |
return $converted_string; |
|
| 373 |
} |
|
| 374 |
|
|
| 375 |
// Function to convert a string from html-entities to umlauts |
|
| 376 |
// and encode htmlspecialchars |
|
| 377 |
function entities_to_umlauts($string) {
|
|
| 378 |
$charset = get_wbcharset(); |
|
| 379 |
// there's no GB2312 or ISO-8859-11 encoding in php's mb_* functions |
|
| 380 |
if (strcmp($charset,"GB2312") == 0) {
|
|
| 381 |
if(function_exists('iconv')) {
|
|
| 382 |
$string=mb_convert_encoding($string,'UTF-8','HTML-ENTITIES'); |
|
| 383 |
$string=iconv("UTF-8","GB2312",$string);
|
|
| 384 |
} else {
|
|
| 385 |
$string=mb_convert_encoding($string,'UTF-8','HTML-ENTITIES'); |
|
| 386 |
$string=my_mysql_iconv($string, 'utf8', 'gb2312'); |
|
| 387 |
} |
|
| 388 |
} elseif (strcmp($charset,"ISO-8859-11") == 0) {
|
|
| 389 |
if(function_exists('iconv')) {
|
|
| 390 |
$string=mb_convert_encoding($string,'UTF-8','HTML-ENTITIES'); |
|
| 391 |
$string=iconv("UTF-8","ISO-8859-11",$string);
|
|
| 392 |
} else {
|
|
| 393 |
$string=mb_convert_encoding($string,'UTF-8','HTML-ENTITIES'); |
|
| 394 |
$string=my_mysql_iconv($string, 'utf8', 'tis620'); |
|
| 395 |
} |
|
| 396 |
} else {
|
|
| 397 |
$string=mb_convert_encoding($string,$charset,'HTML-ENTITIES'); |
|
| 398 |
} |
|
| 399 |
$string=htmlspecialchars($string); |
|
| 400 |
return($string); |
|
| 401 |
} |
|
| 402 |
|
|
| 403 |
// Function to convert a string from umlauts to html-entities |
|
| 404 |
// and encode htmlspecialchars |
|
| 405 |
function umlauts_to_entities($string) {
|
|
| 406 |
$charset=get_wbcharset(); |
|
| 407 |
// there's no GB2312 or ISO-8859-11 encoding in php's mb_* functions |
|
| 408 |
if (strcmp($charset,"GB2312") == 0) {
|
|
| 409 |
if(function_exists('iconv')) {
|
|
| 410 |
$string=iconv("GB2312","UTF-8",$string);
|
|
| 411 |
$charset="UTF-8"; |
|
| 412 |
} else {
|
|
| 413 |
$string=my_mysql_iconv($string, 'gb2312', 'utf8'); |
|
| 414 |
$charset="UTF-8"; |
|
| 415 |
} |
|
| 416 |
} elseif (strcmp($charset,"ISO-8859-11") == 0) {
|
|
| 417 |
if(function_exists('iconv')) {
|
|
| 418 |
$string=iconv("ISO-8859-11","UTF-8",$string);
|
|
| 419 |
$charset="UTF-8"; |
|
| 420 |
} else {
|
|
| 421 |
$string=my_mysql_iconv($string, 'tis620', 'utf8'); |
|
| 422 |
$charset="UTF-8"; |
|
| 423 |
} |
|
| 424 |
} |
|
| 425 |
$string=mb_convert_encoding($string,'HTML-ENTITIES',$charset); |
|
| 426 |
$string=mb_convert_encoding($string,'UTF-8','HTML-ENTITIES'); |
|
| 427 |
$string=htmlspecialchars($string,ENT_QUOTES); |
|
| 428 |
$string=mb_convert_encoding($string,'HTML-ENTITIES','UTF-8'); |
|
| 429 |
return($string); |
|
| 430 |
} |
|
| 431 |
|
|
| 432 |
// translate any "latin" html-entities to their plain 7bit equivalents |
|
| 433 |
function entities_to_7bit($string) {
|
|
| 434 |
require(WB_PATH.'/framework/convert.php'); |
|
| 435 |
$string = strtr($string, $conversion_array); |
|
| 436 |
return($string); |
|
| 437 |
} |
|
| 438 |
|
|
| 341 | 439 |
// Function to convert a page title to a page filename |
| 342 | 440 |
function page_filename($string) {
|
| 343 |
// First, translate any non-english characters to their english equivalents |
|
| 344 |
require(WB_PATH.'/framework/convert.php'); |
|
| 345 |
$string = strtr($string, $conversion_array); |
|
| 441 |
$string = entities_to_7bit(umlauts_to_entities($string)); |
|
| 346 | 442 |
// Now replace spaces with page spcacer |
| 347 | 443 |
$string = str_replace(' ', PAGE_SPACER, $string);
|
| 348 | 444 |
// Now remove all bad characters |
| ... | ... | |
| 371 | 467 |
|
| 372 | 468 |
// Function to convert a desired media filename to a clean filename |
| 373 | 469 |
function media_filename($string) {
|
| 374 |
// First, translate any non-english characters to their english equivalents |
|
| 375 |
require(WB_PATH.'/framework/convert.php'); |
|
| 376 |
$string = strtr($string, $conversion_array); |
|
| 470 |
$string = entities_to_7bit(umlauts_to_entities($string)); |
|
| 377 | 471 |
// Now remove all bad characters |
| 378 | 472 |
$bad = array( |
| 379 | 473 |
'\'', // ' |
Also available in: Unified diff
Fixed issues with some languages when UTF8 is used.
Removed all htmlentites from the code (revoked changeset [396]) and added new functions instead.
Adapted the search and the highlighting to use the new functions.
With this changes WB can now be used with charset UTF8 for all languages.