devol/wallabag
1
0
Fork 0
mirror of https://github.com/wallabag/wallabag.git synced 2024-12-15 09:57:41 +01:00
Code Issues Packages Projects Releases Wiki Activity

Bugfix: Sanitize the title of a saved webpage from invalid UTF-8 characters

Browse Source
This commit is contained in:
Tobi823 2018-09-18 15:04:19 +02:00
parent 759c91940b
commit d76a5a6d60
1 changed files with 23 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
src/Wallabag/CoreBundle/Helper/ContentProxy.php
View File

@ -53,6 +53,7 @@ class ContentProxy
if ((empty($content) || false === $this->validateContent($content)) && false === $disableContentUpdate) {
$fetchedContent = $this->graby->fetchContent($url);
$fetchedContent['title'] = $this->sanitizeUTF8Text($fetchedContent['title']);
// when content is imported, we have information in $content
// in case fetching content goes bad, we'll keep the imported information instead of overriding them
@ -68,6 +69,28 @@ class ContentProxy
$this->stockEntry($entry, $content);
}
/**
* Remove invalid UTF-8 characters from the given string in following steps:
* - try to interpret the given string as ISO-8859-1, convert it to UTF-8 and return it (if its valid)
* - simply remove every invalid UTF-8 character and return the result (https://stackoverflow.com/a/1433665)
* @param String $rawText
* @return string
*/
private function sanitizeUTF8Text(String $rawText) {
if (mb_check_encoding($rawText, 'utf-8')) {
return $rawText; // return because its valid utf-8 text
}
// we assume that $text is encoded in ISO-8859-1 (and not the similar Windows-1252 or other encoding)
$convertedText = utf8_encode($rawText);
if (mb_check_encoding($convertedText, 'utf-8')) {
return $convertedText;
}
// last resort: simply remove invalid UTF-8 character because $rawText can have some every exotic encoding
return iconv("UTF-8", "UTF-8//IGNORE", $rawText);
}
/**
* Use a Symfony validator to ensure the language is well formatted.
*