Arabic encoding (mojibake) checker
Does your Arabic turn into strange symbols because of the wrong encoding?
- Checks your page
- Reads:HTML
- 1 rule
- Last updated
What we keep
The result appears here
Enter your page’s address and press “Check the page”. This is what the tool runs on it:
- Garbled Arabic text (mojibake)not started
What it checks
- UTF-8 Arabic read as Windows-1252, the commonest case:
منتجfor «منتج». - Windows-1256 Arabic (the old Arabic code page) read as Windows-1252:
ÇáÚÑÈíÉfor «العربية». - That a word counts only when decoding it again gives Arabic letters and no Latin ones, so «Café» and «señor» are left alone.
- The first run of garbled words in each text, with what it said.
Example
<p>آراء العملاء في عطر الورد الطائفي خلال الشهر الماضي.</p> <blockquote>منتج رائع وصلني بسرعة</blockquote>
<p>آراء العملاء في عطر الورد الطائفي خلال الشهر الماضي.</p> <blockquote>منتج رائع وصلني بسرعة</blockquote>
How to fix
- Save every file as UTF-8 and serve it with
Content-Type: text/html; charset=utf-8; add<meta charset="utf-8" />as the first tag in<head>. - If only one part is garbled (a review, a product name, an imported field), the data was stored or imported in the wrong encoding: re-import it as UTF-8, or convert the database column (
utf8mb4in MySQL). - If the garbled text came from a feed, a plugin or an import file, fix the source file's encoding, not the page.
- After fixing, clear the page cache.
- On Salla, Zid and Shopify, garbled text usually came in through a product import: re-import the file as UTF-8 (in Excel, save as "CSV UTF-8").
For example:
<meta charset="utf-8" />FAQ
Because the bytes are UTF-8 and the program reading them thinks they are Windows-1252 (or Latin-1). Each Arabic letter is two bytes in UTF-8, and each byte shows as one Latin symbol.
This tool reads the result: garbled text that is on the page. The cause is almost always a charset mismatch, but a page can declare UTF-8 correctly and still hold garbled text that was stored that way.
Related, but different: question marks mean the letters were lost when the data was converted to an encoding that cannot hold Arabic. They cannot be recovered; the source must be re-imported.
No, but it shows what each garbled word said, so that you can check the repair.
What this tool runs
- Garbled Arabic text (mojibake)serious
ar-mojibake
How we measure
We read the visible text of the page's HTML as the server sends it, decode it as the browser does, and look for words made of Latin letters such as Ø, Ù, Ç and á. Each word is turned back into its Windows-1252 bytes and read as UTF-8 or Windows-1256; it counts only when the result is two or more Arabic letters and no Latin letter. See the rule's page for the details.