Text orientation problem

[AhmedBenmoussa]

Hello,

I'm the TM of the [ar-DZ] language. I would report a problem.
I have a little problem with the punctuations, dots, commas, etc... As you know the Arabic language is written from right to left. When I started translating the tokens, everything was good (the dots, commas, special characters % ...) were all at the right place. But when I validated the tokens, it changed. the sentence's orientation has changed from (right -> left) to (left -> wright). They are mixed.
Here's an exemple:

maybe it's just a text orientation problem. maybe changing the text orientation (to right->left) when the language chosen is Arabic will resolve the problem. like in css:

div {
direction: rtl;
} 

I found a trick to write them properly. To do that I have to start writing the tokens from the end. When I validate them, the sentences retrieve their right places. The problem with that method is: What if the developers change the text orientation. They will be all mixed again and I will have to do the work again.

What should I do ? writing them normally without changing them and hoping that the developers will resolve the problem. Or should I write them with the second method with the fear that if the developers change the text orientation they will be mixed again.

Thank you,
Ahmed.