ZF-9211: 2nd message for Zend_Validate_File_IsImage is different between English file and Russian one.
Description
I find difference between English file and Russian one in resource/language .
I think following message for Russian at line 139, {quote} The mimetype of file '%value%' could not been detected {quote}
should be {quote} The mimetype of file '%value%' could not be detected {quote}
Comments
Posted by Thomas Weidner (thomas) on 2010-02-19T11:09:00.000+0000
I am no russian docu member and I don't speak russian. Unassigning the issue from me.
Posted by Maghiel Dijksman (maghiel) on 2010-02-20T12:17:29.000+0000
Patch attached.
I've checked the other languages to be sure, they are all correct.
Posted by Thomas Weidner (thomas) on 2010-02-20T12:41:19.000+0000
Maghiel: How can you check other languages? Do you speak 8 different languages? ;-)
Hint: When the source changes than also the translation will change. In german for example "has be" and "has been" have different meaning in translation. And many other languages behave the same way. So simply changing the source is not correct as long as you are not a native russian speaker.
Posted by Maghiel Dijksman (maghiel) on 2010-02-20T13:40:26.000+0000
@Thomas: clear! I assumed it was just a simple question of a miss-spelled key. (I still think it's a misspelled key, as all languages contained the same source, except for Russian. But I cannot be sure so you're definatly right i cannot resolve this issue).
Well then Reopened and unassigned from me :p
Posted by Maghiel Dijksman (maghiel) on 2010-02-20T13:41:03.000+0000
A Russian speaking member should check the russian translation.
Posted by Dolf Schimmel (Freeaqingme) (freak) on 2010-02-20T18:15:10.000+0000
Thomas, the key differs from the value. The value should obviously be checked by someone who speaks Russian (pref. natively. Alex?), but the keys should be the same in all language files.
Posted by Thomas Weidner (thomas) on 2010-02-21T07:33:39.000+0000
Dolf: The unittest shows that the key differs from the original english resource.
This means that the translation has been made by using a wrong key. Therefor also the value has to be seen as incorrect as long as it has been verified by a russian speaking person.
Posted by Dolf Schimmel (Freeaqingme) (freak) on 2010-02-21T07:34:35.000+0000
Correct, but then it still makes sense to at least correct the key?
Posted by Thomas Weidner (thomas) on 2010-02-21T07:45:11.000+0000
No one corrected the key, because then the value is incorrect. And thats the reason why the unittest fails.
I don't get your problem.
To be on the correct side we could in the meantime delete the related key/value pair, but why making the work of others more problematic?
Posted by Maghiel Dijksman (maghiel) on 2010-02-21T08:22:47.000+0000
Well, i did 'correct' it, but then took it back :p
Posted by Satoru Yoshida (satoruyoshida) on 2010-02-28T02:35:42.000+0000
Thank All for comments, I am not native, but I have just checked both key string and value one for SVN r20377 .
Posted by Thomas Weidner (thomas) on 2010-02-28T08:09:07.000+0000
@Saturo: Note that when you would have done this for german, then you would have broken the validation as the meaning for the translation would have changed.
Please note that only a native translator should change the key and value pair, otherwise the validation is surely broken.