Closed Bug 886715 Opened 11 years ago Closed 11 years ago

[MMS] [Regression] After changing language all strings should be in new language

Categories

(Firefox OS Graveyard :: Gaia::SMS, defect)

ARM
Gonk (Firefox OS)
defect
Not set
normal

Tracking

(blocking-b2g:koi+)

RESOLVED FIXED
1.1 QE3 (26jun)
blocking-b2g koi+

People

(Reporter: borjasalguero, Assigned: gnarf)

References

Details

(Keywords: regression, Whiteboard: MMS_TEF, )

Attachments

(2 files, 1 obsolete file)

Attached image Issue
Changing the language is not working properly for the composer-header. Check attachment.
Summary: [MMS] [Regression] MMS Header is not localized → [MMS] [Regression] MMS Header in composer is not localized
Attached file Pull Request (obsolete) —
Attachment #767080 - Flags: review?
Assignee: nobody → fbsc
This is a regression, leo+?
blocking-b2g: --- → leo?
Whiteboard: MMS_TEF, TaipeiWW
Assignee: fbsc → gnarf37
Comment on attachment 767080 [details]
Pull Request

Corey is going to take a look because this is happening in other places as well.
Attachment #767080 - Flags: review?
We have detected that this is happening in more places, so Corey is going to take a look. Dietrich, this is a regression, leo+?
Flags: needinfo?(dietrich)
Summary: [MMS] [Regression] MMS Header in composer is not localized → [MMS] [Regression] Localization is not working properly in some scenarios
Keywords: regression
Summary: [MMS] [Regression] Localization is not working properly in some scenarios → [MMS] [Regression] After changing language all strings should be in new language
There are many strings, generally the header in the messages view, the download mms strings, etc are all not properly localized when the language is changed while the app is running. I'm sure there are more as well
blocking-b2g: leo? → leo+
Target Milestone: --- → 1.1 QE3 (26jun)
I'd like to argue for this not being QE3 - and also not being leo+

Are there many people who will be switching language with open applications and expecting everything to change immediately?  The app fixes itself after changing pages anyway.

Visibility seems pretty low, and complexity of the fix is fairly high (because it affects much more than just the header)

Renominating.
blocking-b2g: leo+ → leo?
Comment on attachment 767080 [details]
Pull Request

Only solves one of the problem cases, and there are also better ways to solve.
Attachment #767080 - Attachment is obsolete: true
Status: NEW → ASSIGNED
leo-, as most users will not change languages often, and as comment 6 notes that it can be fixed within itself after changing pages within the app.   renom if you feel otherwise with a comment.
blocking-b2g: leo? → -
Whiteboard: MMS_TEF, TaipeiWW → MMS_TEF,
Flags: needinfo?(dietrich)
Depends on: 882591
This might be solved by the patch I’ve done for bug 898992:
https://github.com/mozilla-b2g/gaia/pull/11304

Corey, please ask me to review your patch if you work on this — or feel free to re-assign this bug to me.
Depends on: 898992
(In reply to Fabien Cazenave [:kaze] from comment #9)
> This might be solved by the patch I’ve done for bug 898992:
> https://github.com/mozilla-b2g/gaia/pull/11304
> 
> Corey, please ask me to review your patch if you work on this — or feel free
> to re-assign this bug to me.

I'm using that work to fix this.  The problem is we do a lot of mozL10n.get(string) and stuff innerText with it.  Were we should be instead using .localize or .translate and setting the right l10n properties on the elements.  I'm just doing a giant audit of all our uses of l10n for this ticket to make sure we get all of the places.
If you want to take a look at it - I've been pushing up my progress pretty much every major .js file -  https://github.com/gnarf/gaia/compare/886715
Depends on: 902363
Attached file pull request on github
Looking for feedback from kaze/rick, and review from julien on this pull.

Any other input also welcome - thanks!
Attachment #789138 - Flags: review?(felash)
Attachment #789138 - Flags: feedback?(waldron.rick)
Attachment #789138 - Flags: feedback?(kaze)
Comment on attachment 789138 [details] [review]
pull request on github

made a first pass but I have not looked at the test yet, neither tried it in the device. I keep this for tomorrow.

Reviewint this I've seen a lot of dialogs whose markup is inserted using JS whereas we could have it in index.html instead.
Comment on attachment 789138 [details] [review]
pull request on github

patch works well to me, but still some questions have to be answered regarding some functionality to be moved to l10n.js
Dear mozilla, marked koi? to make sure these issues ok for v1.2
blocking-b2g: - → koi?
Comment on attachment 789138 [details] [review]
pull request on github

r=me with the latest nit

check that the tests still pass locally before merging !
Attachment #789138 - Flags: review?(felash)
Attachment #789138 - Flags: review+
Attachment #789138 - Flags: feedback?(waldron.rick)
Attachment #789138 - Flags: feedback?(kaze)
master: https://github.com/mozilla-b2g/gaia/commit/359b24a12724f1f8227e4d5b99a60bc3050846af
Status: ASSIGNED → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
blocking-b2g: koi? → koi+
FTR, this patch introduces a new user-facing string (“view-attachment”) and should add it to the *.properties file. Filing a follow-up.
Depends on: 911709
Yeah, we eventually renamed it to "view-attachment-other".
Attachment mime type: text/plain → text/x-github-pull-request
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: