830521 - with precompiled l10n we fire "localized" event twice

Status: RESOLVED FIXED

(Firefox OS Graveyard :: Gaia, defect)

Comment 1

[Zibi Braniecki [:zbraniecki][:gandalf]]

Attachment: Pointer to Github pull request: https://github.com/mozilla-b2g/gaia/pull/7602

Comment 2

[Zibi Braniecki [:zbraniecki][:gandalf]]

ok. this bug is not as innocent as I thought. Because we rarely remove the listener, we actually fire all the code that depends on the event in all apps twice, which contributes to startup timing.

Comment 3

[James Lal [:lightsofapollo]]

A good workaround to this is to unregister the localized handler when starting your app (which you should really do anyway since the event will fire again later if the user changes locales...)

Comment 4

[David Scravaglieri [:scravag]]

:kaze, what do you think about that ?

Comment 5

[Fabien Cazenave [:kaze]]

Like I wrote on m.d.gaia when djf reported this bug a few days ago, I sure agree it should be fixed. This patch works but it relies on the fact that `loadLocale' is called twice — which is what should be fixed, precisely.

Give me a couple hours, I’ll r+ Gandalf’s patch if I can’t find any better fix.

Comment 6

[Zibi Braniecki [:zbraniecki][:gandalf]]

(In reply to Fabien Cazenave [:kaze] from comment #5)
>  This patch works but it relies on the fact that
> `loadLocale' is called twice

um. no. loadLocale is called once. It just calls fireL10nReadyEvent twice without this patch.

Comment 7

[Vivien Nicolas (:vingtetun) (:21) - (NOT reading bugmails, needinfo? please)]

Looking at the code I would say the right fix will be to omit the second argument of loadLocale at line 942. The early return in the patch will not set the state of l10n as 'complete'.

Comment 8

[Zibi Braniecki [:zbraniecki][:gandalf]]

(In reply to Vivien Nicolas (:vingtetun) from comment #7)
> Looking at the code I would say the right fix will be to omit the second
> argument of loadLocale at line 942. The early return in the patch will not
> set the state of l10n as 'complete'.

Good point.
Updated the pull request

Comment 9

[Fabien Cazenave [:kaze]]

Comment on attachment 702000 [details]
Pointer to Github pull request: https://github.com/mozilla-b2g/gaia/pull/7602

I much prefer this approach. Gandalf, Vivien, thanks for your help!

[Approval Request Comment]
Bug caused by (feature/regressing bug #): 
User impact if declined: 
Testing completed: 
Risk to taking this patch (and alternatives if risky):

Comment 10

[Vivien Nicolas (:vingtetun) (:21) - (NOT reading bugmails, needinfo? please)]

https://github.com/mozilla-b2g/gaia/commit/d194ce5bd86a6ed16d453d02d0cb7842e47e2ad3