320147 - Japanese L10N update for Thunderbird 1.5 release

Status: RESOLVED FIXED

(Mozilla Localizations :: ja / Japanese, defect)

Description

[dynamis (Tomoya ASAI)]

This is a bug to request approval about Japanese l10n update.
# trademark-relating problem will not be included here

Comment 1

[dynamis (Tomoya ASAI)]

Attachment: ja.diff

This patch is a little large but includes some critical updates and this change
must be included before 1.5 release.

Comment 2

[dynamis (Tomoya ASAI)]

Attachment: ja-JP-mac.diff

same for ja-JP-mac

Comment 3

[Axel Hecht [:Pike]]

Comment on attachment 205746 [details] [diff] [review]
ja.diff

We won't take changes to browser or toolkit et al at this point.

Comment 4

[dynamis (Tomoya ASAI)]

Attachment: ja.diff

This new diff includes only one fix against firefox relating resource:
# other diff relating firefox is removed

The one fix is:
diff -r -u -w ja/toolkit/chrome/global/intl.css ja.new/toolkit/chrome/global/intl.css
--- ja/toolkit/chrome/global/intl.css	2005-07-07 06:06:31.631000000 +0900
+++ ja.new/toolkit/chrome/global/intl.css	2005-07-07 06:06:31.631000000 +0900
@@ -12,5 +12,9 @@
 }#reportForm #description {
   max-height: 6.8em;
 }
+/* see bug 306166 */
+#OCSPDialog {
+  min-width: 40em;
+}
 
Without this, japanese firefox cannot set OSCP service URL on Mac, so far.
So, I request approval about this patch, which includes one fix against firefox.

Comment 5

[dynamis (Tomoya ASAI)]

Of cource, the workaround in intl.css is needed for thunderbird 1.5 too.

Comment 6

[dynamis (Tomoya ASAI)]

Attachment: ja.diff

Removed fix in intl.css (patch for thunderbird only files).

Comment 7

[Axel Hecht [:Pike]]

bug 306166 suggests that this was a font size change somewhere, can you track down
the original change? That will be vital for my assessment of the land we're in 
here, I wonder if that was a post fx1.5 l10n-impact change.

Comment 8

[dynamis (Tomoya ASAI)]

Attachment: ja-JP-mac.diff

Same for ja-JP-mac.

Comment 9

[Axel Hecht [:Pike]]

Comment on attachment 205904 [details] [diff] [review]
ja.diff

Please submit minimal patches. This patch has complete files that are just affected by single comment changes.

There is a conversion from utf-8 to java encoding in at least one property file, is that an actual change or just noise? If so, it shouldn't be in approval patches.

Comment 10

[dynamis (Tomoya ASAI)]

Conversion from utf-8 to escaped-unicode is actual change.
Though some of them are very minor fix but all diff within the patch is actual.
# 3 ja/ja-JP-mac file for thunderbird was not escaped properly.
# This was cased by bug of my script which generates ja/ja-JP-mac from one set
# of files.

Comment 11

[dynamis (Tomoya ASAI)]

(In reply to comment #7)
> bug 306166 suggests that this was a font size change somewhere, can you track
> down
> the original change? That will be vital for my assessment of the land we're in 
> here, I wonder if that was a post fx1.5 l10n-impact change.

See bug 306166#c8 about font size change. I think we shouldn't use different
font size only for OCSPDialog and we should change the window.width:
http://lxr.mozilla.org/mozilla1.8/source/toolkit/locales/en-US/chrome/mozapps/preferences/ocsp.dtd#1
  -<!ENTITY window.width                 "38em">
  +<!ENTITY window.width                 "40em">
http://lxr.mozilla.org/l10n-mozilla1.8/source/ja/toolkit/chrome/mozapps/preferences/ocsp.dtd#1
  -<!ENTITY window.width                   "38em">
  +<!ENTITY window.width                   "40em">
# I withdraw the change in intl.css

Comment 12

[dynamis (Tomoya ASAI)]

Attachment: ja.minimal.diff

(In reply to comment #9)
> (From update of attachment 205904 [details] [diff] [review] [edit])
> Please submit minimal patches. This patch has complete files that are just
> affected by single comment changes.

Sorry taking your time including quite minor change in my patch.
This is a patch without comment only changes but if possible, I'd like to
commit previous patch.
So if you denied previous patch because it include encoding change in properties, which is actually needed as I wrote in #10, not because comment only changes,
please ignore this and approve previous one (attachment 205904 [details] [diff] [review]).
# this new patch is made just by removing comment only diff from previous one
# substantial change is same as attachment 205904 [details] [diff] [review]
# of cource, this patch doesn't affect firefox same as previous one

Comment 13

[Axel Hecht [:Pike]]

Comment on attachment 205964 [details] [diff] [review]
ja.minimal.diff

I was really close to give this an a-. Please don't use sucky tools that mess with source format more than code changes.
We're in tight timing here and noisy patches carry a HUGE risk.

Please submit a patch for the dialog size only, too. I wonder why the screen shot for fx looks different to the tb one (though that was deerpark for fx, so it might look different in release).
I want to chat with mscott on wether we need to take that or not, could we get screenshots for real fx1.5 and the current nightly builds in here so that we have all info at hand?

Comment 14

[Axel Hecht [:Pike]]

Comment on attachment 205906 [details] [diff] [review]
ja-JP-mac.diff

please land the minimal change

Comment 15

[dynamis (Tomoya ASAI)]

Attachment: ja-JP-mac.minimal.diff

same as ja.minimal.diff for ja-JP-mac.

Comment 16

[dynamis (Tomoya ASAI)]

Thanks approving. Committed minimal changes for ja/ja-JP-mac.

(In reply to comment #13)
> I wonder why the screen
> shot for fx looks different to the tb one (though that was deerpark for fx, so
> it might look different in release).
What you mean? If you are talking about difference between attachment 194033 [details]
and 200439 in the bug 306166, the fx one is en-US and the tb one is ja-JP-mac.
That cause different height required.
# first text of the dialog is a little longer in ja-JP-mac

> I want to chat with mscott on wether we need to take that or not, could we get
> screenshots for real fx1.5 and the current nightly builds in here so that we
> have all info at hand?
I'm requesting my peer to check and take screenshots...
After that, I'll submit a patch and attach screenshots.
# I don't have mac now... :(

Comment 17

[Hiro]

Attachment: Firefox1.5 Final English version

Comment 18

[Hiro]

Attachment: Firefox1.5 Final Japanese version

Comment 19

[Hiro]

Attachment: Firefox1.5 L10n nightly version

Use package:
http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly/latest-mozilla1.8-1.5-l10n/firefox-1.5.ja-JP-mac.mac.dmg

Comment 20

[Hiro]

Attachment: Firefox1.5 nightly + new JLP

Firefox1.5 nightlybuild and the following language package.
http://ftp.mozilla-japan.org/pub/mozilla-japan/firefox/development/1.5/firefox-1.5.ja-JP-mac.langpack-1.5.0.1.xpi

Comment 21

[dynamis (Tomoya ASAI)]

Attachment: ocspdialog.diff

This is the patch to fix OCSPDialog for ja/ja-JP-mac.

(In reply to comment #20)
> Created an attachment (id=206078) [edit]
> Firefox1.5 nightly + new JLP
> 
> Firefox1.5 nightlybuild and the following language package.
This means that he took the last screenshot with langpack which includes change
of the window.width. As you see, this patch will fix the dialog on ja-JP-mac.
# of cource this patch will fix the dialog on thunderbird too

Thanks crot0. ;)

Comment 22

[Axel Hecht [:Pike]]

Comment on attachment 206090 [details] [diff] [review]
ocspdialog.diff

This is unfortunate, but I want to stick to a single platform here.

I chatted with dveditz, he looked at the screenshots, and apart from "this is crappy UI per se" (he was opposed to this being a confirm dialog), it's not a security issue in his mind, folks should hardly have to use that part.

Comment 23

[dynamis (Tomoya ASAI)]

Attachment: Firefox1.5 Final Japanese version on Linux

Firefox1.5 Final Japanese version on CentOS 4.1 (Red Hat compatible Linux).
Theme of system is BlueCurve, *default* theme of CentOS (RHEL).

It's enough with some theme but not with the others including most important
default theme, BlueCurve. :(
# Sorry I didn't use BlueCurve and didn't find this problem exists on linux too.

(In reply to comment #22)
> (From update of attachment 206090 [details] [diff] [review] [edit])
> This is unfortunate, but I want to stick to a single platform here.

If you denied the patch because it includes change against ja, not only
ja-JP-mac, approve it please since ja (linux) also needs to be fixed.

> I chatted with dveditz, he looked at the screenshots, and apart from "this is
> crappy UI per se" (he was opposed to this being a confirm dialog), it's not a
> security issue in his mind, folks should hardly have to use that part.

If the patch is rejected because dveditz didn't treat this as a security issue
nor enough important issuce, I'll wait next chance.

Comment 24

[dynamis (Tomoya ASAI)]

Attachment: Firefox1.5 Final Japanese version on Linux (fixed)

40em is enough to fix the problem on linux.

I won't attach them but I checked not only BlueCurve but all the other theme
included in CentOS by default.
# Dialogs of latest nightly are same as Firefox 1.5 Final.

Comment 25

[Axel Hecht [:Pike]]

"the platform" was in this case firefox and thunderbird being released from the 
same code base, it was not a decision on mac vs something.

Comment 26

[Axel Hecht [:Pike]]

This is not my bug, probably fixed, too.

Comment 27

[dynamis (Tomoya ASAI)]

--> RESOLVED FIXED