Clean up Noto Sans UI stuff.

Roboto has shipped to stable channel for a while and it works fine. This CL cleans up old Noto Sans UI font stuff.

This is done by the following steps:
 1) Remove the helper function that used to dynamically replace 'Noto Sans UI' with 'Roboto'.
 2) Remove disable-roboto-font-ui flag in chrome://flags page.
 3) Modify the default font of ChromeOS nativeUI and WebUI from 'Noto Sans UI' to 'Roboto'.

BUG=482042
TEST=Manually checked on device

Committed: https://crrev.com/43ed3b00314af1c5b689816c6ba12cbf461cef71
Cr-Commit-Position: refs/heads/master@{#329171}

[xdai1, 2015-05-06 17:01:00]

xdai@chromium.org changed reviewers:
+ avi@chromium.org, derat@chromium.org, jshin@chromium.org

[xdai1, 2015-05-06 17:01:01]

Could you help to review the CL please? Thanks!

[Avi (use Gerrit), 2015-05-06 17:03:38]

On 2015/05/06 17:01:01, xdai1 wrote:
> Could you help to review the CL please? Thanks!

Can you clarify who should be reviewing what?

[Daniel Erat, 2015-05-06 17:08:18]

thanks! this mostly looks fine to me, but i had one question. jungshik, do you
know the answer?

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
File ui/strings/translations/app_locale_settings_ja.xtb (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
ui/strings/translations/app_locale_settings_ja.xtb:15: <if expr="chromeos and
_google_chrome">
do you know what the reason is for additionally checking _google_chrome here?
most other translations just check the chromeos flag. i'm worried that this
means that we won't be defining IDS_UI_FONT_FAMILY_CROS on non-official builds.

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
File ui/strings/translations/app_locale_settings_zh-TW.xtb (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
ui/strings/translations/app_locale_settings_zh-TW.xtb:15: <if expr="chromeos and
_google_chrome">
same question here -- why does zh-CN just use chromeos but zh-TW uses
_google_chrome as well?

[xdai1, 2015-05-06 17:12:33]

On 2015/05/06 17:03:38, Avi wrote:
> On 2015/05/06 17:01:01, xdai1 wrote:
> > Could you help to review the CL please? Thanks!
> 
> Can you clarify who should be reviewing what?

You're the owner of all these files, so could you help to review all these
files?
jshin@ and derat@ are also supposed to review all these files since jshin@ is
the previous owner of font helper function, and derat@ are actively working on
font issues these days.

Thanks for your help!

[Daniel Erat, 2015-05-06 17:15:46]

On 2015/05/06 17:12:33, xdai1 wrote:
> On 2015/05/06 17:03:38, Avi wrote:
> > On 2015/05/06 17:01:01, xdai1 wrote:
> > > Could you help to review the CL please? Thanks!
> > 
> > Can you clarify who should be reviewing what?
> 
> You're the owner of all these files, so could you help to review all these
> files?
> jshin@ and derat@ are also supposed to review all these files since jshin@ is
> the previous owner of font helper function, and derat@ are actively working on
> font issues these days.
> 
> Thanks for your help!

(clarification: i don't actively work on fonts anymore; i just fix stuff that
bugs me.) :-)

[Avi (use Gerrit), 2015-05-06 18:16:46]

So the code is all nice. But I'm confused about the xtb files. I thought those
were generated by the translation console. Are we supposed to be hacking on them
by hand?

[xdai1, 2015-05-06 20:42:52]

On 2015/05/06 18:16:46, Avi wrote:
> So the code is all nice. But I'm confused about the xtb files. I thought those
> were generated by the translation console. Are we supposed to be hacking on
them
> by hand?

Seems there are two kinds of xtb files here: app_locale_settings_*.xtb and
ui_strings_*.xtb. The former one are configuration files and should be
maintained by hand and the latter one should be generated by the translation
console. Am I right, Jungshik?

[Avi (use Gerrit), 2015-05-06 22:00:03]

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
File ui/strings/app_locale_settings.grd (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
ui/strings/app_locale_settings.grd:259: Roboto, Noto Sans UI, 13px
Given the environs, where Roboto will always be available, do we still want to
list Noto Sans?

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
ui/strings/app_locale_settings.grd:264: Roboto, 'Noto Sans UI', sans-serif
ditto, plus the same question for all the other xtb files.

[Daniel Erat, 2015-05-06 22:04:10]

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
File ui/strings/app_locale_settings.grd (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
ui/strings/app_locale_settings.grd:259: Roboto, Noto Sans UI, 13px
On 2015/05/06 22:00:02, Avi wrote:
> Given the environs, where Roboto will always be available, do we still want to
> list Noto Sans?

jungshik can confirm, but i believe that noto sans ui's coverage vastly exceeds
roboto's. if that's correct, we probably want to use it for fallback.

[Avi (use Gerrit), 2015-05-06 22:06:19]

lgtm

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
File ui/strings/app_locale_settings.grd (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
ui/strings/app_locale_settings.grd:259: Roboto, Noto Sans UI, 13px
On 2015/05/06 22:04:10, Daniel Erat wrote:
> On 2015/05/06 22:00:02, Avi wrote:
> > Given the environs, where Roboto will always be available, do we still want
to
> > list Noto Sans?
> 
> jungshik can confirm, but i believe that noto sans ui's coverage vastly
exceeds
> roboto's. if that's correct, we probably want to use it for fallback.

But then, in all those other files, you're changing the size too. If for some of
those regions you're going to definitely hit Noto Sans because of the lack of
code point coverage, this then turns out to be a UI-affecting change.

[Avi (use Gerrit), 2015-05-06 22:06:20]

lgtm

[Avi (use Gerrit), 2015-05-06 22:06:42]

Gah. not lgtm yet. Darn shifting buttons.

[Daniel Erat, 2015-05-06 22:10:32]

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
File ui/strings/app_locale_settings.grd (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
ui/strings/app_locale_settings.grd:259: Roboto, Noto Sans UI, 13px
On 2015/05/06 22:06:18, Avi wrote:
> On 2015/05/06 22:04:10, Daniel Erat wrote:
> > On 2015/05/06 22:00:02, Avi wrote:
> > > Given the environs, where Roboto will always be available, do we still
want
> to
> > > list Noto Sans?
> > 
> > jungshik can confirm, but i believe that noto sans ui's coverage vastly
> exceeds
> > roboto's. if that's correct, we probably want to use it for fallback.
> 
> But then, in all those other files, you're changing the size too. If for some
of
> those regions you're going to definitely hit Noto Sans because of the lack of
> code point coverage, this then turns out to be a UI-affecting change.

you're right. the currently checked-in (and shipping) code calls
ReplaceNotoSansWithRobotoIfEnabled() to replace "Noto Sans UI" with "Roboto" in
these strings, so this change should probably do the same.

[jungshik at Google, 2015-05-07 13:45:16]

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
File ui/strings/app_locale_settings.grd (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
ui/strings/app_locale_settings.grd:259: Roboto, Noto Sans UI, 13px
On 2015/05/06 22:10:32, Daniel Erat wrote:
> On 2015/05/06 22:06:18, Avi wrote:
> > On 2015/05/06 22:04:10, Daniel Erat wrote:
> > > On 2015/05/06 22:00:02, Avi wrote:
> > > > Given the environs, where Roboto will always be available, do we still
> want
> > to
> > > > list Noto Sans?
> > > 
> > > jungshik can confirm, but i believe that noto sans ui's coverage vastly
> > exceeds
> > > roboto's. if that's correct, we probably want to use it for fallback.
> > 
> > But then, in all those other files, you're changing the size too. If for
some
> of
> > those regions you're going to definitely hit Noto Sans because of the lack
of
> > code point coverage, this then turns out to be a UI-affecting change.
> 
> you're right. the currently checked-in (and shipping) code calls
> ReplaceNotoSansWithRobotoIfEnabled() to replace "Noto Sans UI" with "Roboto"
in
> these strings, so this change should probably do the same.

Actually, Roboto (relatively new version of Roboto as checked in to Chrome OS
[1] ) matches or even exceeds the character repertoire of Noto Sans UI. So, for
languages written in LGC, we don't need 'Noto Sans UI' as a fallback. 


[1] There's an update delivered very recently and we have to update once more,
but what's checked in is 'recent' enough.

[jungshik at Google, 2015-05-07 13:50:59]

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
File ui/strings/translations/app_locale_settings_am.xtb (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
ui/strings/translations/app_locale_settings_am.xtb:7: <translation
id="IDS_WEB_FONT_FAMILY">Roboto, Noto Sans UI, Noto Sans Ethiopic,
sans-serif</translation>
As mentioned in the previous comment, you can drop 'Noto Sans UI' here and
elsewhere.

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
File ui/strings/translations/app_locale_settings_ja.xtb (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
ui/strings/translations/app_locale_settings_ja.xtb:15: <if expr="chromeos and
_google_chrome">
On 2015/05/06 17:08:18, Daniel Erat wrote:
> do you know what the reason is for additionally checking _google_chrome here?
> most other translations just check the chromeos flag. i'm worried that this
> means that we won't be defining IDS_UI_FONT_FAMILY_CROS on non-official
builds.

Chromium OS does have all the fonts listed here. So, I don't see there's any 
reason to enclose it with _google_chrome? Please, remove the condition. 

Also, drop Noto Sans UI as elsewhere.

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
File ui/strings/translations/app_locale_settings_zh-TW.xtb (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
ui/strings/translations/app_locale_settings_zh-TW.xtb:15: <if expr="chromeos and
_google_chrome">
On 2015/05/06 17:08:18, Daniel Erat wrote:
> same question here -- why does zh-CN just use chromeos but zh-TW uses
> _google_chrome as well?

Again, no reason to check _google_chrome.

[xdai1, 2015-05-08 22:22:14]

Addressed the comments. Please take another look at the CL, thanks!

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
File ui/strings/app_locale_settings.grd (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/app_locale_s...
ui/strings/app_locale_settings.grd:259: Roboto, Noto Sans UI, 13px
On 2015/05/07 13:45:15, Jungshik at google wrote:
> On 2015/05/06 22:10:32, Daniel Erat wrote:
> > On 2015/05/06 22:06:18, Avi wrote:
> > > On 2015/05/06 22:04:10, Daniel Erat wrote:
> > > > On 2015/05/06 22:00:02, Avi wrote:
> > > > > Given the environs, where Roboto will always be available, do we still
> > want
> > > to
> > > > > list Noto Sans?
> > > > 
> > > > jungshik can confirm, but i believe that noto sans ui's coverage vastly
> > > exceeds
> > > > roboto's. if that's correct, we probably want to use it for fallback.
> > > 
> > > But then, in all those other files, you're changing the size too. If for
> some
> > of
> > > those regions you're going to definitely hit Noto Sans because of the lack
> of
> > > code point coverage, this then turns out to be a UI-affecting change.
> > 
> > you're right. the currently checked-in (and shipping) code calls
> > ReplaceNotoSansWithRobotoIfEnabled() to replace "Noto Sans UI" with "Roboto"
> in
> > these strings, so this change should probably do the same.
> 
> Actually, Roboto (relatively new version of Roboto as checked in to Chrome OS
> [1] ) matches or even exceeds the character repertoire of Noto Sans UI. So,
for
> languages written in LGC, we don't need 'Noto Sans UI' as a fallback. 
> 
> 
> [1] There's an update delivered very recently and we have to update once more,
> but what's checked in is 'recent' enough. 

Done.

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
File ui/strings/translations/app_locale_settings_ja.xtb (right):

https://codereview.chromium.org/1130613002/diff/20001/ui/strings/translations...
ui/strings/translations/app_locale_settings_ja.xtb:15: <if expr="chromeos and
_google_chrome">
On 2015/05/07 13:50:59, Jungshik at google wrote:
> On 2015/05/06 17:08:18, Daniel Erat wrote:
> > do you know what the reason is for additionally checking _google_chrome
here?
> > most other translations just check the chromeos flag. i'm worried that this
> > means that we won't be defining IDS_UI_FONT_FAMILY_CROS on non-official
> builds.
> 
> Chromium OS does have all the fonts listed here. So, I don't see there's any 
> reason to enclose it with _google_chrome? Please, remove the condition. 
> 
> Also, drop Noto Sans UI as elsewhere. 

Done.

[Daniel Erat, 2015-05-08 22:42:29]

lgtm

[Avi (use Gerrit), 2015-05-08 23:11:09]

My objections are addressed. LGTM

[xdai1, 2015-05-11 16:11:18]

The CQ bit was checked by xdai@chromium.org

[commit-bot: I haz the power, 2015-05-11 16:15:05]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1130613002/40001

[commit-bot: I haz the power, 2015-05-11 17:35:07]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-05-11 17:37:19]

Patchset 3 (id:??) landed as
https://crrev.com/43ed3b00314af1c5b689816c6ba12cbf461cef71
Cr-Commit-Position: refs/heads/master@{#329171}