Remove the concept of user stylesheets.

Remove the concept of user stylesheets.

-The support for this has been removed from Chromium already.
This is now just deleting dead code.
-Changed TestRunner::addUserStyleSheet to TestRunner::injectStyleSheet.
It wasn't actually injecting a user style sheet and tests use it.
-Also, add an insertStyleSheet method to WebDocument so that we
can eventually get rid of the UserStyleLevel enum.
-Deleted all the tests that were explicitly testing user styles.
This also means we can delete insertAuthorCSS, since it was only
used in the user stylesheet tests.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=161671

[ojan, 2013-11-08 20:14:21]

+dglazkov for public/OWNERS.

[dglazkov, 2013-11-08 20:20:10]

On 2013/11/08 20:14:21, ojan wrote:
> +dglazkov for public/OWNERS.

Lgtm.

[vsevik, 2013-11-08 20:39:00]

Inspector still has some code to handle user stylesheets (see
https://code.google.com/p/chromium/codesearch#search/&q=stylesheetorigin&sq=p...,
stylesheet origin could be "user" and this notion spreads across DevTools
back-end, protocol and front-end.
We should remove all this dead code now but this could probably be handled in a
separate CL. apavlov@ could also have some more comments.

https://codereview.chromium.org/66383005/diff/1/LayoutTests/inspector/styles/...
File LayoutTests/inspector/styles/inject-stylesheet.html (left):

https://codereview.chromium.org/66383005/diff/1/LayoutTests/inspector/styles/...
LayoutTests/inspector/styles/inject-stylesheet.html:16:
window.testRunner.addUserStyleSheet(styleSheet, true);
Reading the test description it seems that it should be removed completely.

[commit-bot: I haz the power, 2013-11-08 22:26:49]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ojan@chromium.org/66383005/70001

[commit-bot: I haz the power, 2013-11-08 23:10:57]

Retried try job too often on linux_blink_rel for step(s) webkit_unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_blin...

[ojan, 2013-11-09 00:37:47]

On 2013/11/08 23:10:57, I haz the power (commit-bot) wrote:
> Retried try job too often on linux_blink_rel for step(s) webkit_unit_tests
>
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_blin...

I had to add back in some of this code because it turns out that watched
selectors use the userStyle RuleSet. Added it back in, renamed it to make it
clear it's only for watched selectors and added FIXMEs.

[esprehn, 2013-11-09 00:42:52]

On 2013/11/09 00:37:47, ojan wrote:
> On 2013/11/08 23:10:57, I haz the power (commit-bot) wrote:
> > Retried try job too often on linux_blink_rel for step(s) webkit_unit_tests
> >
>
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_blin...
> 
> I had to add back in some of this code because it turns out that watched
> selectors use the userStyle RuleSet. Added it back in, renamed it to make it
> clear it's only for watched selectors and added FIXMEs.

lgtm

[commit-bot: I haz the power, 2013-11-09 00:44:45]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ojan@chromium.org/66383005/260001

[commit-bot: I haz the power, 2013-11-09 01:38:54]

Change committed as 161671

[apavlov, 2013-11-09 07:51:03]

https://codereview.chromium.org/66383005/diff/1/LayoutTests/inspector/styles/...
File LayoutTests/inspector/styles/inject-stylesheet.html (left):

https://codereview.chromium.org/66383005/diff/1/LayoutTests/inspector/styles/...
LayoutTests/inspector/styles/inject-stylesheet.html:16:
window.testRunner.addUserStyleSheet(styleSheet, true);
On 2013/11/08 20:39:01, vsevik wrote:
> Reading the test description it seems that it should be removed completely.

IIRC, this tests stylesheet injection by extensions, so is still valid.

[benm (inactive), 2014-01-24 02:22:35]

Ojan - is there a bug discussing the removal of user stylesheets in
Chromium/Blink?

Thanks
Ben

[ojan, 2014-01-30 19:54:58]

Nope, no bug.


On Thu, Jan 23, 2014 at 6:22 PM, <benm@chromium.org> wrote:

> Ojan - is there a bug discussing the removal of user stylesheets in
> Chromium/Blink?
>
> Thanks
> Ben
>
> https://codereview.chromium.org/66383005/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[benm (inactive), 2014-01-31 11:03:44]

Is there any discussion about why this was removed and whether a
replacement is available or being looked into?

Thanks


On 30 January 2014 11:54, Ojan Vafai <ojan@chromium.org> wrote:

> Nope, no bug.
>
>
> On Thu, Jan 23, 2014 at 6:22 PM, <benm@chromium.org> wrote:
>
>> Ojan - is there a bug discussing the removal of user stylesheets in
>> Chromium/Blink?
>>
>> Thanks
>> Ben
>>
>> https://codereview.chromium.org/66383005/
>>
>
>

To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[abarth-chromium, 2014-01-31 16:04:06]

On 2014/01/31 11:03:44, benm wrote:
> Is there any discussion about why this was removed and whether a
> replacement is available or being looked into?

I don't think there was much discussion about it.  The description says the code
wasn't used by Chromium, which means there wasn't any reason to keep it. 
Instead of creating user style sheets, users can install extensions which
contain style sheets.

[benm (inactive), 2014-01-31 16:16:29]

OK. The alternative seems to be WebView::injectStyleSheet, however that is
specifically documented as treating the injected CSS as an author
stylesheet; I was looking for something that would be able to override
author stylesheets (without injecting another author stylesheet with
!important rules). I thought user stylesheets were a way to achieve this -
any thoughts on alternatives?

Thanks!



On 31 January 2014 16:04, <abarth@chromium.org> wrote:

> On 2014/01/31 11:03:44, benm wrote:
>
>> Is there any discussion about why this was removed and whether a
>> replacement is available or being looked into?
>>
>
> I don't think there was much discussion about it.  The description says
> the code
> wasn't used by Chromium, which means there wasn't any reason to keep it.
> Instead of creating user style sheets, users can install extensions which
> contain style sheets.
>
>
> https://codereview.chromium.org/66383005/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[ojan, 2014-01-31 19:17:51]

The point of this was to get rid of the concept of user stylesheets
entirely so we can simplify the CSS code in blink since user stylesheets
were not well supported anyways. So, author stylesheets is intentionally
the only way to do this now.


On Fri, Jan 31, 2014 at 8:16 AM, Ben Murdoch <benm@chromium.org> wrote:

> OK. The alternative seems to be WebView::injectStyleSheet, however that is
> specifically documented as treating the injected CSS as an author
> stylesheet; I was looking for something that would be able to override
> author stylesheets (without injecting another author stylesheet with
> !important rules). I thought user stylesheets were a way to achieve this -
> any thoughts on alternatives?
>
> Thanks!
>
>
>
>
> On 31 January 2014 16:04, <abarth@chromium.org> wrote:
>
>> On 2014/01/31 11:03:44, benm wrote:
>>
>>> Is there any discussion about why this was removed and whether a
>>> replacement is available or being looked into?
>>>
>>
>> I don't think there was much discussion about it.  The description says
>> the code
>> wasn't used by Chromium, which means there wasn't any reason to keep it.
>> Instead of creating user style sheets, users can install extensions which
>> contain style sheets.
>>
>>
>> https://codereview.chromium.org/66383005/
>>
>
>

To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[Prontector, 2014-02-28 23:33:15]

User style sheets were the easiest way to fix the Helvetica bug in Chrome on
Windows. Should've fixed that bug instead of removing a perfectly fine feature
for no reason. Now every website using Helvetica is f*cked up again and I have
to code myself an extension. Thanks.

[adam.benett, 2014-03-11 22:02:05]

On 2014/02/28 23:33:15, Prontector wrote:
> User style sheets were the easiest way to fix the Helvetica bug in Chrome on
> Windows. Should've fixed that bug instead of removing a perfectly fine feature
> for no reason. Now every website using Helvetica is f*cked up again and I have
> to code myself an extension. Thanks.

My project was previously using user stylesheets to hide Chrome's printing
options that is still shown briefly even when using --kiosk-printing.  Does
anyone know of an alternative solution to hide this?  Should --kiosk-printing be
changed to hide the print options when automatically using the defaults.  Now
we're stuck in a situation where 33 takes away a previously possible capability
(unless I've overlooked something).

[adam.benett, 2014-03-11 22:22:31]

On 2014/03/11 22:02:05, adam.benett wrote:
> On 2014/02/28 23:33:15, Prontector wrote:
> > User style sheets were the easiest way to fix the Helvetica bug in Chrome on
> > Windows. Should've fixed that bug instead of removing a perfectly fine
feature
> > for no reason. Now every website using Helvetica is f*cked up again and I
have
> > to code myself an extension. Thanks.
> 
> My project was previously using user stylesheets to hide Chrome's printing
> options that is still shown briefly even when using --kiosk-printing.  Does
> anyone know of an alternative solution to hide this?  Should --kiosk-printing
be
> changed to hide the print options when automatically using the defaults.  Now
> we're stuck in a situation where 33 takes away a previously possible
capability
> (unless I've overlooked something).

Nevermind.. looks like I found it up above.

[tom.krier, 2014-09-08 20:58:12]

On 2014/03/11 22:22:31, adam.benett wrote:
> On 2014/03/11 22:02:05, adam.benett wrote:
> > On 2014/02/28 23:33:15, Prontector wrote:
> > > User style sheets were the easiest way to fix the Helvetica bug in Chrome
on
> > > Windows. Should've fixed that bug instead of removing a perfectly fine
> feature
> > > for no reason. Now every website using Helvetica is f*cked up again and I
> have
> > > to code myself an extension. Thanks.
> > 
> > My project was previously using user stylesheets to hide Chrome's printing
> > options that is still shown briefly even when using --kiosk-printing.  Does
> > anyone know of an alternative solution to hide this?  Should
--kiosk-printing
> be
> > changed to hide the print options when automatically using the defaults. 
Now
> > we're stuck in a situation where 33 takes away a previously possible
> capability
> > (unless I've overlooked something).
> 
> Nevermind.. looks like I found it up above.

What did you find that helped you out with the hiding the printing option for
Chrome?
We are stuck on v32 because if we upgrade we lose the custom.css ability and are
looking for a replacement option.