Re: [webkit-dev] LayoutTestHelper

2012-05-03 Thread Ryosuke Niwa
Could you rename all these helper methods and programs to have more
descriptive names?

None of this may not have happened if we had more semantically precise
program/method names.

On Mon, Apr 30, 2012 at 12:33 PM, Dirk Pranke dpra...@chromium.org wrote:

 On Sun, Apr 29, 2012 at 6:20 PM, Maciej Stachowiak m...@apple.com wrote:
 
  On Apr 29, 2012, at 5:49 PM, Maciej Stachowiak m...@apple.com wrote:
 
 
  Hi folks,
 
  new-run-webkit-tests seems to mess with the system color profile on
 Mac, even when not running pixel tests. Historically, I believe we did this
 only when running pixel tests. I noticed that this is because it launches
 the LayoutTestHelper tool unconditionally, and in addition to changing the
 color profile on Mac, it also changes font smoothing settings on Windows.
 Does anyone know whether this work is required when running non-pixel
 tests? If not, I'd like to change NRWT to only launch LayoutTestHelper in
 pixel mode, and perhaps also rename LayoutTestHelper to PixelTestHelper.
 
  I went ahead assumed that this tool was never necessary for non-pixel
 tests, if I'm wrong, please comment here: 
 https://bugs.webkit.org/show_bug.cgi?id=81729
 

 You're wrong :). I commented on the bug as well, but for the record
 the chromium android port,at least, requires that that method be
 called regardless of whether pixel tests are being run or not. I had
 thought at some point that the chromium layout test helper did some
 things that were needed on windows regardless of whether the pixel
 tests were being run, but my memory might be failing me.

 It was always the intent that the start_helper() hook be generic.
 However, that routine was written before we had a setup_test_run()
 method, so it's probable that we could move all of the non-pixel logic
 into that method.

 -- Dirk
 ___
 webkit-dev mailing list
 webkit-dev@lists.webkit.org
 http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev

___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


Re: [webkit-dev] LayoutTestHelper

2012-05-03 Thread Maciej Stachowiak

Better naming/documentation may help. But I think the root problem here was 
that the Chromium Android LayoutTestHelper does not exist in the WebKit 
repository, as far as I can tell. So there was no reasonable way for anyone to 
consider it when making changes.

All the LayoutTestHelpers that exist in svn.webkit.org could usefully be 
renamed to PixelTestHelper at the very least, or perhaps to something even more 
specific. Also, the mac and chromium-mac versions should likely be merged. They 
are almost identical and I accidentally fixed a bug in one and not the other.

Regards,
Maciej

On May 3, 2012, at 12:31 AM, Ryosuke Niwa rn...@webkit.org wrote:

 Could you rename all these helper methods and programs to have more 
 descriptive names?
 
 None of this may not have happened if we had more semantically precise 
 program/method names.
 
 On Mon, Apr 30, 2012 at 12:33 PM, Dirk Pranke dpra...@chromium.org wrote:
 On Sun, Apr 29, 2012 at 6:20 PM, Maciej Stachowiak m...@apple.com wrote:
 
  On Apr 29, 2012, at 5:49 PM, Maciej Stachowiak m...@apple.com wrote:
 
 
  Hi folks,
 
  new-run-webkit-tests seems to mess with the system color profile on Mac, 
  even when not running pixel tests. Historically, I believe we did this 
  only when running pixel tests. I noticed that this is because it launches 
  the LayoutTestHelper tool unconditionally, and in addition to changing the 
  color profile on Mac, it also changes font smoothing settings on Windows. 
  Does anyone know whether this work is required when running non-pixel 
  tests? If not, I'd like to change NRWT to only launch LayoutTestHelper in 
  pixel mode, and perhaps also rename LayoutTestHelper to PixelTestHelper.
 
  I went ahead assumed that this tool was never necessary for non-pixel 
  tests, if I'm wrong, please comment here: 
  https://bugs.webkit.org/show_bug.cgi?id=81729
 
 
 You're wrong :). I commented on the bug as well, but for the record
 the chromium android port,at least, requires that that method be
 called regardless of whether pixel tests are being run or not. I had
 thought at some point that the chromium layout test helper did some
 things that were needed on windows regardless of whether the pixel
 tests were being run, but my memory might be failing me.
 
 It was always the intent that the start_helper() hook be generic.
 However, that routine was written before we had a setup_test_run()
 method, so it's probable that we could move all of the non-pixel logic
 into that method.
 
 -- Dirk
 ___
 webkit-dev mailing list
 webkit-dev@lists.webkit.org
 http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev
 

___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


Re: [webkit-dev] LayoutTestHelper

2012-05-03 Thread Dirk Pranke
On Thu, May 3, 2012 at 7:35 AM, Maciej Stachowiak m...@apple.com wrote:

 Better naming/documentation may help. But I think the root problem here was
 that the Chromium Android LayoutTestHelper does not exist in the WebKit
 repository, as far as I can tell. So there was no reasonable way for anyone
 to consider it when making changes.


I'm not trying to make a big deal out of this, but the python code for
the android port is checked in :)

http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/layout_tests/port/chromium_android.py

-- Dirk
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


Re: [webkit-dev] LayoutTestHelper

2012-05-03 Thread Maciej Stachowiak

On May 3, 2012, at 8:54 AM, Dirk Pranke dpra...@chromium.org wrote:

 On Thu, May 3, 2012 at 7:35 AM, Maciej Stachowiak m...@apple.com wrote:
 
 Better naming/documentation may help. But I think the root problem here was
 that the Chromium Android LayoutTestHelper does not exist in the WebKit
 repository, as far as I can tell. So there was no reasonable way for anyone
 to consider it when making changes.
 
 
 I'm not trying to make a big deal out of this, but the python code for
 the android port is checked in :)
 
 http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/layout_tests/port/chromium_android.py

I see, I was looking for something that would start a LayoutTestHelper tool. It 
seems like Android has a different interpretation of what start_helper means. I 
would like to fix all this stuff but I am not sure how to fix the naming. Do 
you have a suggestion for a better name for the port-specific pre-test setup 
hook?

Regards,
Maciej

___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


Re: [webkit-dev] LayoutTestHelper

2012-05-03 Thread Dirk Pranke
On Thu, May 3, 2012 at 9:16 AM, Maciej Stachowiak m...@apple.com wrote:

 On May 3, 2012, at 8:54 AM, Dirk Pranke dpra...@chromium.org wrote:

 On Thu, May 3, 2012 at 7:35 AM, Maciej Stachowiak m...@apple.com wrote:

 Better naming/documentation may help. But I think the root problem here was
 that the Chromium Android LayoutTestHelper does not exist in the WebKit
 repository, as far as I can tell. So there was no reasonable way for anyone
 to consider it when making changes.


 I'm not trying to make a big deal out of this, but the python code for
 the android port is checked in :)

 http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/layout_tests/port/chromium_android.py

 I see, I was looking for something that would start a LayoutTestHelper tool. 
 It seems like Android has a different interpretation of what start_helper 
 means. I would like to fix all this stuff but I am not sure how to fix the 
 naming. Do you have a suggestion for a better name for the port-specific 
 pre-test setup hook?


As I mentioned earlier in the thread, there is now a setup_test_run()
hook as well in the Port interface (and a clean_up_test_run() hook) --
you can see examples of use in
Tools/Scripts/webkitpy/layout_tests/port/gtk.py -- so we could
probably move the android code to that and then change the name or at
least the doc string of start_helper to be clear that it's
pixel-test-specific.

-- Dirk
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


Re: [webkit-dev] LayoutTestHelper

2012-04-30 Thread Dirk Pranke
On Sun, Apr 29, 2012 at 6:20 PM, Maciej Stachowiak m...@apple.com wrote:

 On Apr 29, 2012, at 5:49 PM, Maciej Stachowiak m...@apple.com wrote:


 Hi folks,

 new-run-webkit-tests seems to mess with the system color profile on Mac, 
 even when not running pixel tests. Historically, I believe we did this only 
 when running pixel tests. I noticed that this is because it launches the 
 LayoutTestHelper tool unconditionally, and in addition to changing the color 
 profile on Mac, it also changes font smoothing settings on Windows. Does 
 anyone know whether this work is required when running non-pixel tests? If 
 not, I'd like to change NRWT to only launch LayoutTestHelper in pixel mode, 
 and perhaps also rename LayoutTestHelper to PixelTestHelper.

 I went ahead assumed that this tool was never necessary for non-pixel tests, 
 if I'm wrong, please comment here: 
 https://bugs.webkit.org/show_bug.cgi?id=81729


You're wrong :). I commented on the bug as well, but for the record
the chromium android port,at least, requires that that method be
called regardless of whether pixel tests are being run or not. I had
thought at some point that the chromium layout test helper did some
things that were needed on windows regardless of whether the pixel
tests were being run, but my memory might be failing me.

It was always the intent that the start_helper() hook be generic.
However, that routine was written before we had a setup_test_run()
method, so it's probable that we could move all of the non-pixel logic
into that method.

-- Dirk
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


[webkit-dev] LayoutTestHelper

2012-04-29 Thread Maciej Stachowiak

Hi folks,

new-run-webkit-tests seems to mess with the system color profile on Mac, even 
when not running pixel tests. Historically, I believe we did this only when 
running pixel tests. I noticed that this is because it launches the 
LayoutTestHelper tool unconditionally, and in addition to changing the color 
profile on Mac, it also changes font smoothing settings on Windows. Does anyone 
know whether this work is required when running non-pixel tests? If not, I'd 
like to change NRWT to only launch LayoutTestHelper in pixel mode, and perhaps 
also rename LayoutTestHelper to PixelTestHelper.

Regards,
Maciej

___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


Re: [webkit-dev] LayoutTestHelper

2012-04-29 Thread Maciej Stachowiak

On Apr 29, 2012, at 5:49 PM, Maciej Stachowiak m...@apple.com wrote:

 
 Hi folks,
 
 new-run-webkit-tests seems to mess with the system color profile on Mac, even 
 when not running pixel tests. Historically, I believe we did this only when 
 running pixel tests. I noticed that this is because it launches the 
 LayoutTestHelper tool unconditionally, and in addition to changing the color 
 profile on Mac, it also changes font smoothing settings on Windows. Does 
 anyone know whether this work is required when running non-pixel tests? If 
 not, I'd like to change NRWT to only launch LayoutTestHelper in pixel mode, 
 and perhaps also rename LayoutTestHelper to PixelTestHelper.

I went ahead assumed that this tool was never necessary for non-pixel tests, if 
I'm wrong, please comment here: https://bugs.webkit.org/show_bug.cgi?id=81729

Regards,
Maciej

___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev