Re: [webkit-dev] Using Git for WebKit development

2011-01-18 Thread Philippe Normand
On Tue, 2011-01-18 at 11:26 +0300, Konstantin Tokarev wrote:
 Hi all,
 
 When I'm trying to cherry-pick or rebase something in my WebKit git clone,
 I constantly end up with conflicts in changelog files
 
 How do you fight this problem?
 

Add in your .git/config:

[merge changelog]
driver = Tools/Scripts/resolve-ChangeLogs --merge-driver %O %A %
B

Philippe


signature.asc
Description: This is a digitally signed message part
___
webkit-dev mailing list
webkit-dev@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev


Re: [webkit-dev] Using Git for WebKit development

2011-01-18 Thread Jeremy Orlow
The info is on https://trac.webkit.org/wiki/UsingGitWithWebKit as well.

On Tue, Jan 18, 2011 at 8:42 AM, Philippe Normand pnorm...@igalia.comwrote:

 On Tue, 2011-01-18 at 11:26 +0300, Konstantin Tokarev wrote:
  Hi all,
 
  When I'm trying to cherry-pick or rebase something in my WebKit git
 clone,
  I constantly end up with conflicts in changelog files
 
  How do you fight this problem?
 

 Add in your .git/config:

 [merge changelog]
driver = Tools/Scripts/resolve-ChangeLogs --merge-driver %O %A %
 B

 Philippe

 ___
 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] Using Git for WebKit development

2011-01-18 Thread Konstantin Tokarev


18.01.2011, 12:52, Jeremy Orlow jor...@chromium.org:
 The info is on https://trac.webkit.org/wiki/UsingGitWithWebKit as well.

 On Tue, Jan 18, 2011 at 8:42 AM, Philippe Normand pnorm...@igalia.com wrote:
 On Tue, 2011-01-18 at 11:26 +0300, Konstantin Tokarev wrote:
 Hi all,

 When I'm trying to cherry-pick or rebase something in my WebKit git clone,
 I constantly end up with conflicts in changelog files

 How do you fight this problem?


 Add in your .git/config:

 [merge changelog]
        driver = Tools/Scripts/resolve-ChangeLogs --merge-driver %O %A %
 B


Thanks for all replies!


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


[webkit-dev] question re: including some python-licensed code into the tree

2011-01-18 Thread Dirk Pranke
Hi all,

In the course of working on new-run-webkit-tests, I find myself
needing to implement a variant of some code normally provided in the
Python standard library. Attempting to implement this in a clean room
manner will be painful and nonobvious, so I'd just as soon just
cutpaste the relevant code over and modify it. However, doing so
would require me to include the appropriate license info, and I'm told
that the PSF license may not necessarily be already approved for
inclusion into the tree.

Anyone have any thoughts on this? The code in question is the
implementation of the os.walk() routine (about 10 lines of code),
which needs to be emulated using an in-memory implementation of a
filesystem that uses a dictionary of files. the PSF license is here:
http://docs.python.org/license.html . It is GPL-compatible but more
liberal.

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


[webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread Darin Adler
It seems that a lot of people are making script tests with non-standard 
wrappers. And not adding exceptions to the make-script-test-wrappers script. I 
ran the script and it created 4 files, and modified 33 others.

Any ideas on how to improve the situation?

-- Darin

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


Re: [webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread James Robinson
On Tue, Jan 18, 2011 at 4:30 PM, Darin Adler da...@apple.com wrote:

 It seems that a lot of people are making script tests with non-standard
 wrappers. And not adding exceptions to the make-script-test-wrappers
 script. I ran the script and it created 4 files, and modified 33 others.

 Any ideas on how to improve the situation?


I didn't even know such a script existed.  What are the advantages of using
it vs generating script tests in other ways?  I see that the use is
documented here:
https://trac.webkit.org/wiki/Writing%20Layout%20Tests%20for%20DumpRenderTreebut
there aren't any motivations listed for it.

- James


 -- Darin


 ___
 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] make-script-test-wrappers not being maintained

2011-01-18 Thread Ryosuke Niwa
On Tue, Jan 18, 2011 at 4:30 PM, Darin Adler da...@apple.com wrote:

 It seems that a lot of people are making script tests with non-standard
 wrappers. And not adding exceptions to the make-script-test-wrappers
 script. I ran the script and it created 4 files, and modified 33 others.

 Any ideas on how to improve the situation?


What kind of diffs you're getting for the wrappers?

One test I had to use a non-standard wrapper was when the test included some
RTL text and I had to specify text encoding type (utf-8) by a meta tag.

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


Re: [webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread Simon Fraser
On Jan 18, 2011, at 4:30 PM, Darin Adler wrote:

 It seems that a lot of people are making script tests with non-standard 
 wrappers. And not adding exceptions to the make-script-test-wrappers script. 
 I ran the script and it created 4 files, and modified 33 others.

I'm probably guilty of some of those. I had totally forgotten about 
make-script-test-wrappers.
 
 Any ideas on how to improve the situation?

I usually take an existing .html file for a script test and modify it. It would 
be good if the generated HTML had a big !-- THIS IS AUTOGENERATED, DO NOT COPY 
OR EDIT -- in it.

Simon

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


[webkit-dev] Leopard Bot has been broken for 5 days.

2011-01-18 Thread Eric Seidel
I ran webkit-patch failure-reason on Leopard Intel Release (Tests) and got:

SUCCESS: Build 26596 (r75728) was the first to show failures:
set([u'fast/blockflow/broken-ideograph-small-caps.html'])
Suspect revisions:
r75726:
  http://trac.webkit.org/changeset/75726
  Bug: 52364 (https://bugs.webkit.org/show_bug.cgi?id=52364)
  Author: Csaba Osztrogonác o...@webkit.org
  Reviewer: Darin Adler da...@apple.com
  Committer: Csaba Osztrogonác o...@webkit.org
r75727:
  http://trac.webkit.org/changeset/75727
  Bug: None (None)
  Author: Tony Chang t...@chromium.org
  Reviewer: None
  Committer: Tony Chang t...@chromium.org
r75728:
  http://trac.webkit.org/changeset/75728
  Bug: 52317 (https://bugs.webkit.org/show_bug.cgi?id=52317)
  Author: Dimitri Glazkov dglaz...@chromium.org
  Reviewer: None
  Committer: Dimitri Glazkov dglaz...@chromium.org
Explained all results for Leopard Intel Release (Tests)

But those don't actually look related to the failure.

I suspect the bot has a configuration issue.

Could someone who has access to the bot give it a kick?

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


Re: [webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread Ojan Vafai
On Tue, Jan 18, 2011 at 4:46 PM, Simon Fraser simon.fra...@apple.comwrote:

 On Jan 18, 2011, at 4:30 PM, Darin Adler wrote:
  Any ideas on how to improve the situation?

 I usually take an existing .html file for a script test and modify it. It
 would be good if the generated HTML had a big !-- THIS IS AUTOGENERATED, DO
 NOT COPY OR EDIT -- in it.


I agree this would be valuable.

If you need a non-standard wrapper, don't put the JS file in the
script-tests directory. If you need shared JS resources, they should go in
the resources directory, not in the script-tests directory. We should not
need any exclusions AFAIK and should remove the concept of exclusions from
the generator script entirely.

Also, there's been a number of discussions about this on webkit-dev. There
seemed to be general consensus then that we should eliminate or simplify
this generator script. All we need to do is fix js-test-pre.js to use onload
events to eliminate the need for js-test-post.js. Then we don't need the
generator script and can just write these tests by hand.

Pure script test (possibly generated by a script):

!DocType html

script src=path/to/js-test-pre.js/script

script src=path/to/test.js/script

Regular test (hand rolled):

!DocType html

script src=path/to/js-test-pre.js/script
divhtml that demonstrates bug/div
script
// Script that verifies correctness or otherwise excersizes bug.
/script

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


Re: [webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread Darin Adler
On Jan 18, 2011, at 4:36 PM, James Robinson wrote:

 What are the advantages of using it vs generating script tests in other ways?

The concept is that you write a script test and the wrapper is automatically 
generated. The advantage is that you don’t have to write the wrapper yourself.

The long term plan was that we might come up with a completely different way of 
making wrappers. For many JavaScript-only tests, we might come up with a way to 
run the script tests directly without involving WebCore at all. Even for the 
other tests we might stop checking in wrappers.

The wrappers have the benefit that they let you easily run the test in any web 
browser. For run-webkit-tests, though, they probably aren’t needed.

-- Darin

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


Re: [webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread Darin Adler
On Jan 18, 2011, at 4:49 PM, Ojan Vafai wrote:

 On Tue, Jan 18, 2011 at 4:46 PM, Simon Fraser simon.fra...@apple.com wrote:
 It would be good if the generated HTML had a big !-- THIS IS AUTOGENERATED, 
 DO NOT COPY OR EDIT -- in it.
 
 I agree this would be valuable.

We should do it.

 If you need a non-standard wrapper, don't put the JS file in the script-tests 
 directory.

That’s a good idea. We should have done that when we created the script-tests 
directory. The exclusions list predates that directory.

 If you need shared JS resources, they should go in the resources directory, 
 not in the script-tests directory.

That’s already done, I think.

 We should not need any exclusions AFAIK and should remove the concept of 
 exclusions from the generator script entirely.

Yes.

 Also, there's been a number of discussions about this on webkit-dev.

I didn’t remember that.

 There seemed to be general consensus then that we should eliminate or 
 simplify this generator script.

Interesting.

 All we need to do is fix js-test-pre.js to use onload events to eliminate the 
 need for js-test-post.js.

Maybe that will work for almost all the tests.

-- Darin

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


Re: [webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread Alexey Proskuryakov

18.01.2011, в 16:30, Darin Adler написал(а):

 It seems that a lot of people are making script tests with non-standard 
 wrappers. And not adding exceptions to the make-script-test-wrappers script. 
 I ran the script and it created 4 files, and modified 33 others.
 
 Any ideas on how to improve the situation?

I think that a style bot rule complaining about new files in script-tests 
directories (outside fast/js) would be great. If there are no new script tests, 
then there are no incorrectly prepared ones.

- WBR, Alexey Proskuryakov

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


Re: [webkit-dev] question re: including some python-licensed code into the tree

2011-01-18 Thread Adam Barth
In the past, we've used autoinstall to make use of library code that
doesn't have a compatible license, assuming that the library is freely
distributed on the Internet.  However, that approach does not seem
appropriate for this particular use.

Adam


On Tue, Jan 18, 2011 at 3:59 PM, Dirk Pranke dpra...@chromium.org wrote:
 Hi all,

 In the course of working on new-run-webkit-tests, I find myself
 needing to implement a variant of some code normally provided in the
 Python standard library. Attempting to implement this in a clean room
 manner will be painful and nonobvious, so I'd just as soon just
 cutpaste the relevant code over and modify it. However, doing so
 would require me to include the appropriate license info, and I'm told
 that the PSF license may not necessarily be already approved for
 inclusion into the tree.

 Anyone have any thoughts on this? The code in question is the
 implementation of the os.walk() routine (about 10 lines of code),
 which needs to be emulated using an in-memory implementation of a
 filesystem that uses a dictionary of files. the PSF license is here:
 http://docs.python.org/license.html . It is GPL-compatible but more
 liberal.

 -- 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] Leopard Bot has been broken for 5 days.

2011-01-18 Thread Adam Barth
I investigated this issue for a while.  Disabling the test just causes
the next test to fail.  I'm not very familiar with this code.  I can
spend more time investigating, but having someone familiar with
ideographs would likely be more efficient.

Adam


On Tue, Jan 18, 2011 at 4:49 PM, Eric Seidel e...@webkit.org wrote:
 I ran webkit-patch failure-reason on Leopard Intel Release (Tests) and 
 got:

 SUCCESS: Build 26596 (r75728) was the first to show failures:
 set([u'fast/blockflow/broken-ideograph-small-caps.html'])
 Suspect revisions:
 r75726:
  http://trac.webkit.org/changeset/75726
  Bug: 52364 (https://bugs.webkit.org/show_bug.cgi?id=52364)
  Author: Csaba Osztrogonác o...@webkit.org
  Reviewer: Darin Adler da...@apple.com
  Committer: Csaba Osztrogonác o...@webkit.org
 r75727:
  http://trac.webkit.org/changeset/75727
  Bug: None (None)
  Author: Tony Chang t...@chromium.org
  Reviewer: None
  Committer: Tony Chang t...@chromium.org
 r75728:
  http://trac.webkit.org/changeset/75728
  Bug: 52317 (https://bugs.webkit.org/show_bug.cgi?id=52317)
  Author: Dimitri Glazkov dglaz...@chromium.org
  Reviewer: None
  Committer: Dimitri Glazkov dglaz...@chromium.org
 Explained all results for Leopard Intel Release (Tests)

 But those don't actually look related to the failure.

 I suspect the bot has a configuration issue.

 Could someone who has access to the bot give it a kick?

 -eric
 ___
 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] Leopard Bot has been broken for 5 days.

2011-01-18 Thread Mihai Parparita
If this is the same as http://webkit.org/b/51807, then it's actually
been broken for a couple of weeks.

Mihai

On Tue, Jan 18, 2011 at 10:12 PM, Adam Barth aba...@webkit.org wrote:
 I investigated this issue for a while.  Disabling the test just causes
 the next test to fail.  I'm not very familiar with this code.  I can
 spend more time investigating, but having someone familiar with
 ideographs would likely be more efficient.

 Adam


 On Tue, Jan 18, 2011 at 4:49 PM, Eric Seidel e...@webkit.org wrote:
 I ran webkit-patch failure-reason on Leopard Intel Release (Tests) and 
 got:

 SUCCESS: Build 26596 (r75728) was the first to show failures:
 set([u'fast/blockflow/broken-ideograph-small-caps.html'])
 Suspect revisions:
 r75726:
  http://trac.webkit.org/changeset/75726
  Bug: 52364 (https://bugs.webkit.org/show_bug.cgi?id=52364)
  Author: Csaba Osztrogonác o...@webkit.org
  Reviewer: Darin Adler da...@apple.com
  Committer: Csaba Osztrogonác o...@webkit.org
 r75727:
  http://trac.webkit.org/changeset/75727
  Bug: None (None)
  Author: Tony Chang t...@chromium.org
  Reviewer: None
  Committer: Tony Chang t...@chromium.org
 r75728:
  http://trac.webkit.org/changeset/75728
  Bug: 52317 (https://bugs.webkit.org/show_bug.cgi?id=52317)
  Author: Dimitri Glazkov dglaz...@chromium.org
  Reviewer: None
  Committer: Dimitri Glazkov dglaz...@chromium.org
 Explained all results for Leopard Intel Release (Tests)

 But those don't actually look related to the failure.

 I suspect the bot has a configuration issue.

 Could someone who has access to the bot give it a kick?

 -eric
 ___
 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

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


Re: [webkit-dev] make-script-test-wrappers not being maintained

2011-01-18 Thread Maciej Stachowiak

On Jan 18, 2011, at 4:50 PM, Darin Adler wrote:

 On Jan 18, 2011, at 4:43 PM, Ryosuke Niwa wrote:
 
 What kind of diffs you're getting for the wrappers?
 
 All kinds. Most are some small difference between the template and the 
 generated wrapper. An HTML5 DOCTYPE instead of the one in the wrapper. 

Off-topic, but... we really should change the script to use the HTML5 doctype.

 - Maciej

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