build DRT, webkit_unit_tests from individual gyp files.

build DRT, webkit_unit_tests from individual gyp files.

This change toggles the build_webkit_exes_from_webkit_gyp file to 0,
so that we try to build DumpRenderTree and webkit_unit_tests from their
corresponding .gyp files (WebKitUnitTests.gyp and Tools.gyp) rather than
the old WebKit.gyp file. This breaks the circular dependencies between
the executables which depend on webkit_support, which depends on
WebKit.gyp. We now only use the 'webkit' target itself in WebKit.gyp.

R=tony@chromium.org
BUG=105826
TEST=waterfall stays green

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=118366

[Dirk Pranke, 2012-01-19 00:12:15]

please take a look?

[tony, 2012-01-19 00:21:46]

http://codereview.chromium.org/8889039/diff/5001/build/all.gyp
File build/all.gyp (left):

http://codereview.chromium.org/8889039/diff/5001/build/all.gyp#oldcode51
build/all.gyp:51: '../third_party/WebKit/Source/WebKit/chromium/WebKit.gyp:*',
Should we pull in all the targets in DumpRenderTree.gyp?

http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
File webkit/tools/test_shell/test_shell.gypi (right):

http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
webkit/tools/test_shell/test_shell.gypi:209: 'pull_in_copy_TestNetscapePlugIn',
It's a bit unfortunate that this is defined in a different file.  Maybe it
should be defined in test_shell.gypi instead?

[Dirk Pranke, 2012-01-19 03:16:04]

http://codereview.chromium.org/8889039/diff/5001/build/all.gyp
File build/all.gyp (left):

http://codereview.chromium.org/8889039/diff/5001/build/all.gyp#oldcode51
build/all.gyp:51: '../third_party/WebKit/Source/WebKit/chromium/WebKit.gyp:*',
On 2012/01/19 00:21:46, tony wrote:
> Should we pull in all the targets in DumpRenderTree.gyp?

My thinking here was that webkit/webkit.gyp:* should be expected to pull in
everything we need, but I suppose that that doesn't actually list ImageDiff (for
example), so maybe I'm mistaken. I'll add the dependencies back in and we can
delete later if it makes sense to do so.

http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
File webkit/tools/test_shell/test_shell.gypi (right):

http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
webkit/tools/test_shell/test_shell.gypi:209: 'pull_in_copy_TestNetscapePlugIn',
On 2012/01/19 00:21:46, tony wrote:
> It's a bit unfortunate that this is defined in a different file.  Maybe it
> should be defined in test_shell.gypi instead?

I'm not sure I see much advantage to putting it here rather than in webkit.gyp;
the target is referenced from other files as well (see chrome_tests.gypi). If it
was only used/needed by test_shell I would agree, but for now it seems better to
keep it next to the other pull_in_* targets. 

Also, if I moved that target here I'd have to have
build_webkit_exes_from_webkit_gyp logic in two places. If you still want me to
move the target here, can it wait until when I delete the conditional logic from
common.gypi and webkit.gyp, and I can move it then?

[Dirk Pranke, 2012-01-19 03:39:40]

please take another look?

[tony, 2012-01-19 18:22:05]

LGTM

http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
File webkit/tools/test_shell/test_shell.gypi (right):

http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
webkit/tools/test_shell/test_shell.gypi:209: 'pull_in_copy_TestNetscapePlugIn',
On 2012/01/19 03:16:04, Dirk Pranke wrote:
> On 2012/01/19 00:21:46, tony wrote:
> > It's a bit unfortunate that this is defined in a different file.  Maybe it
> > should be defined in test_shell.gypi instead?
> 
> I'm not sure I see much advantage to putting it here rather than in
webkit.gyp;
> the target is referenced from other files as well (see chrome_tests.gypi). If
it
> was only used/needed by test_shell I would agree, but for now it seems better
to
> keep it next to the other pull_in_* targets. 

chrome_tests.gypi is different, it explicitly names webkit.gyp.  If we wanted to
build test_shell in an upstream only checkout, this wouldn't work because we
don't include webkit.gyp in an upstream checkout (only webkit_support.gyp).

Anyway, this is fine for now, just a bit unfortunate.  We can refactor later if
we need to.

http://codereview.chromium.org/8889039/diff/6007/webkit/webkit.gyp
File webkit/webkit.gyp (right):

http://codereview.chromium.org/8889039/diff/6007/webkit/webkit.gyp#newcode41
webkit/webkit.gyp:41: }, {
Nit: Please add a comment here that this is
build_webkit_exes_from_webkit_gyp==0.

[Dirk Pranke, 2012-01-19 19:41:32]

On 2012/01/19 18:22:05, tony wrote:
> LGTM
> 
>
http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
> File webkit/tools/test_shell/test_shell.gypi (right):
> 
>
http://codereview.chromium.org/8889039/diff/5001/webkit/tools/test_shell/test...
> webkit/tools/test_shell/test_shell.gypi:209:
'pull_in_copy_TestNetscapePlugIn',
> On 2012/01/19 03:16:04, Dirk Pranke wrote:
> > On 2012/01/19 00:21:46, tony wrote:
> > > It's a bit unfortunate that this is defined in a different file.  Maybe it
> > > should be defined in test_shell.gypi instead?
> > 
> > I'm not sure I see much advantage to putting it here rather than in
> webkit.gyp;
> > the target is referenced from other files as well (see chrome_tests.gypi).
If
> it
> > was only used/needed by test_shell I would agree, but for now it seems
better
> to
> > keep it next to the other pull_in_* targets. 
> 
> chrome_tests.gypi is different, it explicitly names webkit.gyp.  If we wanted
to
> build test_shell in an upstream only checkout, this wouldn't work because we
> don't include webkit.gyp in an upstream checkout (only webkit_support.gyp).
> 

Good points. I will fix this when I yank the conditional.

> Anyway, this is fine for now, just a bit unfortunate.  We can refactor later
if
> we need to.
> 
> http://codereview.chromium.org/8889039/diff/6007/webkit/webkit.gyp
> File webkit/webkit.gyp (right):
> 
> http://codereview.chromium.org/8889039/diff/6007/webkit/webkit.gyp#newcode41
> webkit/webkit.gyp:41: }, {
> Nit: Please add a comment here that this is
> build_webkit_exes_from_webkit_gyp==0.

Will do.