List IDL compiler sources as inputs to GN's idl_impl action

List IDL compiler sources as inputs to GN's idl_impl action

This simply copies the inputs of the similar idl_compiler action, and
fixes a dependency issue where e.g. UnionTypes{Core,Modules}.* would not
be regenerated properly when the compiler sources were modified.

BUG=240176, 321462
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=192014

[Jens Widell, 2015-03-17 12:17:32]

jl@opera.com changed reviewers:
+ bashi@chromium.org, haraken@chromium.org

[Jens Widell, 2015-03-17 12:17:33]

PTAL

Spurious GN bot compilation issues observed here:
https://codereview.chromium.org/1011093002/#ps20001

[vivekg, 2015-03-17 12:58:48]

On 2015/03/17 12:17:33, Jens Widell wrote:
> PTAL
> 
> Spurious GN bot compilation issues observed here:
> https://codereview.chromium.org/1011093002/#ps20001

LGTM. Now that we are here, can we also make gn to utilize v8.gypi from custom
modules binding. Currently there exist a separate v8.gni which is not the case
for core custom binding.

[Jens Widell, 2015-03-17 13:42:08]

On 2015/03/17 12:58:48, vivekg_ wrote:
> On 2015/03/17 12:17:33, Jens Widell wrote:
> > PTAL
> > 
> > Spurious GN bot compilation issues observed here:
> > https://codereview.chromium.org/1011093002/#ps20001
> 
> LGTM.

Thanks for reviewing!

> Now that we are here, can we also make gn to utilize v8.gypi from custom
> modules binding. Currently there exist a separate v8.gni which is not the case
> for core custom binding.

Yeah, might do that in a separate CL. It's unclear to me why it is that way,
though. I'm not that familiar with GN.

https://codereview.chromium.org/402883003 added the duplication (including the
comments about it), with no hints as to why, but Brett was obviously aware of
what he was doing. Brett also wrote the GN files that imports
core/v8/custom/custom.gypi, so I'm inclined to think there's something going on
here that I don't understand.

[vivekg, 2015-03-17 14:18:58]

On 2015/03/17 at 13:42:08, jl wrote:
> On 2015/03/17 12:58:48, vivekg_ wrote:
> > On 2015/03/17 12:17:33, Jens Widell wrote:
> > > PTAL
> > > 
> > > Spurious GN bot compilation issues observed here:
> > > https://codereview.chromium.org/1011093002/#ps20001
> > 
> > LGTM.
> 
> Thanks for reviewing!
> 
> > Now that we are here, can we also make gn to utilize v8.gypi from custom
> > modules binding. Currently there exist a separate v8.gni which is not the
case
> > for core custom binding.
> 
> Yeah, might do that in a separate CL. It's unclear to me why it is that way,
though. I'm not that familiar with GN.
> 
> https://codereview.chromium.org/402883003 added the duplication (including the
comments about it), with no hints as to why, but Brett was obviously aware of
what he was doing. Brett also wrote the GN files that imports
core/v8/custom/custom.gypi, so I'm inclined to think there's something going on
here that I don't understand.

Yeah sure, makes sense. And not super urgent as well... :)

[haraken, 2015-03-17 15:03:55]

LGTM

[Jens Widell, 2015-03-17 15:04:14]

The CQ bit was checked by jl@opera.com

[commit-bot: I haz the power, 2015-03-17 15:04:43]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1014793002/1

[commit-bot: I haz the power, 2015-03-17 15:07:21]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=192014