Add lots of GN targets

Add lots of GN targets:
  components/autofill
  components/metrics
  components/policy
  components/variations
  third_party/libphonenumber
  third_party/mt19937ar
  ui/gfx/ipc

These are all requried to compile chrome/browser in some form or another.

Fills out all grit_defines (also required for chrome/browser) and adds associated flags for them.

Add the ability for generated_extensions_api to not generate a bundle.

Remove mt19937ar from .gitignore (it's just checked into the tree)

R=ajwong@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=281057

[brettw, 2014-07-02 00:17:01]

=

[chromium-reviews, 2014-07-02 00:29:07]

I noticed that src/BUILD.gn has //third_part/libaddressinput, but there is
no src/third_party/libaddressinput/BUILD. gn. Does it use
libaddressinput.gyp directly?
On Jul 1, 2014 5:17 PM, <brettw@chromium.org> wrote:

> Reviewers: awong,
>
> Message:
> =
>
> Description:
> Add lots of GN targets:
>   components/autofill
>   components/metrics
>   components/policy
>   components/variations
>   third_party/libphonenumber
>   third_party/mt19937ar
>   ui/gfx/ipc
>
> These are all requried to compile chrome/browser in some form or another.
>
> Fills out all grit_defines (also required for chrome/browser) and adds
> associated flags for them.
>
> Add the ability for generated_extensions_api to not generate a bundle.
>
> Remove mt19937ar from .gitignore (it's just checked into the tree)
>
> Please review this at https://codereview.chromium.org/365793002/
>
> SVN Base: svn://svn.chromium.org/chrome/trunk/src
>
> Affected files (+1355, -91 lines):
>   M .gitignore
>   M BUILD.gn
>   M build/config/BUILD.gn
>   M build/config/features.gni
>   M build/config/linux/BUILD.gn
>   M build/json_schema_compile.gypi
>   M build/secondary/tools/grit/grit_rule.gni
>   M chrome/browser/BUILD.gn
>   A chrome/browser/performance_monitor/BUILD.gn
>   A chrome/browser/safe_browsing/BUILD.gn
>   M chrome/chrome_browser.gypi
>   M chrome/common/BUILD.gn
>   M chrome/common/extensions/api/BUILD.gn
>   M components/BUILD.gn
>   M components/autofill.gypi
>   A components/autofill/content/browser/BUILD.gn
>   A components/autofill/content/common/BUILD.gn
>   A components/autofill/content/renderer/BUILD.gn
>   A components/autofill/core/browser/BUILD.gn
>   A components/autofill/core/common/BUILD.gn
>   M components/metrics.gypi
>   M components/metrics/BUILD.gn
>   M components/policy.gypi
>   A components/policy/BUILD.gn
>   A components/policy/core/browser/BUILD.gn
>   A components/policy/core/common/BUILD.gn
>   M components/policy/policy_browser.gypi
>   M components/policy/policy_common.gypi
>   M components/policy/proto/BUILD.gn
>   M components/variations.gypi
>   A components/variations/BUILD.gn
>   A + components/variations/proto/BUILD.gn
>   M extensions/common/api/BUILD.gn
>   M extensions/generated_extensions_api.gni
>   M net/BUILD.gn
>   A third_party/libphonenumber/BUILD.gn
>   M third_party/libphonenumber/libphonenumber.gyp
>   A + third_party/mt19937ar/BUILD.gn
>   M third_party/protobuf/proto_library.gni
>   M ui/accessibility/BUILD.gn
>   A + ui/gfx/ipc/BUILD.gn
>   M ui/gfx/ipc/gfx_ipc.gyp
>
>
>

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

[brettw, 2014-07-02 00:34:25]

No, libaddressinput's BUILD.gn file was already added in the previous patch, but
I forgot to hook it up.

[please use gerrit instead, 2014-07-02 00:36:35]

On 2014/07/02 00:34:25, brettw wrote:
> No, libaddressinput's BUILD.gn file was already added in the previous patch

That's great! I love building with gn :)

[awong, 2014-07-02 01:14:40]

1 real comment, and 1 wistful thought.

https://codereview.chromium.org/365793002/diff/60001/build/secondary/tools/gr...
File build/secondary/tools/grit/grit_rule.gni (right):

https://codereview.chromium.org/365793002/diff/60001/build/secondary/tools/gr...
build/secondary/tools/grit/grit_rule.gni:116: if (enable_printing != 0) {
This logic is odd given that enable_printing sounds like a boolean.  Do you know
the domain of values for enable_printing? Can we explain why "== 1" translates
to "enable_full_printing"?

https://codereview.chromium.org/365793002/diff/60001/build/secondary/tools/gr...
build/secondary/tools/grit/grit_rule.gni:152: }
I almost want the ability to take an array of strings, check for the
corresponding boolean, and then add them into another array. Sounds too
complicated for the GN language though. :-/

[awong, 2014-07-02 01:44:21]

LGTM if comment is addressed.


On Tue, Jul 1, 2014 at 6:14 PM, <ajwong@chromium.org> wrote:

> 1 real comment, and 1 wistful thought.
>
>
> https://codereview.chromium.org/365793002/diff/60001/
> build/secondary/tools/grit/grit_rule.gni
> File build/secondary/tools/grit/grit_rule.gni (right):
>
> https://codereview.chromium.org/365793002/diff/60001/
> build/secondary/tools/grit/grit_rule.gni#newcode116
> build/secondary/tools/grit/grit_rule.gni:116: if (enable_printing != 0)
> {
> This logic is odd given that enable_printing sounds like a boolean.  Do
> you know the domain of values for enable_printing? Can we explain why
> "== 1" translates to "enable_full_printing"?
>
> https://codereview.chromium.org/365793002/diff/60001/
> build/secondary/tools/grit/grit_rule.gni#newcode152
> build/secondary/tools/grit/grit_rule.gni:152: }
> I almost want the ability to take an array of strings, check for the
> corresponding boolean, and then add them into another array. Sounds too
> complicated for the GN language though. :-/
>
> https://codereview.chromium.org/365793002/
>

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

[awong, 2014-07-02 01:45:45]

LGTM.

<rant>why did we break the e-mail LGTM flow? grrr.</rant>

[brettw, 2014-07-02 02:43:12]

https://codereview.chromium.org/365793002/diff/60001/build/secondary/tools/gr...
File build/secondary/tools/grit/grit_rule.gni (right):

https://codereview.chromium.org/365793002/diff/60001/build/secondary/tools/gr...
build/secondary/tools/grit/grit_rule.gni:116: if (enable_printing != 0) {
On 2014/07/02 01:14:40, awong wrote:
> This logic is odd given that enable_printing sounds like a boolean.  Do you
know
> the domain of values for enable_printing? Can we explain why "== 1" translates
> to "enable_full_printing"?

Yeah, I've been wanting to rename it. Currently it matches the GYP exactly.
Another case like this I renamed to foo_mode. I still plan to get to it, but I
will have to update all the files that reference it so it's not super trivial.

The reason that =1 means "full" is that it used to be 0/1 like everything else,
and then the Android people came by and hacked on a partial mode and 0 and 1
were taken, so they used 2. Yes, this is very sad.

[brettw, 2014-07-02 17:37:42]

Committed patchset #5 manually as r281057.