[mojo] Add typemap and variant support to generators

[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1524613002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=569669
Committed: https://crrev.com/103fd974609482f3179a7a2c15d2d21760524fb2
Cr-Commit-Position: refs/heads/master@{#365674}

[Ken Rockot(use gerrit already), 2015-12-12 04:50:35]

Description was changed from

==========
[mojo] Generate chromium and blink variants

This adds support to mojom build rules for generating
chromium and blink bindings variants. To make the build rules
easier to write (and faster to run), the C++ generator now
*always* emits -chromium and -blink suffixed .h/.cc files
in addition to the regular ones.

All symbols in the generated variants are enclosed in an
additional nested namespace ("chromium" or "blink"),
and mojom rules generate two new sources sets. e.g. for
a mojom target named "interfaces", there are now three
source sets: interfaces, interfaces_chromium,
and interfaces_blink.

BUG=
==========

to

==========
[mojo] Generate chromium and blink variants

This adds support to mojom build rules for generating
chromium and blink bindings variants. To make the build rules
easier to write (and faster to run), the C++ generator now
*always* emits -chromium and -blink suffixed .h/.cc files
in addition to the regular ones.

All symbols in the generated variants are enclosed in an
additional nested namespace ("chromium" or "blink"),
and mojom rules generate two new sources sets. e.g. for
a mojom target named "interfaces", there are now three
source sets: interfaces, interfaces_chromium,
and interfaces_blink.

Part of a series of changes to reuse existing serialization
traits with new Mojo bindings:

  1. https://codereview.chromium.org/1524613002
  2. https://codereview.chromium.org/1515423002
  3. This CL
  4. https://codereview.chromium.org/1518293002

BUG=
==========

[Ken Rockot(use gerrit already), 2015-12-14 07:12:22]

Description was changed from

==========
[mojo] Generate chromium and blink variants

This adds support to mojom build rules for generating
chromium and blink bindings variants. To make the build rules
easier to write (and faster to run), the C++ generator now
*always* emits -chromium and -blink suffixed .h/.cc files
in addition to the regular ones.

All symbols in the generated variants are enclosed in an
additional nested namespace ("chromium" or "blink"),
and mojom rules generate two new sources sets. e.g. for
a mojom target named "interfaces", there are now three
source sets: interfaces, interfaces_chromium,
and interfaces_blink.

Part of a series of changes to reuse existing serialization
traits with new Mojo bindings:

  1. https://codereview.chromium.org/1524613002
  2. https://codereview.chromium.org/1515423002
  3. This CL
  4. https://codereview.chromium.org/1518293002

BUG=
==========

to

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1524613002
  2. https://codereview.chromium.org/1515423002
  3. https://codereview.chromium.org/1517043004
  4. https://codereview.chromium.org/1524693002
  5. https://codereview.chromium.org/1520153002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=
==========

[Ken Rockot(use gerrit already), 2015-12-14 07:12:45]

Description was changed from

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1524613002
  2. https://codereview.chromium.org/1515423002
  3. https://codereview.chromium.org/1517043004
  4. https://codereview.chromium.org/1524693002
  5. https://codereview.chromium.org/1520153002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=
==========

to

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1524613002
  2. https://codereview.chromium.org/1515423002
  3. This CL
  4. https://codereview.chromium.org/1524693002
  5. https://codereview.chromium.org/1520153002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=
==========

[Ken Rockot(use gerrit already), 2015-12-14 18:40:57]

rockot@chromium.org changed reviewers:
+ yzshen@chromium.org

[Ken Rockot(use gerrit already), 2015-12-14 18:40:57]

Please take a look!

[Ken Rockot(use gerrit already), 2015-12-14 18:45:58]

Description was changed from

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1524613002
  2. https://codereview.chromium.org/1515423002
  3. This CL
  4. https://codereview.chromium.org/1524693002
  5. https://codereview.chromium.org/1520153002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=
==========

to

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1526533002
  6. https://codereview.chromium.org/1524703002

BUG=
==========

[Ken Rockot(use gerrit already), 2015-12-14 20:18:47]

Description was changed from

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1526533002
  6. https://codereview.chromium.org/1524703002

BUG=
==========

to

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1524613002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=
==========

[Ken Rockot(use gerrit already), 2015-12-14 21:08:33]

Description was changed from

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1524613002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=
==========

to

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1524613002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=569669
==========

[yzshen1, 2015-12-15 00:09:10]

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
File mojo/public/tools/bindings/generators/cpp_templates/module.h.tmpl (right):

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
mojo/public/tools/bindings/generators/cpp_templates/module.h.tmpl:12:
format(variant_path|upper|replace("/","_")|replace(".","_")|replace("-", "_"))
%}
Please consider following the 80-char rule.

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
File mojo/public/tools/bindings/generators/mojom_java_generator.py (right):

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
mojo/public/tools/bindings/generators/mojom_java_generator.py:509: return
Please output some message or assert, so it is easier for people to find out the
reason if they don't get the generated files.

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
File mojo/public/tools/bindings/generators/mojom_js_generator.py (right):

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
mojo/public/tools/bindings/generators/mojom_js_generator.py:388: if
self.variant:
Please see my comment in mojom_java_generator.py

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
File mojo/public/tools/bindings/mojom.gni (right):

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
mojo/public/tools/bindings/mojom.gni:28: #   typemaps (optional)
What about GYP? :) I don't see the corresponding changes in GYP.

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
File mojo/public/tools/bindings/mojom_bindings_generator.py (right):

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
mojo/public/tools/bindings/mojom_bindings_generator.py:62: elif
generator_name.endswith(".py"):
This elif is probably not useful.

https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
mojo/public/tools/bindings/mojom_bindings_generator.py:95: def
LoadTypemaps(self, typemaps):
Please consider writing tests in:
https://code.google.com/p/chromium/codesearch#chromium/src/mojo/public/tools/...

(Not surprisingly, these tests are not run on bots yet. But no doubt that should
be done. :) crbug.com/547566)

https://codereview.chromium.org/1517043004/diff/120001/mojo/public/tools/bind...
File mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl
(right):

https://codereview.chromium.org/1517043004/diff/120001/mojo/public/tools/bind...
mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl:12:
format(variant_path|upper|replace("/","_")|replace(".","_")|replace("-", "_"))
%}
nit: I understand that we don't have style guide for this language. But I
personally prefer to follow the 80-char rule to make it a little more readable
when possible.

[Ken Rockot(use gerrit already), 2015-12-15 01:52:35]

On 2015/12/15 at 00:09:10, yzshen wrote:
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> File mojo/public/tools/bindings/generators/cpp_templates/module.h.tmpl
(right):
> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> mojo/public/tools/bindings/generators/cpp_templates/module.h.tmpl:12:
format(variant_path|upper|replace("/","_")|replace(".","_")|replace("-", "_"))
%}
> Please consider following the 80-char rule.

I generally tried to do this in most places, but missed some spots. Fixed the
header guards.

> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> File mojo/public/tools/bindings/generators/mojom_java_generator.py (right):
> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> mojo/public/tools/bindings/generators/mojom_java_generator.py:509: return
> Please output some message or assert, so it is easier for people to find out
the reason if they don't get the generated files.
> 

Alright, JS and Java generators now raise exceptions if you try to run them for
variants. Build rules have changed to
explicitly request only C++ bindings from the generator script when variant is
set.

>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> File mojo/public/tools/bindings/generators/mojom_js_generator.py (right):
> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> mojo/public/tools/bindings/generators/mojom_js_generator.py:388: if
self.variant:
> Please see my comment in mojom_java_generator.py
> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> File mojo/public/tools/bindings/mojom.gni (right):
> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> mojo/public/tools/bindings/mojom.gni:28: #   typemaps (optional)
> What about GYP? :) I don't see the corresponding changes in GYP.

:((((((

I forgot about GYP. What a fun adventure that was! Fixed, and also added some
missing bindings tests to the target.

> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> File mojo/public/tools/bindings/mojom_bindings_generator.py (right):
> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> mojo/public/tools/bindings/mojom_bindings_generator.py:62: elif
generator_name.endswith(".py"):
> This elif is probably not useful.

Agreed and removed.

> 
>
https://codereview.chromium.org/1517043004/diff/100001/mojo/public/tools/bind...
> mojo/public/tools/bindings/mojom_bindings_generator.py:95: def
LoadTypemaps(self, typemaps):
> Please consider writing tests in:
>
https://code.google.com/p/chromium/codesearch#chromium/src/mojo/public/tools/...

Hmm. I agree we should have tests but I think I'll defer to a future CL if you
don't mind. After taking a brief look I think there's still a bit of work needed
just to get a basic test suite up and running. WDYT?

> 
> (Not surprisingly, these tests are not run on bots yet. But no doubt that
should be done. :) crbug.com/547566)
> 
>
https://codereview.chromium.org/1517043004/diff/120001/mojo/public/tools/bind...
> File
mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl
(right):
> 
>
https://codereview.chromium.org/1517043004/diff/120001/mojo/public/tools/bind...
> mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl:12:
format(variant_path|upper|replace("/","_")|replace(".","_")|replace("-", "_"))
%}
> nit: I understand that we don't have style guide for this language. But I
personally prefer to follow the 80-char rule to make it a little more readable
when possible.

[yzshen1, 2015-12-15 23:02:53]

> Hmm. I agree we should have tests but I think I'll defer to a future CL if you
> don't mind. After taking a brief look I think there's still a bit of work
needed
> just to get a basic test suite up and running. WDYT?

Sure.

LGTM

https://codereview.chromium.org/1517043004/diff/140001/mojo/public/tools/bind...
File mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl
(right):

https://codereview.chromium.org/1517043004/diff/140001/mojo/public/tools/bind...
mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl:13:
replace("/","_")|replace(".","_")|replace("-", "_")) %}
nit: the indent is a little unusual.

(And also in module.h.tmpl.)

[Ken Rockot(use gerrit already), 2015-12-16 00:49:06]

On 2015/12/15 at 23:02:53, yzshen wrote:
> > Hmm. I agree we should have tests but I think I'll defer to a future CL if
you
> > don't mind. After taking a brief look I think there's still a bit of work
needed
> > just to get a basic test suite up and running. WDYT?
> 
> Sure.
> 
> LGTM
> 
>
https://codereview.chromium.org/1517043004/diff/140001/mojo/public/tools/bind...
> File
mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl
(right):
> 
>
https://codereview.chromium.org/1517043004/diff/140001/mojo/public/tools/bind...
> mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl:13:
replace("/","_")|replace(".","_")|replace("-", "_")) %}
> nit: the indent is a little unusual.
> 
> (And also in module.h.tmpl.)

OK - I adjusted it slightly, so that "format(" stays on the first line, with its
arguments on the next line.

I'm not sure what the proper indentation would be, but my rationale for this is
that all template code seems to exist in its own indentation space. i.e.:

{%- foo %}

would be considered to be indented 0 spaces and 

{%-   foo %}

would be considered a 2-space indent. If we assume those definitions, and we
assume the normal C++/JS style of indenting 4 spaces on a continuation line, I'd
expect to read something like:

{%- really_long_foo_thing_that_takes_too_much_space(
        arguments, indented, four_spaces, like_so) -%}


WDYT?

[yzshen1, 2015-12-16 00:52:51]

On 2015/12/16 00:49:06, Ken Rockot wrote:
> On 2015/12/15 at 23:02:53, yzshen wrote:
> > > Hmm. I agree we should have tests but I think I'll defer to a future CL if
> you
> > > don't mind. After taking a brief look I think there's still a bit of work
> needed
> > > just to get a basic test suite up and running. WDYT?
> > 
> > Sure.
> > 
> > LGTM
> > 
> >
>
https://codereview.chromium.org/1517043004/diff/140001/mojo/public/tools/bind...
> > File
> mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl
> (right):
> > 
> >
>
https://codereview.chromium.org/1517043004/diff/140001/mojo/public/tools/bind...
> >
mojo/public/tools/bindings/generators/cpp_templates/module-internal.h.tmpl:13:
> replace("/","_")|replace(".","_")|replace("-", "_")) %}
> > nit: the indent is a little unusual.
> > 
> > (And also in module.h.tmpl.)
> 
> OK - I adjusted it slightly, so that "format(" stays on the first line, with
its
> arguments on the next line.
> 
> I'm not sure what the proper indentation would be, but my rationale for this
is
> that all template code seems to exist in its own indentation space. i.e.:
> 
> {%- foo %}
> 
> would be considered to be indented 0 spaces and 
> 
> {%-   foo %}
> 
> would be considered a 2-space indent. If we assume those definitions, and we
> assume the normal C++/JS style of indenting 4 spaces on a continuation line,
I'd
> expect to read something like:
> 
> {%- really_long_foo_thing_that_takes_too_much_space(
>         arguments, indented, four_spaces, like_so) -%}
> 
> 
> WDYT?

Sounds good! That is what I have been doing, too. Sorry for the bikeshedding! :)

[Ken Rockot(use gerrit already), 2015-12-16 05:49:30]

The CQ bit was checked by rockot@chromium.org

[Ken Rockot(use gerrit already), 2015-12-16 05:49:31]

The patchset sent to the CQ was uploaded after l-g-t-m from yzshen@chromium.org
Link to the patchset: https://codereview.chromium.org/1517043004/#ps200001
(title: "gyp deps")

[commit-bot: I haz the power, 2015-12-16 05:50:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1517043004/200001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1517043004/200001

[commit-bot: I haz the power, 2015-12-16 07:08:46]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-12-16 07:08:47]

Try jobs failed on following builders:
  android_clang_dbg_recipe on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/android_clang_d...)

[Ken Rockot(use gerrit already), 2015-12-16 16:30:19]

The CQ bit was checked by rockot@chromium.org

[Ken Rockot(use gerrit already), 2015-12-16 16:30:32]

The CQ bit was unchecked by rockot@chromium.org

[Ken Rockot(use gerrit already), 2015-12-16 17:18:16]

The CQ bit was checked by rockot@chromium.org

[Ken Rockot(use gerrit already), 2015-12-16 17:18:16]

The patchset sent to the CQ was uploaded after l-g-t-m from yzshen@chromium.org
Link to the patchset: https://codereview.chromium.org/1517043004/#ps220001
(title: "merge")

[Ken Rockot(use gerrit already), 2015-12-16 17:19:41]

The CQ bit was unchecked by rockot@chromium.org

[commit-bot: I haz the power, 2015-12-16 17:19:47]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1517043004/220001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1517043004/220001

[Ken Rockot(use gerrit already), 2015-12-16 21:50:40]

The CQ bit was checked by rockot@chromium.org

[Ken Rockot(use gerrit already), 2015-12-16 21:50:40]

The patchset sent to the CQ was uploaded after l-g-t-m from yzshen@chromium.org
Link to the patchset: https://codereview.chromium.org/1517043004/#ps250001
(title: "set proper upstream")

[commit-bot: I haz the power, 2015-12-16 21:51:30]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1517043004/250001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1517043004/250001

[commit-bot: I haz the power, 2015-12-16 23:46:26]

Committed patchset #14 (id:250001)

[commit-bot: I haz the power, 2015-12-16 23:47:10]

Description was changed from

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1524613002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=569669
==========

to

==========
[mojo] Add typemap and variant support to generators

This introduces two new flags for the mojom bindings generator:

  --typemap can be used to provide typemap (JSON) files
    to the bindings generator.

  --variant specifies the name of the bindings variant to emit.

Both of these new flags only affect C++ generation, and specifying
--variant at all prevents JS or Java bindings from being emitted.

Both of these new flags have GN mojom template variables associated
with them as well.

Part of a series of changes to support custom mojom serialization:

  1. https://codereview.chromium.org/1515423002
  2. This CL
  3. https://codereview.chromium.org/1524693002
  4. https://codereview.chromium.org/1520153002
  5. https://codereview.chromium.org/1524613002
  6. https://codereview.chromium.org/1526533002
  7. https://codereview.chromium.org/1524703002

BUG=569669

Committed: https://crrev.com/103fd974609482f3179a7a2c15d2d21760524fb2
Cr-Commit-Position: refs/heads/master@{#365674}
==========

[commit-bot: I haz the power, 2015-12-16 23:47:11]

Patchset 14 (id:??) landed as
https://crrev.com/103fd974609482f3179a7a2c15d2d21760524fb2
Cr-Commit-Position: refs/heads/master@{#365674}