NOT FOR LANDING: Thread the needle through all webmodules.

NOT FOR LANDING: Thread the needle through all webmodules.

BUG=

[dglazkov, 2016-12-22 05:20:51]

Wired up the agent and the fake-generated code.

[dglazkov, 2016-12-22 05:25:10]

Better comments.

[dglazkov, 2016-12-22 05:26:03]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-22 05:26:15]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-22 05:33:00]

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

[commit-bot: I haz the power, 2016-12-22 05:33:00]

Dry run: Try jobs failed on following builders:
  chromium_presubmit on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[dglazkov, 2016-12-22 16:15:16]

Needle threaded all the way.

[dglazkov, 2016-12-22 16:15:17]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-22 16:15:35]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[dglazkov, 2016-12-22 16:24:46]

dglazkov@chromium.org changed reviewers:
+ esprehn@chromium.org, haraken@chromium.org, ojan@chromium.org

[commit-bot: I haz the power, 2016-12-22 16:27:10]

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

[commit-bot: I haz the power, 2016-12-22 16:27:11]

Dry run: Try jobs failed on following builders:
  chromeos_amd64-generic_chromium_compile_only_ng on
master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_amd64-...)

[dglazkov, 2016-12-22 16:36:13]

PTAL, my pseudo-design doc :)

I figured instead of writing a long doc and have you imagine how everything
fits, I'll write a minimal patch that spans the whole problem, and work on it as
a code review!

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/element.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/element.h:20: namespace api {
Should this be just web::Element? I feel like I am getting to fancy with
namespaces here. I could even go with blink::WebElement, but the "Web" prefix
looks like a namespace :)

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/element.h:30: void SetId(AtomicString id) {
element()->setIdAttribute(id); }
Here, the situation will get funky when the type that's being passed needs to be
unwrapped from blink::Type to web::api::Type. The concept that I came up with so
far was a "BFF scheme": each class A that needs to unwrap class B is declared in
class B as a friend.

However, this means that I can't do a one-pass code generation: friendship is
now known until after A is generated, so class B will need to be fixed up with
"friend" declarations on a second pass. That seems to be sucky.

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/event_target.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/event_target.h:28: using
InlinedGlobalMarkingVisitor = blink::InlinedGlobalMarkingVisitor;
This is sort of a hack. Maybe I could patch the macros to explicitly reference
namespaces?

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/event_target.h:34: //                      
AddEventListenerOptions* options) {
The idea of generating web::AddEventListenerOptions seems daunting. Maybe we
could come up with some compromise here, where only the "main" (what does that
mean?!) types are being generated, and most of blink:: types could be reused?

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/node.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/node.h:32: // TODO(dglazkov): Will need .cc
to compile this.
This sucks. I really had hope to avoid making source files with the API, but the
cross-referencing in IDLs is pervasive.

[dglazkov, 2016-12-22 16:39:10]

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/fake_the_gen.py (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/fake_the_gen.py:2: """Fake code generator! It's
awesome!"""
The fake generator gives me the ability to hand-write the files that will be
generated and actually compile them, so that we could iterate on the design.

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/element.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/element.h:30: void SetId(AtomicString id) {
element()->setIdAttribute(id); }
On 2016/12/22 at 16:36:13, dglazkov wrote:
> unwrapped from blink::Type to web::api::Type.

Correction from web::api::Type to blink::Type (the other way around).

[dglazkov, 2016-12-22 16:44:48]

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/element.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/element.h:30: void SetId(AtomicString id) {
element()->setIdAttribute(id); }
On 2016/12/22 at 16:36:13, dglazkov wrote:
> friendship is now known

Correction: friendship is NOT known.

[dglazkov, 2016-12-22 16:46:51]

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/Sour...
File third_party/WebKit/Source/web/api/frame.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/Sour...
third_party/WebKit/Source/web/api/frame.h:15: class Frame : public
blink::GarbageCollected<Frame> {
Frame will be hand-written and act as the conduit into the world of generated
file API.

[haraken, 2016-12-23 15:16:42]

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/element.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/element.h:20: namespace api {
On 2016/12/22 16:36:13, dglazkov wrote:
> Should this be just web::Element? I feel like I am getting to fancy with
> namespaces here. I could even go with blink::WebElement, but the "Web" prefix
> looks like a namespace :)

Personally I'd prefer avoiding nested namespaces. web::Element will work.

blink::WebElement would be confusing since we're aiming at replacing existing
WebXXX with web-modules.

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/element.h:30: void SetId(AtomicString id) {
element()->setIdAttribute(id); }
On 2016/12/22 16:44:48, dglazkov wrote:
> On 2016/12/22 at 16:36:13, dglazkov wrote:
> > friendship is now known
> 
> Correction: friendship is NOT known.

Would it be an option to introduce a toBlinkType() method to all web::Type
class? (c.f., we define a toV8() method for blink::Type => v8::Type conversion
and toImpl() for v8::Type => blink::Type conversion.)

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/event_target.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/event_target.h:28: using
InlinedGlobalMarkingVisitor = blink::InlinedGlobalMarkingVisitor;
On 2016/12/22 16:36:13, dglazkov wrote:
> This is sort of a hack. Maybe I could patch the macros to explicitly reference
> namespaces?

Sounds good.

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/event_target.h:34: //                      
AddEventListenerOptions* options) {
On 2016/12/22 16:36:13, dglazkov wrote:
> The idea of generating web::AddEventListenerOptions seems daunting. Maybe we
> could come up with some compromise here, where only the "main" (what does that
> mean?!) types are being generated, and most of blink:: types could be reused?

Agreed. Trying to redefine everything in the web namespace sounds too much to
me.

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
File third_party/WebKit/fake_gen/web/api/node.h (right):

https://codereview.chromium.org/2591803002/diff/60001/third_party/WebKit/fake...
third_party/WebKit/fake_gen/web/api/node.h:32: // TODO(dglazkov): Will need .cc
to compile this.
On 2016/12/22 16:36:13, dglazkov wrote:
> This sucks. I really had hope to avoid making source files with the API, but
the
> cross-referencing in IDLs is pervasive.

Personally I'm okay with generating all methods in a cpp file. I'm assuming that
customers of web-modules won't be really performance-sensitive.

[dglazkov, 2016-12-23 20:50:07]

Fix the build?

[dglazkov, 2016-12-23 20:50:08]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-23 20:50:24]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-23 21:02:29]

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

[commit-bot: I haz the power, 2016-12-23 21:02:29]

Dry run: Try jobs failed on following builders:
  cast_shell_android on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/cast_shell_a...)

[dglazkov, 2016-12-23 22:07:10]

Cut the namespaces down.

[dglazkov, 2016-12-23 22:07:11]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-23 22:07:22]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-23 22:59:05]

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

[commit-bot: I haz the power, 2016-12-23 22:59:06]

Dry run: Try jobs failed on following builders:
  android_n5x_swarming_rel on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_n5x_...)

[dglazkov, 2016-12-24 21:17:58]

Fix all the red.

[dglazkov, 2016-12-24 21:17:58]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-24 21:18:07]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-24 22:02:59]

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

[commit-bot: I haz the power, 2016-12-24 22:03:00]

Dry run: Try jobs failed on following builders:
  android_n5x_swarming_rel on master.tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_n5x_...)

[dglazkov, 2016-12-26 02:38:50]

No more Christmas, just green.

[dglazkov, 2016-12-26 02:38:51]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-26 02:39:01]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-26 04:21:34]

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

[commit-bot: I haz the power, 2016-12-26 04:21:35]

Dry run: This issue passed the CQ dry run.

[dglazkov, 2016-12-30 18:04:47]

Added rest argument handling.

[dglazkov, 2016-12-30 18:04:48]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-30 18:04:59]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-30 19:45:50]

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

[commit-bot: I haz the power, 2016-12-30 19:45:51]

Dry run: This issue passed the CQ dry run.

[dglazkov, 2017-01-01 21:11:30]

Implemented variant-based union type alternative.

[dglazkov, 2017-01-01 21:11:33]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-01-01 21:11:44]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-01-01 21:25:53]

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

[commit-bot: I haz the power, 2017-01-01 21:25:54]

Dry run: Try jobs failed on following builders:
  cast_shell_linux on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/cast_shell_linu...)

[dglazkov, 2017-02-12 19:27:24]

Updated with latest generated code.

[dglazkov, 2017-02-12 19:27:25]

The CQ bit was checked by dglazkov@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2017-02-12 19:27:36]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2017-02-12 19:41:49]

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

[commit-bot: I haz the power, 2017-02-12 19:41:53]

Dry run: Try jobs failed on following builders:
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[esprehn, 2017-02-15 02:03:02]

This looks really cool

https://codereview.chromium.org/2591803002/diff/200001/third_party/WebKit/fak...
File third_party/WebKit/fake_gen/web/api/document.h (right):

https://codereview.chromium.org/2591803002/diff/200001/third_party/WebKit/fak...
third_party/WebKit/fake_gen/web/api/document.h:41: class SVGScriptElement :
public Node {
public Element ? seems you're missing the right super class here

[haraken, 2017-02-15 02:15:14]

Is this ready for review?

[dglazkov, 2017-02-15 03:00:15]

On 2017/02/15 at 02:15:14, haraken wrote:
> Is this ready for review?

Not at all. This is just my canary for testing out the generator code and
general approach.

[dglazkov, 2017-02-15 03:04:55]

On 2017/02/15 at 02:03:02, esprehn wrote:
> This looks really cool
> 
>
https://codereview.chromium.org/2591803002/diff/200001/third_party/WebKit/fak...
> File third_party/WebKit/fake_gen/web/api/document.h (right):
> 
>
https://codereview.chromium.org/2591803002/diff/200001/third_party/WebKit/fak...
> third_party/WebKit/fake_gen/web/api/document.h:41: class SVGScriptElement :
public Node {
> public Element ? seems you're missing the right super class here

Yeah, this is just a stub to play with wtf::Variant :D

[dglazkov, 2017-02-17 03:28:41]

AddEventListener sketched out.