Clean up compute_interfaces_info.py and generate_global_constructors.py

Clean up compute_interfaces_info.py and generate_global_constructors.py

This post-split CL:
1. adds documentation and renames functions,
2. makes the call to generate_global_constructors generic
　(so we don't need to add a flag for each global scope!),
3. only passes main interfaces to generate_global_constructors;
　this means we don't need to check for partial interfaces,
　and means that in a followup I can remove [NoInterfaceObject]
　from testing IDLs and implemented interfaces, since we now just don't feed
　these into generate_global_constructors!

2 followup CLs:
* Split out EventInterfaces.in
* Remove [NoInterfaceObject] from tests and dependencies
　(no longer needed, as discussed above)

Followup to:
Split generate_global_constructors.py from compute_interfaces_info.py
https://codereview.chromium.org/173503009/

Split from:
Split generate_global_constructors.py out of compute_dependencies.py
https://codereview.chromium.org/173803006/

BUG=341748
R=haraken

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=168126

[Nils Barth (inactive), 2014-02-21 13:29:17]

Here's 3/3.
You're right, it's much easier to review the logic changes
when the renaming and splitting are in separate CLs.
Sorry about that! (^_^;;)

[haraken, 2014-02-21 13:37:06]

LGTM

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
File Source/bindings/scripts/compute_interfaces_info.py (right):

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
Source/bindings/scripts/compute_interfaces_info.py:59:
'inherited_extended_attributes': inherited extended attributes

BTW, did you support [WillBeGarbageCollected]?

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
Source/bindings/scripts/compute_interfaces_info.py:69:
'dependencies_full_paths': paths to dependencies (for merging into main)

full_paths_of_dependencies ?

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
Source/bindings/scripts/compute_interfaces_info.py:70:
'dependencies_include_paths': paths for use in C++ #include directives

include_paths_of_dependencies ?

[Nils Barth (inactive), 2014-02-21 13:51:57]

Thanks!
One naming reply.

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
File Source/bindings/scripts/compute_interfaces_info.py (right):

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
Source/bindings/scripts/compute_interfaces_info.py:59:
'inherited_extended_attributes': inherited extended attributes
On 2014/02/21 13:37:07, haraken wrote:
> BTW, did you support [WillBeGarbageCollected]?

Yup, it's just down under INHERITED_EXTENDED_ATTRIBUTES!

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
Source/bindings/scripts/compute_interfaces_info.py:69:
'dependencies_full_paths': paths to dependencies (for merging into main)
On 2014/02/21 13:37:07, haraken wrote:
> full_paths_of_dependencies ?

It feels more natural as '..._paths' than 'paths_of_...'.
I think it's an English word order issue, but it's also reflected in GYP.

Suffixing is more consistent with gyp, where '_files' is special
and means "paths", so you see lots of '..._idl_files'.

Similarly, you often say "file name, file path, full path", etc.:
usually it's (adjective/attributive noun) (noun).
Putting "of" shows it's opposite of normal order.
(Compare 'name_of_file' to 'file_name'.)

[haraken, 2014-02-21 13:53:19]

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
File Source/bindings/scripts/compute_interfaces_info.py (right):

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
Source/bindings/scripts/compute_interfaces_info.py:59:
'inherited_extended_attributes': inherited extended attributes
On 2014/02/21 13:51:57, Nils Barth wrote:
> On 2014/02/21 13:37:07, haraken wrote:
> > BTW, did you support [WillBeGarbageCollected]?
> 
> Yup, it's just down under INHERITED_EXTENDED_ATTRIBUTES!

Thanks!

https://codereview.chromium.org/174953005/diff/1/Source/bindings/scripts/comp...
Source/bindings/scripts/compute_interfaces_info.py:69:
'dependencies_full_paths': paths to dependencies (for merging into main)
On 2014/02/21 13:51:57, Nils Barth wrote:
> On 2014/02/21 13:37:07, haraken wrote:
> > full_paths_of_dependencies ?
> 
> It feels more natural as '..._paths' than 'paths_of_...'.
> I think it's an English word order issue, but it's also reflected in GYP.
> 
> Suffixing is more consistent with gyp, where '_files' is special
> and means "paths", so you see lots of '..._idl_files'.
> 
> Similarly, you often say "file name, file path, full path", etc.:
> usually it's (adjective/attributive noun) (noun).
> Putting "of" shows it's opposite of normal order.
> (Compare 'name_of_file' to 'file_name'.)

Makes sense.

[Nils Barth (inactive), 2014-02-28 09:35:00]

The CQ bit was checked by nbarth@chromium.org

[commit-bot: I haz the power, 2014-02-28 09:35:10]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nbarth@chromium.org/174953005/100001

[commit-bot: I haz the power, 2014-02-28 13:17:44]

Change committed as 168126