Add a flag to the compiler for compiling polymer code.

Add a flag to the compiler for compiling polymer code.

Note that I did not add the checkbox compiled_resources.gyp to third_party/closure_compiler/compiled_resources.gyp because there are still some errors to dig through.

BUG=486240
Committed: https://crrev.com/0216cb3dd6369731c269304f04279914d0a39682
Cr-Commit-Position: refs/heads/master@{#329476}

[Jeremy Klein, 2015-05-11 23:55:23]

jlklein@chromium.org changed reviewers:
+ dbeam@chromium.org

[Jeremy Klein, 2015-05-11 23:55:23]

This depends on https://codereview.chromium.org/1136093004/ for the Polymer
externs.

[Dan Beam, 2015-05-12 01:09:06]

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:60: "--extra_annotation_name=group",
arguable that we should be doing:

  _POLYMER_ARGS = ["--polymer_pass"] + [
    "--extra_annotation_name=%s" % a for a in [
      "attribute", "status", "element", "homepage", "submodule", "group",
    ]
  ]

and also with --jscomp_error (maybe it'd have to be):

  _JSCOMP_ERRORS = [
    "accessControls",
    ...
  ]

  _COMMON_CLOSURE_ARGS = [
    "--accept_const_keyword",
    ...
  ] + [
    "--jscomp_error=%s" % j for j in _JSCOMP_ERROR
  ]

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:96: current_dir =
os.path.join(os.path.dirname(__file__))
can you move this to a global and re-use?

[Dan Beam, 2015-05-12 01:10:58]

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:416:
"../polymer/v0_8/components-chromium/polymer-externs/polymer.externs.js")
80 col wrap

[Dan Beam, 2015-05-12 01:12:25]

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:416:
"../polymer/v0_8/components-chromium/polymer-externs/polymer.externs.js")
i think you might also want to use os.path.join() for this rather than
hardcoding / in the path (this wont work on windoze... if we were ever to run
this script as a presubmit that'd matter)

[Jeremy Klein, 2015-05-12 03:58:23]

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:60: "--extra_annotation_name=group",
On 2015/05/12 01:09:06, Dan Beam wrote:
> arguable that we should be doing:
> 
>   _POLYMER_ARGS = ["--polymer_pass"] + [
>     "--extra_annotation_name=%s" % a for a in [
>       "attribute", "status", "element", "homepage", "submodule", "group",
>     ]
>   ]
> 
> and also with --jscomp_error (maybe it'd have to be):
> 
>   _JSCOMP_ERRORS = [
>     "accessControls",
>     ...
>   ]
> 
>   _COMMON_CLOSURE_ARGS = [
>     "--accept_const_keyword",
>     ...
>   ] + [
>     "--jscomp_error=%s" % j for j in _JSCOMP_ERROR
>   ]
>   

Done.

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:96: current_dir =
os.path.join(os.path.dirname(__file__))
On 2015/05/12 01:09:06, Dan Beam wrote:
> can you move this to a global and re-use?

Done.

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:416:
"../polymer/v0_8/components-chromium/polymer-externs/polymer.externs.js")
On 2015/05/12 01:12:25, Dan Beam wrote:
> i think you might also want to use os.path.join() for this rather than
> hardcoding / in the path (this wont work on windoze... if we were ever to run
> this script as a presubmit that'd matter)

Done.

https://codereview.chromium.org/1112403006/diff/120001/third_party/closure_co...
third_party/closure_compiler/compile.py:416:
"../polymer/v0_8/components-chromium/polymer-externs/polymer.externs.js")
On 2015/05/12 01:10:57, Dan Beam wrote:
> 80 col wrap

Done.

[Dan Beam, 2015-05-12 04:22:37]

general question either of us should've asked beforehand:
can we just turn on Polymer for everything?

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:20: CURRENT_DIR =
os.path.join(os.path.dirname(__file__))
_CURRENT_DIR

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:27: _COMMON_JSCOMP_ERROR = [
I don't think you really need _COMMON unless there's another list of errors...,
also _JSCOMP_ERRORS (<-- S)?

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:57: # Extra flags used when compiling
polymer code.
Extra @jsDocAnnotations used when compiling Polymer code.

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:71: _POLYMER_EXTRA_ANNOTATIONS
2 "annotations" in the same line (e.g.

  "--extra_annotation_name=%s" % a for a in _POLYMER_EXTRA_ANNOTATIONS

is probably enough)

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:105: polymer: Whether the Polymer pass
is being run.
  polymer: Whether the code requires Polymer support.

[Dan Beam, 2015-05-12 04:32:49]

On 2015/05/12 04:22:37, Dan Beam wrote:
> general question either of us should've asked beforehand:
> can we just turn on Polymer for everything?

doesn't seem to make a significant time difference.  let's just turn Polymer
stuff globally.  maybe you'll catch more weirdness bugs...

[Jeremy Klein, 2015-05-12 04:47:49]

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:20: CURRENT_DIR =
os.path.join(os.path.dirname(__file__))
On 2015/05/12 04:22:37, Dan Beam wrote:
> _CURRENT_DIR

Done.

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:27: _COMMON_JSCOMP_ERROR = [
On 2015/05/12 04:22:37, Dan Beam wrote:
> I don't think you really need _COMMON unless there's another list of
errors...,
> also _JSCOMP_ERRORS (<-- S)?

There are a few in _STRICT_CLOSURE_ARGS.

Worth doing the list comprehension there too?

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:57: # Extra flags used when compiling
polymer code.
On 2015/05/12 04:22:37, Dan Beam wrote:
> Extra @jsDocAnnotations used when compiling Polymer code.

Done.

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:71: _POLYMER_EXTRA_ANNOTATIONS
On 2015/05/12 04:22:37, Dan Beam wrote:
> 2 "annotations" in the same line (e.g.
> 
>   "--extra_annotation_name=%s" % a for a in _POLYMER_EXTRA_ANNOTATIONS
> 
> is probably enough)

Done.

https://codereview.chromium.org/1112403006/diff/140001/third_party/closure_co...
third_party/closure_compiler/compile.py:105: polymer: Whether the Polymer pass
is being run.
On 2015/05/12 04:22:37, Dan Beam wrote:
>   polymer: Whether the code requires Polymer support.

Removed.

[Dan Beam, 2015-05-12 04:55:55]

lgtm

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
third_party/closure_compiler/compile.py:19: 
extra \n (\n\n between file-level globals)

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
third_party/closure_compiler/compile.py:68: ]
make it prettier

] + [
  "--jscomp_error=%s" % err for err in _COMMON_JSCOMP_ERRORS
] + [
  "--extra_annotation_name=%s" % a for a in _POLYMER_EXTRA_ANNOTATIONS
]

OR

_COMMON_JSCOMP_ERRORS = [
  "--jscomp_error=%e" for e in [
    "accessControls",
    ...
  ]
]

# ... later ...

] + _COMMON_JSCOMP_ERRORS + _POLYMER_EXTRA_ANNOTATIONS

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
third_party/closure_compiler/compile.py:411: polymer_externs =
os.path.join(_CURRENT_DIR, os.pardir, 'polymer',
hack: if you use os.path.dirname(_CURRENT_DIR) instead of .join(os.pardir) you
can skip the os.path.abspath() call

[Jeremy Klein, 2015-05-12 05:46:08]

Also fixed the --no-single-file case. PTAL

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
File third_party/closure_compiler/compile.py (right):

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
third_party/closure_compiler/compile.py:19: 
On 2015/05/12 04:55:55, Dan Beam wrote:
> extra \n (\n\n between file-level globals)

Done.

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
third_party/closure_compiler/compile.py:68: ]
On 2015/05/12 04:55:55, Dan Beam wrote:
> make it prettier
> 
> ] + [
>   "--jscomp_error=%s" % err for err in _COMMON_JSCOMP_ERRORS
> ] + [
>   "--extra_annotation_name=%s" % a for a in _POLYMER_EXTRA_ANNOTATIONS
> ]
> 
> OR
> 
> _COMMON_JSCOMP_ERRORS = [
>   "--jscomp_error=%e" for e in [
>     "accessControls",
>     ...
>   ]
> ]
> 
> # ... later ...
> 
> ] + _COMMON_JSCOMP_ERRORS + _POLYMER_EXTRA_ANNOTATIONS

Done.

https://codereview.chromium.org/1112403006/diff/160001/third_party/closure_co...
third_party/closure_compiler/compile.py:411: polymer_externs =
os.path.join(_CURRENT_DIR, os.pardir, 'polymer',
On 2015/05/12 04:55:55, Dan Beam wrote:
> hack: if you use os.path.dirname(_CURRENT_DIR) instead of .join(os.pardir) you
> can skip the os.path.abspath() call

Done.

[Dan Beam, 2015-05-12 18:48:34]

beautiful. lgtm

[Jeremy Klein, 2015-05-12 18:52:37]

The CQ bit was checked by jlklein@chromium.org

[Jeremy Klein, 2015-05-12 18:52:45]

The CQ bit was unchecked by jlklein@chromium.org

[Jeremy Klein, 2015-05-12 20:27:30]

The CQ bit was checked by jlklein@chromium.org

[commit-bot: I haz the power, 2015-05-12 20:29:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1112403006/200001

[commit-bot: I haz the power, 2015-05-12 20:38:28]

Committed patchset #11 (id:200001)

[commit-bot: I haz the power, 2015-05-12 20:39:09]

Patchset 11 (id:??) landed as
https://crrev.com/0216cb3dd6369731c269304f04279914d0a39682
Cr-Commit-Position: refs/heads/master@{#329476}