Refactor dmprof: Split dmprof.py into modules.

Refactor dmprof: Split dmprof.py into modules.

No changes in behavior and the main code. Changes are:

- .*Commands are moved into subcommands/.
- Other classes and functions are moved into into lib/.
- Constants are defined in their appropriate files in lib/.
- External libraries are imported via lib/.
- "Command" is renamed to "SubCommand".

BUG=123750
NOTRY=True

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=211976

[Dai Mikurube (NOT FULLTIME), 2013-07-16 09:15:11]

Hi maruel@ and peria@,

It's the big refactoring of dmprof.py with no changes in the main code.  Only
code moving and some renames.

Could you take a look when you have time?

[peria, 2013-07-16 10:30:11]

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
File tools/deep_memory_profiler/lib/bucket.py (right):

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
tools/deep_memory_profiler/lib/bucket.py:13: BUCKET_ID = 5
please put comments what those constants mean.

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
File tools/deep_memory_profiler/lib/ordered_dict.py (right):

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
tools/deep_memory_profiler/lib/ordered_dict.py:5: # TODO(dmikurube): Remove this
once Python 2.7 is required.
Remove this *file*?

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
File tools/deep_memory_profiler/subcommands/cat.py (right):

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
tools/deep_memory_profiler/subcommands/cat.py:14: BUCKET_ID = 5
Cannot those constants be imported from Bucket like other sub commands?

[Dai Mikurube (NOT FULLTIME), 2013-07-16 10:42:39]

Thanks! Addressed the comments.

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
File tools/deep_memory_profiler/lib/bucket.py (right):

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
tools/deep_memory_profiler/lib/bucket.py:13: BUCKET_ID = 5
On 2013/07/16 10:30:11, peria wrote:
> please put comments what those constants mean.

Done.

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
File tools/deep_memory_profiler/lib/ordered_dict.py (right):

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
tools/deep_memory_profiler/lib/ordered_dict.py:5: # TODO(dmikurube): Remove this
once Python 2.7 is required.
On 2013/07/16 10:30:11, peria wrote:
> Remove this *file*?

Done.

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
File tools/deep_memory_profiler/subcommands/cat.py (right):

https://codereview.chromium.org/19346002/diff/6001/tools/deep_memory_profiler...
tools/deep_memory_profiler/subcommands/cat.py:14: BUCKET_ID = 5
On 2013/07/16 10:30:11, peria wrote:
> Cannot those constants be imported from Bucket like other sub commands?

Done.

[peria, 2013-07-16 11:50:32]

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
File tools/deep_memory_profiler/lib/bucket.py (right):

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/bucket.py:14: BUCKET_ID = 5
If all constant numbers are indexes, they should be sorted.

[M-A Ruel, 2013-07-16 13:31:32]

mostly lgtm, feel free to commit after addressing these.

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
File tools/deep_memory_profiler/lib/__init__.py (right):

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/__init__.py:5: import lib.bucket
I don't think it's a good idea here.

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
File tools/deep_memory_profiler/lib/bucket.py (right):

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/bucket.py:14: BUCKET_ID = 5
On 2013/07/16 11:50:32, peria wrote:
> If all constant numbers are indexes, they should be sorted.

In general, I use the following idiom:

VIRTUAL, COMMITTED, ALLOC_COUNT, FREE_COUNT, _, BUCKET_ID = range(6)

Usually I skip 0 so I'd use range(1, 7) to reduce the likelihood of silently
accepting bad input values.

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
File tools/deep_memory_profiler/lib/exception.py (right):

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/exception.py:1: # Copyright 2013 The Chromium
Authors. All rights reserved.
Maybe name it exceptions.py ?

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/exception.py:7: super(EmptyDumpException,
self).__init__()
Just FYI, you could go simpler with:

class EmptyDumpException(Exception):
  def __str__(self):
    return repr(self.args[0])

[Dai Mikurube (NOT FULLTIME), 2013-07-17 05:55:06]

Thank you! Addressed the comments.

What do you think, Peria-san?

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
File tools/deep_memory_profiler/lib/__init__.py (right):

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/__init__.py:5: import lib.bucket
On 2013/07/16 13:31:32, Marc-Antoine Ruel wrote:
> I don't think it's a good idea here.

Just removed them.

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
File tools/deep_memory_profiler/lib/bucket.py (right):

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/bucket.py:14: BUCKET_ID = 5
On 2013/07/16 13:31:32, Marc-Antoine Ruel wrote:
> On 2013/07/16 11:50:32, peria wrote:
> > If all constant numbers are indexes, they should be sorted.
> 
> In general, I use the following idiom:
> 
> VIRTUAL, COMMITTED, ALLOC_COUNT, FREE_COUNT, _, BUCKET_ID = range(6)
> 
> Usually I skip 0 so I'd use range(1, 7) to reduce the likelihood of silently
> accepting bad input values.

Good catch. Thanks.

The index is applied for results of str.split(), so we should start from 0.

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
File tools/deep_memory_profiler/lib/exception.py (right):

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/exception.py:1: # Copyright 2013 The Chromium
Authors. All rights reserved.
On 2013/07/16 13:31:32, Marc-Antoine Ruel wrote:
> Maybe name it exceptions.py ?

Done.

https://codereview.chromium.org/19346002/diff/12001/tools/deep_memory_profile...
tools/deep_memory_profiler/lib/exception.py:7: super(EmptyDumpException,
self).__init__()
On 2013/07/16 13:31:32, Marc-Antoine Ruel wrote:
> Just FYI, you could go simpler with:
> 
> class EmptyDumpException(Exception):
>   def __str__(self):
>     return repr(self.args[0])

Done.

[peria, 2013-07-17 06:05:04]

lgtm

[Dai Mikurube (NOT FULLTIME), 2013-07-17 06:06:32]

Thanks! Committing.

[commit-bot: I haz the power, 2013-07-17 06:08:42]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dmikurube@chromium.org/19346002/38001

[commit-bot: I haz the power, 2013-07-17 06:29:01]

Change committed as 211976