Remove the weak dictionary that maps tab objects to it markers

Remove the weak dictionary that maps tab objects to it markers in tracing_timeline_data.

We add the list of tab's markers to traces as metadata and later reconstruct the maps from tab markers to renderer thread in trace_event_importer. This makes it possible to reconstruct the timeline model with the mapping between tabs & renderer threads from the raw json tracing data. 

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=275671

[ernstm, 2014-06-02 22:07:03]

Can we distinguish more clearly between "ids" and "markers" in variable names,
etc.? "markers" are the timeline events and "ids" are the strings uniquely
identifying a tab. Example: chrome_browser_backend.py:299 adds a an entry
'tabIdMarkers' which actually store the tabIds. The previous implementation
wasn't very strict about this distinction either, but I think  the introduction
of tab.id is a good opportunity to clean this up.

[nednguyen, 2014-06-02 22:23:29]

On 2014/06/02 22:07:03, ernstm wrote:
> Can we distinguish more clearly between "ids" and "markers" in variable names,
> etc.? "markers" are the timeline events and "ids" are the strings uniquely
> identifying a tab. Example: chrome_browser_backend.py:299 adds a an entry
> 'tabIdMarkers' which actually store the tabIds. The previous implementation
> wasn't very strict about this distinction either, but I think  the
introduction
> of tab.id is a good opportunity to clean this up.

I agree, however the boundary isn't quite clear sometimes. I put the name
'tabIdMarkers' there to specify that they are tab's ids which are also markers
that can be found among the timeline events. The implementation of timeline
model constructing the map from the tabs' ids to rendered threads relies on that
assumption.

[ernstm, 2014-06-02 22:36:33]

On 2014/06/02 22:23:29, nednguyen wrote:
> On 2014/06/02 22:07:03, ernstm wrote:
> > Can we distinguish more clearly between "ids" and "markers" in variable
names,
> > etc.? "markers" are the timeline events and "ids" are the strings uniquely
> > identifying a tab. Example: chrome_browser_backend.py:299 adds a an entry
> > 'tabIdMarkers' which actually store the tabIds. The previous implementation
> > wasn't very strict about this distinction either, but I think  the
> introduction
> > of tab.id is a good opportunity to clean this up.
> 
> I agree, however the boundary isn't quite clear sometimes. I put the name
> 'tabIdMarkers' there to specify that they are tab's ids which are also markers
> that can be found among the timeline events. The implementation of timeline
> model constructing the map from the tabs' ids to rendered threads relies on
that
> assumption.

I would like to use the following terminology:
The tab has an "id" (a string)
The events created by console.time are "markers". The name of these markers is
the tab's "id".

That means a "marker" is a TimelineEvent, while an "id" is a string. There is a
one-to-one mapping between the two, but it would still be nice to have
consistent variable and dictionary entry names.

[nednguyen, 2014-06-02 22:38:07]

On 2014/06/02 22:36:33, ernstm wrote:
> On 2014/06/02 22:23:29, nednguyen wrote:
> > On 2014/06/02 22:07:03, ernstm wrote:
> > > Can we distinguish more clearly between "ids" and "markers" in variable
> names,
> > > etc.? "markers" are the timeline events and "ids" are the strings uniquely
> > > identifying a tab. Example: chrome_browser_backend.py:299 adds a an entry
> > > 'tabIdMarkers' which actually store the tabIds. The previous
implementation
> > > wasn't very strict about this distinction either, but I think  the
> > introduction
> > > of tab.id is a good opportunity to clean this up.
> > 
> > I agree, however the boundary isn't quite clear sometimes. I put the name
> > 'tabIdMarkers' there to specify that they are tab's ids which are also
markers
> > that can be found among the timeline events. The implementation of timeline
> > model constructing the map from the tabs' ids to rendered threads relies on
> that
> > assumption.
> 
> I would like to use the following terminology:
> The tab has an "id" (a string)
> The events created by console.time are "markers". The name of these markers is
> the tab's "id".
> 
> That means a "marker" is a TimelineEvent, while an "id" is a string. There is
a
> one-to-one mapping between the two, but it would still be nice to have
> consistent variable and dictionary entry names.

Ah, got it. Will change these naming.

[nduca, 2014-06-03 00:05:51]

https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
File tools/telemetry/telemetry/core/backends/chrome/chrome_browser_backend.py
(left):

https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
tools/telemetry/telemetry/core/backends/chrome/chrome_browser_backend.py:30:
class ChromeBrowserBackend(browser_backend.BrowserBackend):
you have added a method to the base class of browser backend, please add it to
the base as well and make it raise NotImplemented

https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
File tools/telemetry/telemetry/core/timeline/model.py (right):

https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
tools/telemetry/telemetry/core/timeline/model.py:76: #pylint: disable=E0202
no. please mutate the dictionary rather than letting the entire dictionary be
overridden. you won't be able to prevent it from being edited after import but
thats probably fine. even better, dont expose this map at all and instead add a
AddMappingFromTabToRenderThread(...) and you dont need this public at all

[nednguyen, 2014-06-03 15:39:24]

On 2014/06/03 00:05:51, nduca wrote:
>
https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
> File tools/telemetry/telemetry/core/backends/chrome/chrome_browser_backend.py
> (left):
> 
>
https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
> tools/telemetry/telemetry/core/backends/chrome/chrome_browser_backend.py:30:
> class ChromeBrowserBackend(browser_backend.BrowserBackend):
> you have added a method to the base class of browser backend, please add it to
> the base as well and make it raise NotImplemented
I think I didn't add any new method. Maybe I am missing something here?

>
https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
> File tools/telemetry/telemetry/core/timeline/model.py (right):
> 
>
https://codereview.chromium.org/309533016/diff/1/tools/telemetry/telemetry/co...
> tools/telemetry/telemetry/core/timeline/model.py:76: #pylint: disable=E0202
> no. please mutate the dictionary rather than letting the entire dictionary be
> overridden. you won't be able to prevent it from being edited after import but
> thats probably fine. even better, dont expose this map at all and instead add
a
> AddMappingFromTabToRenderThread(...) and you dont need this public at all

Done.

[nduca, 2014-06-06 16:06:25]

you still didnt' add a debugger_url property to the base class of the backend

[nduca, 2014-06-06 16:44:07]

ugh, i was wrong, there isn't a base class for backends. can we move id to web
contents and rename the backend field from debugger_url to id? lgtm with those.

[nednguyen, 2014-06-06 18:05:08]

On 2014/06/06 16:44:07, nduca wrote:
> ugh, i was wrong, there isn't a base class for backends. can we move id to web
> contents 
Done
> and rename the backend field from debugger_url to id? lgtm with those.

I use inspector_backend.id field for tab's id. I also implement the id for
web_driver_tab_backend which NonImplemented error. I will make a subsequent
patch to create an abstract tab_backend that both inspector_backend &
web_driver_backend inherit from.

[nednguyen, 2014-06-06 23:55:10]

On 2014/06/06 18:05:08, nednguyen wrote:
> On 2014/06/06 16:44:07, nduca wrote:
> > ugh, i was wrong, there isn't a base class for backends. can we move id to
web
> > contents 
> Done
> > and rename the backend field from debugger_url to id? lgtm with those.
> 
> I use inspector_backend.id field for tab's id. I also implement the id for
> web_driver_tab_backend which NonImplemented error. I will make a subsequent
> patch to create an abstract tab_backend that both inspector_backend &
> web_driver_backend inherit from.

Partial land in https://codereview.chromium.org/322553003/

Change inspector_backend's id to debugger_id. Mark it unimplemented in
webdriver_tab_backend.

[nednguyen, 2014-06-06 23:55:15]

The CQ bit was checked by nednguyen@google.com

[commit-bot: I haz the power, 2014-06-06 23:57:36]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/nednguyen@google.com/309533016/80001

[nednguyen, 2014-06-07 02:00:53]

The CQ bit was checked by nednguyen@google.com

[commit-bot: I haz the power, 2014-06-07 02:03:13]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/nednguyen@google.com/309533016/100001

[commit-bot: I haz the power, 2014-06-07 05:23:14]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  linux_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/linux_chromium_rel/bu...)
  mac_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/mac_chromium_rel/buil...)
  win_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/win_chromium_rel/buil...)
  win_chromium_x64_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/win_chromium_x64_rel/...)

[commit-bot: I haz the power, 2014-06-07 05:42:20]

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

[commit-bot: I haz the power, 2014-06-07 05:42:20]

Try jobs failed on following builders:
  mac_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/mac_chromium_rel/buil...)

[nednguyen, 2014-06-07 05:50:19]

The CQ bit was checked by nednguyen@google.com

[commit-bot: I haz the power, 2014-06-07 05:52:54]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/nednguyen@google.com/309533016/120001

[commit-bot: I haz the power, 2014-06-07 09:29:18]

Change committed as 275671