780142 - Implement Wifi*Events using code generator

Status: RESOLVED FIXED

(Core :: DOM: Device Interfaces, defect)

Description

[Olli Pettay [:smaug][[e-mail]]]

Wifi*Events are simple interfaces which one should be able to generate automatically.
But since they are compiled only #ifdef MOZ_B2G_RIL, I have no idea how to test
that the implementation actually works.
gwagner, could you perhaps take this bug?

Basically, each event interface should be in their own .idl file, and the same
file should contain the init dictionary.
Then add the events to event_impl_gen.conf.in
(#ifdef should work in that file, at least if you do something similar what is done
to MOZ_JSDEBUGGER in the Makefile.in)

Comment 1

[Chuck Lee [:chucklee, Triggered by NI, Review, and Feedback requests]]

Attachment: 0001. Rewrite WifiStatusChangeEvent

1. Rewrite WifiStatusChangeEvent in codegen style
2. UUID of nsIDOMMozWifiStatusChangeEvent is not change since the structure is not changed.

Comment 2

[Chuck Lee [:chucklee, Triggered by NI, Review, and Feedback requests]]

Attachment: 0002. Rewrite nsDOMMozWifiConnectionInfoEvent

1. Rewrite nsDOMMozWifiConnectionInfoEvent in codegen style
2. UUID of nsDOMMozWifiConnectionInfoEvent is not change since the structure is not changed.

Comment 3

[Chuck Lee [:chucklee, Triggered by NI, Review, and Feedback requests]]

Attachment: 0001. Rewrite WifiStatusChangeEvent. Final patch for inboud. r=smaug

1. Rewrite WifiStatusChangeEvent in codegen style
2. UUID of nsIDOMMozWifiStatusChangeEvent is not change since the structure is not changed.
3. Final patch for inbound. r=smaug

Comment 4

[Chuck Lee [:chucklee, Triggered by NI, Review, and Feedback requests]]

Attachment: 0002. Rewrite nsDOMMozWifiConnectionInfoEvent. Final patch for inboud. r=smaug

1. Rewrite nsDOMMozWifiConnectionInfoEvent in codegen style
2. UUID of nsDOMMozWifiConnectionInfoEvent is not change since the structure is not changed.
3. Final patch for inbound. r=smaug

Comment 5

[Ryan VanderMeulen [:RyanVM]]

I don't see any Try results for these patches, so I've gone ahead and pushed them. I'll land this on inbound if it builds OK.
https://tbpl.mozilla.org/?tree=Try&rev=b1de0f2d6301

Comment 6

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/76836f462a5b
https://hg.mozilla.org/integration/mozilla-inbound/rev/051bc70ec007

Comment 7

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/76836f462a5b
https://hg.mozilla.org/mozilla-central/rev/051bc70ec007

Comment 8

[Chuck Lee [:chucklee, Triggered by NI, Review, and Feedback requests]]

I received a mail warning about memory leak, but using codegen shouldn't such issue.
I can't understand why for now.

Regression  Trace Malloc Leaks increase 7.91% on MacOSX 10.7 Mozilla-Inbound
------------------------------------------------------------------------------
    Previous: avg 2778267.333 stddev 59407.885 of 30 runs up to revision 97474744374c
    New     : avg 2997970.000 stddev 5103.391 of 5 runs since revision e2e341820373
    Change  : +219702.667 (7.91% / z=3.698)
    Graph   : http://mzl.la/SfE0NW

Changeset range: http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=97474744374c&tochange=e2e341820373

Changesets:
  * http://hg.mozilla.org/integration/mozilla-inbound/rev/76836f462a5b
    : Chuck Lee <[e-mail]> - Bug 780142 - Rewrite WifiStatusChangeEvent. r=smaug
    : http://bugzilla.mozilla.org/show_bug.cgi?id=780142

  * http://hg.mozilla.org/integration/mozilla-inbound/rev/051bc70ec007
    : Chuck Lee <[e-mail]> - Bug 780142 - Rewrite MozWifiConnectionInfoEvent. r=smaug
    : http://bugzilla.mozilla.org/show_bug.cgi?id=780142

  * http://hg.mozilla.org/integration/mozilla-inbound/rev/e2e341820373
    : Benjamin Chen <[e-mail]> - Bug 797817 - Update the battery indicator led when receiving battery status change. r=cjones
    : http://bugzilla.mozilla.org/show_bug.cgi?id=797817

Bugs:
  * http://bugzilla.mozilla.org/show_bug.cgi?id=797817 - Otoro LED stays lit after disconnecting from mains following full charge
  * http://bugzilla.mozilla.org/show_bug.cgi?id=780142 - Implement Wifi*Events using code generator

Comment 9

[Olli Pettay [:smaug][[e-mail]]]

This code shouldn't even be compiled on OSX, so the leak increase must be something else.