[webkit-dev] Web Audio API

[Chris Rogers]

Ok, then it looks like we'll go with this for the directory names:

WebCore/webaudio
WebCore/platform/audio

And this for the feature define:
#if ENABLE(WEB_AUDIO)

Thanks everybody,
Chris

On Tue, Aug 24, 2010 at 9:02 PM, Darin Fisher <darin at chromium.org> wrote:

> On Tue, Aug 24, 2010 at 8:55 PM, Eric Carlson <eric.carlson at apple.com>wrote:
>
>>
>> On Aug 24, 2010, at 8:39 PM, Darin Fisher wrote:
>>
>> My objection (and it's only a slight one) was about using "Web" as a
>> prefix for class names defined in WebCore.
>>
>> WebSockets is the main example of the Web prefix used in WebCore, and
>> that's probably because "sockets" by itself would be too confusing.
>>  However, I have found the use of the Web prefix in WebCore to lead to some
>> confusion by itself since WebKit layers tend to use the Web prefix for their
>> classes/interfaces.
>>
>> I realize that the WebCore:: namespace makes this issue technically moot.
>>  I'm just concerned about it being confusing to have WebCore::WebFoo and
>> WebKit API level WebFoo.
>>
>>   I don't think there is any plan to give the *class* name a "Web" prefix,
>> we are just talking about the names of the WebKit folders and the compile
>> flag.
>>
>> eric
>>
>>
>>
> Ah, OK.  Thanks for clearing that up for me.  It seemed like things were
> headed toward Web* classes given the contents of this folder:
> http://trac.webkit.org/browser/trunk/WebCore/websockets
> -Darin
>
>
>
>>
>>
>>
>> On Tue, Aug 24, 2010 at 5:29 PM, Chris Rogers <crogers at google.com> wrote:
>>
>>> Good, it looks like we're getting close.  So we've agreed to how the
>>> files should be split up, but Darin Fisher still was concerned about the
>>> 'web' prefix.
>>>
>>> Darin, was it the directory name WebCore/webaudio that you didn't like
>>> or:
>>> #if ENABLE(WEB_AUDIO)
>>>
>>> Alternatives might be:
>>>
>>> WebCore/audio   or    WebCore/audiocontext
>>> #if ENABLE(AUDIO_CONTEXT)     or      #if ENABLE(AUDIO_API)
>>>
>>> I'm assuming that WebCore/platform/audio we can all agree on...
>>>
>>> Chris
>>>
>>>
>>> On Tue, Aug 24, 2010 at 5:20 PM, Simon Fraser <simon.fraser at apple.com>wrote:
>>>
>>>> On Aug 24, 2010, at 4:47 PM, Chris Rogers wrote:
>>>>
>>>> Hi Simon, thanks for helping here.
>>>>
>>>> By the way, anybody who is interested can look at the files in:
>>>> https://svn.webkit.org/repository/webkit/branches/audio/WebCore/audio/
>>>>
>>>> There are actually a number of audio files which could be considered
>>>> re-usable, although I know of nothing at this time other than the web audio
>>>> API which would use them:
>>>>
>>>> AudioBus.cpp
>>>> Biquad.cpp
>>>> Reverb.cpp
>>>> FFTConvolver.cpp (and other FFT-related files)
>>>>
>>>> possibly also in this category are:
>>>> Cone.cpp
>>>> Distance.cpp
>>>>  MidSide.cpp
>>>> SinWave.cpp
>>>> Panner.cpp (and subclasses)
>>>> (maybe a few others I've missed)
>>>>
>>>> Basically, these are the lowest-level building blocks which the
>>>> higher-level parts (such as AudioContext, and AudioNode) use.  These
>>>> lowest-level building blocks do not (or should not) have any dependencies to
>>>> the higher-level code which implement the actual API (and have IDL files).
>>>>  They also don't have any dependencies on other parts of WebCore, although
>>>> they do use stuff in wtf.
>>>>
>>>> So are you suggesting:
>>>>
>>>> WebCore/webaudio                <--- IDL files and API implementation
>>>> WebCore/platform/audio        <--- lower-level building blocks such as
>>>> AudioBus.cpp
>>>>
>>>>
>>>> Fine by me!
>>>>
>>>>
>>>> then with the conditional
>>>>
>>>> #if WEB_AUDIO            <----- I added an underscore here
>>>>
>>>>
>>>> That would be #if ENABLE(WEB_AUDIO) in the code.
>>>>
>>>> Simon
>>>>
>>>>
>>>
>> _______________________________________________
>> webkit-dev mailing list
>> webkit-dev at lists.webkit.org
>> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-dev/attachments/20100824/37114558/attachment.html>