[Webkit-unassigned] [Bug 67229] ThreadableLoaderClient should support 'didDownloadData' updates

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=67229





--- Comment #14 from Darin Fisher (:fishd, Google) <fishd at chromium.org>  2011-08-31 13:46:31 PST ---
(In reply to comment #12)
> We need a mechanism that wouldn't make clients guess what kind of callbacks
> their ResourceHandle is going to send, and what kind is not going to be sent. 

As I mention above, there is no guessing happening here.  The mechanism is
quite explicit because the client who cares about didDownloadData would have
set downloadAsFile to true.


> It's quite indirect for a client that wants a blob to know that it should 
> listen for didDownloadData instead of didReceiveData.

Maybe I made a mistake in bringing up XMLHttpRequest.responseType("Blob").
The mechanism at the ResourceHandle level is not about Blobs.  It is about
producing files.


> Also, this needs to be done in a way that doesn't expose unwanted concepts to 
> subresource loaders.

I don't understand what this means.  It would help if you could be more specific.


> Finally, I'm still don't think that ResourceHandle is the right class to know 
> about files. Its purpose is to abstract out platform network loading API, and 
> network loading APIs on major platforms don't download to files bypassing 
> memory.

How do you define "major platforms"?  Gecko's network stack and Chromium's
network stack each support this concept, and they are part of some very
popular platforms.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.