Bug 135079 - We don't provide an extension to the temp file used for uploads
Summary: We don't provide an extension to the temp file used for uploads
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Oliver Hunt
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-07-18 17:27 PDT by Oliver Hunt
Modified: 2014-07-21 10:34 PDT (History)
1 user (show)

See Also:


Attachments
Patch (2.07 KB, patch)
2014-07-18 17:29 PDT, Oliver Hunt
ap: review-
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Oliver Hunt 2014-07-18 17:27:53 PDT
We don't provide an extension to the temp file used for uploads
Comment 1 Oliver Hunt 2014-07-18 17:29:19 PDT
Created attachment 235158 [details]
Patch
Comment 2 Alexey Proskuryakov 2014-07-19 10:33:37 PDT
Comment on attachment 235158 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=235158&action=review

> Source/WebKit2/UIProcess/WebPageProxy.cpp:3780
> +    // FIXME: The sandbox extensions should be sent with the DidChooseFilesForOpenPanel message. This
> +    // is gated on a way of passing SandboxExtension::Handles in a Vector.

We have that, SandboxExtension::HandleArray.
Comment 3 Alexey Proskuryakov 2014-07-19 10:34:23 PDT
Comment on attachment 235158 [details]
Patch

I think that we should use the HandleArray.
Comment 4 Oliver Hunt 2014-07-19 13:17:03 PDT
(In reply to comment #3)
> (From update of attachment 235158 [details])
> I think that we should use the HandleArray.

This is a duplicate of the same code path that is used for DidChooseFilesForOpenPanel, this function is simply IOS specific. The core bug here is that both messages don't pass the extensions explicitly, and that should be fixed separately
Comment 5 Alexey Proskuryakov 2014-07-19 14:35:00 PDT
Why should it be fixed separately? The fix should be straightforward, even simpler than the workaround that you are suggesting here.
Comment 6 Oliver Hunt 2014-07-20 00:09:26 PDT
(In reply to comment #5)
> Why should it be fixed separately? The fix should be straightforward, even simpler than the workaround that you are suggesting here.

This code is a duplicate of the code below it, which is already well tested.  The only difference is that this "new" code is in an iOS only code path that does not have this piece of logic.

The "simpler" implementation requires changing the idl, and multiple methods to do a new and untested thing.
Comment 7 Oliver Hunt 2014-07-21 10:09:39 PDT
Committed r171257: <http://trac.webkit.org/changeset/171257>
Comment 8 Alexey Proskuryakov 2014-07-21 10:30:54 PDT
Seriously, committed an r-'ed patch?
Comment 9 Alexey Proskuryakov 2014-07-21 10:34:14 PDT
Oliver pointed out that this was committed before r-; it's just the comment that was posted now.