Bug 68286 – [Chromium] We need an API to expose blob creation to Chromium

• Component: WebKit Misc.
• Status: RESOLVED FIXED

[Jay Civelli]

We need an API to expose blob creation to Chromium.

That is needed for giving access to blobs in Chrome extensions.

[Jay Civelli]

Created attachment 107728 [details]
Patch

[Jay Civelli]

Created attachment 107729 [details]
Patch

[Michael Nordman]

Drive-by-comments only...

Nice to see this doesn't require bunches of code.

  v8::Handle<v8::Value>  createBlobFromFile(...)

Since an opaque v8 handle is the return value, i wonder if the method sig should be tweeked... createBlobFromFileAsV8Handle()?

[Darin Fisher (:fishd, Google)]

Comment on attachment 107729 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=107729&action=review

> Source/WebKit/chromium/public/WebBlob.h:48
> +    static v8::Handle<v8::Value> createBlobFromFile(const WebString& path, long long size);

nit: no need to mention "Blob" in the method name.  just go with WebBlob::createFromFile.

though it may seem heavier than you require, i think it would be better to make
WebBlob be a proper wrapper of WebCore::Blob.  WebCore::Blob is reference counted,
so you can use WebPrivatePtr<T>.  follow the example of WebNode.

then, you can provide a WEBKIT_USING_V8 method name toV8Value().

[Jay Civelli]

Created attachment 108056 [details]
Patch

[Darin Fisher (:fishd, Google)]

Comment on attachment 108056 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=108056&action=review

> Source/WebKit/chromium/public/WebBlob.h:51
> +    virtual ~WebBlob() { reset(); }

nit: this can be non-virtual

> Source/WebKit/chromium/public/WebBlob.h:54
> +    WebBlob(const WebBlob& n) { assign(n); }

nit: the "n" parameter name used in WebNode.h was chosen because "n" stands for "node"
here, perhaps "b" would be a better choice.

> Source/WebKit/chromium/public/WebBlob.h:66
> +    WEBKIT_EXPORT bool equals(const WebBlob&) const;

nit: it's okay to define equals here, but you could also leave it out if you like.
we don't universally provide this for simple WebCore wrappers.

> Source/WebKit/chromium/public/WebBlob.h:69
> +

nit: only one new line here

> Source/WebKit/chromium/public/WebBlob.h:77
> +#if WEBKIT_USING_V8

nit: we normally put WEBKIT_IMPLEMENTATION stuff last in the public section

[Jay Civelli]

Created attachment 108075 [details]
Patch for landing

[WebKit Review Bot]

Comment on attachment 108075 [details]
Patch for landing

Rejecting attachment 108075 [details] from commit-queue.

Failed to run "['/mnt/git/webkit-commit-queue/Tools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '-..." exit_code: 1

Last 500 characters of output:
85e273c316e0d416a2f995b30baa2e9a5ea2dc5a
r95585 = 6e9b58622449856b0904ebe3d271c5190e568d0e
Done rebuilding .git/svn/refs/remotes/origin/master/.rev_map.268f45cc-cd09-0410-ab3c-d52691b4dbfc
First, rewinding head to replay your work on top of it...
Fast-forwarded master to refs/remotes/origin/master.
Updating chromium port dependencies using gclient...

________ running '/usr/bin/python gyp_webkit' in '/mnt/git/webkit-commit-queue/Source/WebKit/chromium'
Updating webkit projects from gyp files...

Full output: http://queues.webkit.org/results/9767473

[Jay Civelli]

Created attachment 108089 [details]
Patch for landing

[WebKit Review Bot]

Comment on attachment 108089 [details]
Patch for landing

Clearing flags on attachment: 108089

Committed r95590: <http://trac.webkit.org/changeset/95590>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.