In the patch for bug #72329 I implemented uploading of BlobData content in POST requests. We are however still missing the capability of making XHR GET request for blob URLs. Fortunately that can be handled completely inside WebCore, and we already have tests for it.
Created attachment 168219 [details] Patch
Comment on attachment 168219 [details] Patch Misnamed expectation file, and two of the tests are flaky and should be investigated.
Created attachment 168691 [details] Patch Reupload. Note two of the tests are flaky and depends on the patch in bug #99178 to become reliable.
Comment on attachment 168691 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=168691&action=review r=me with one minor question. > Source/WebCore/platform/network/qt/ResourceHandleQt.cpp:156 > + blobRegistry().loadResourceSynchronously(request, error, response, data); Hmm, shouldn't this "handle" the case if BlogRegistryImpl::shouldLoadResource returning false (i.e. the cause of loadResourceSynchronously returning false) ?
(In reply to comment #4) > (From update of attachment 168691 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=168691&action=review > > r=me with one minor question. > > > Source/WebCore/platform/network/qt/ResourceHandleQt.cpp:156 > > + blobRegistry().loadResourceSynchronously(request, error, response, data); > > Hmm, shouldn't this "handle" the case if BlogRegistryImpl::shouldLoadResource returning false (i.e. the cause of loadResourceSynchronously returning false) ? We will not be able to "handle" that any better using our platform code. In we will get worse error-codes (unknown protocol instead of GET not allowed).
(In reply to comment #5) > (In reply to comment #4) > > (From update of attachment 168691 [details] [details]) > > View in context: https://bugs.webkit.org/attachment.cgi?id=168691&action=review > > > > r=me with one minor question. > > > > > Source/WebCore/platform/network/qt/ResourceHandleQt.cpp:156 > > > + blobRegistry().loadResourceSynchronously(request, error, response, data); > > > > Hmm, shouldn't this "handle" the case if BlogRegistryImpl::shouldLoadResource returning false (i.e. the cause of loadResourceSynchronously returning false) ? > > We will not be able to "handle" that any better using our platform code. In we will get worse error-codes (unknown protocol instead of GET not allowed). Okay, we won't actually set an error code in either case, but the end result is still the same.
Comment on attachment 168691 [details] Patch Clearing flags on attachment: 168691 Committed r132053: <http://trac.webkit.org/changeset/132053>
All reviewed patches have been landed. Closing bug.