Bug 56715 – [Qt] Refactor handling of deferred loads

• Status: RESOLVED
• Resolution: FIXED
• Component: Page Loading

[Andreas Kling]

QNetworkReplyHandler is delicious spaghetti.

[Andreas Kling]

Created attachment 86271 [details]
Proposed patch

[WebKit Review Bot]

Attachment 86271 [details] did not pass style-queue:

Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/WebCore/ChangeLog', u'Source/WebCor..." exit_code: 1

Source/WebCore/platform/network/qt/QNetworkReplyHandler.h:47:  The parameter name "handle" adds no information, so it should be removed.  [readability/parameter_name] [5]
Total errors found: 1 in 4 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Kenneth Rohde Christiansen]

Comment on attachment 86271 [details]
Proposed patch

View in context: https://bugs.webkit.org/attachment.cgi?id=86271&action=review

Generally looks good. All tests are passing I suppose :-) ?

> Source/WebCore/platform/network/qt/QNetworkReplyHandler.cpp:182
> +    , m_deferStart(true)

Shouldn't it be defers?

> Source/WebCore/platform/network/qt/QNetworkReplyHandler.cpp:184
> +    , m_deferSendResponseIfNeeded(false)

So what is ifNeeded? the defer or the send? tricky how to name these.

[Kenneth Rohde Christiansen]

(In reply to comment #3)
> (From update of attachment 86271 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=86271&action=review
> 
> Generally looks good. All tests are passing I suppose :-) ?
> 
> > Source/WebCore/platform/network/qt/QNetworkReplyHandler.cpp:182
> > +    , m_deferStart(true)
> 
> Shouldn't it be defers?
> 
> > Source/WebCore/platform/network/qt/QNetworkReplyHandler.cpp:184
> > +    , m_deferSendResponseIfNeeded(false)
> 
> So what is ifNeeded? the defer or the send? tricky how to name these.

Btw, I dislike the names loadSynchro.. Also it is not really a kind/mode anymore and it is either sync or async.

[Kenneth Rohde Christiansen]

DataType { AsynchronousData, SynchronousData } ? just thinking here.

[Andreas Kling]

Created attachment 86272 [details]
Proposed patch v2

Updated patch with improved names.
Also sprinkled some assertions over the code.

[Benjamin Poulain]

Comment on attachment 86272 [details]
Proposed patch v2

That looks good to me.

I you have the courage to continue cleaning, it would be nice to make redirection a bit easier to understand. :)
Maybe the constructor should be just calling init() so that the same init() could be called from the redirection called. This recycling of the current QNetworkReplyHandler is confusing :(

[Andreas Kling]

Comment on attachment 86272 [details]
Proposed patch v2

Clearing flags on attachment: 86272

Committed r81560: <http://trac.webkit.org/changeset/81560>

[Andreas Kling]

All reviewed patches have been landed.  Closing bug.