Bug 61079 – Add rel type prerender to distinguish prerender from prefetch

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Gavin Peters]

Add rel type prerender to distinguish prerender from prefetch

[Gavin Peters]

Created attachment 93958 [details]
Patch

[Gavin Peters]

Also, see how I cleaned things up in CachedResourceRequest.cpp?  I like the new cachedResourceTypeToTargetType a lot better.

[WebKit Review Bot]

Attachment 93958 [details] did not pass style-queue:

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

Source/WebCore/loader/cache/CachedResourceLoader.h:73:  The parameter name "priority" adds no information, so it should be removed.  [readability/parameter_name] [5]
Total errors found: 1 in 16 files


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

[Gavin Peters]

Unfortunately, this has tickled me before, the style failure doesn't recognise a default parameter as sufficient reason to name an argument.

[Gavin Peters]

Comment on attachment 93958 [details]
Patch

bad patch.  new upload pending.

[Gavin Peters]

Created attachment 93961 [details]
Patch

[Gavin Peters]

Comment on attachment 93961 [details]
Patch

Better patch.  Thank you [e-mail].

[WebKit Review Bot]

Attachment 93961 [details] did not pass style-queue:

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

Source/WebCore/loader/cache/CachedResourceLoader.h:73:  The parameter name "priority" adds no information, so it should be removed.  [readability/parameter_name] [5]
Total errors found: 1 in 16 files


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

[Early Warning System Bot]

Comment on attachment 93961 [details]
Patch

Attachment 93961 [details] did not pass qt-ews (qt):
Output: http://queues.webkit.org/results/8707547

[Gavin Peters]

Created attachment 93966 [details]
Patch

[Gavin Peters]

Comment on attachment 93966 [details]
Patch

Fix non-prefetching builds!

[WebKit Review Bot]

Attachment 93966 [details] did not pass style-queue:

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

Source/WebCore/loader/cache/CachedResourceLoader.h:73:  The parameter name "priority" adds no information, so it should be removed.  [readability/parameter_name] [5]
Total errors found: 1 in 16 files


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

[Adam Barth]

Comment on attachment 93966 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=93966&action=review

> Source/WebCore/html/HTMLLinkElement.cpp:263
>              priority = ResourceLoadPriorityLow;

We don't want to set priority = ResourceLoadPriorityLow for prerender as well?

[Gavin Peters]

Comment on attachment 93966 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=93966&action=review

>> Source/WebCore/html/HTMLLinkElement.cpp:263
>>              priority = ResourceLoadPriorityLow;
> 
> We don't want to set priority = ResourceLoadPriorityLow for prerender as well?

I think no; the VeryLow that it gets by default is probably correct.  Priority in WebKit loading only controls when requests are sent to the network layer (the lower the priority, the later they go out), and for prerender being late like a prefetch is probably correct; the prerender fetch then won't interfere as much with main resource loading.  In the subresource case, the contrast is that the resource is associated with the current page, and Low priority mirrors what something like an image would get.

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 93966 [details]:

http/tests/websocket/tests/handshake-fail-by-cross-origin.html bug 54147 (author: [e-mail])
The commit-queue is continuing to process your patch.

[WebKit Commit Bot]

Comment on attachment 93966 [details]
Patch

Clearing flags on attachment: 93966

Committed r87020: <http://trac.webkit.org/changeset/87020>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[WebKit Commit Bot]

The commit-queue encountered the following flaky tests while processing attachment 93966 [details]:

http/tests/websocket/tests/close-event.html bug 61240 (author: [e-mail])
The commit-queue is continuing to process your patch.

[Gavin Peters]

I have learned a valuable lesson; always test the chrome side of a webkit patch that has chrome implication before uploading to WebKit: https://bugs.webkit.org/show_bug.cgi?id=61297 catches two mistakes that shouldn't have been in this bug.