RESOLVED FIXED 150097
HTMLPreloadScanner should preload iframes 
https://bugs.webkit.org/show_bug.cgi?id=150097
Summary HTMLPreloadScanner should preload iframes
Chris Dumez
Reported 2015-10-13 13:32:27 PDT
HTMLPreloadScanner should preload iframes to decrease page load time.
Attachments
Patch (6.61 KB, patch)
2015-10-13 14:52 PDT, Chris Dumez 		no flags
DetailsFormatted DiffDiff
Patch (6.88 KB, patch)
2015-10-14 11:36 PDT, Chris Dumez 		no flags
DetailsFormatted DiffDiff
Patch (9.14 KB, patch)
2015-10-15 10:09 PDT, Chris Dumez 		no flags
DetailsFormatted DiffDiff
Archive of layout-test-results from ews106 for mac-mavericks-wk2 (672.88 KB, application/zip)
2015-10-15 10:47 PDT, Build Bot 		no flags
Details
Patch (9.10 KB, patch)
2015-10-16 09:20 PDT, Chris Dumez 		no flags
DetailsFormatted DiffDiff
Show Obsolete (4) View All Add attachment proposed patch, testcase, etc.
Chris Dumez
Comment 1 2015-10-13 13:32:47 PDT
rdar://problem/23094475
Chris Dumez
Comment 2 2015-10-13 14:52:21 PDT
Created attachment 263023 [details] Patch
Chris Dumez
Comment 3 2015-10-13 14:58:03 PDT
I unfortunately did not see an impact on PLT locally.
Antti Koivisto
Comment 4 2015-10-14 00:24:55 PDT
Comment on attachment 263023 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=263023&action=review > LayoutTests/fast/preloader/frame-src.html:13 > +<html> > +<head> > +<script src="../../resources/js-test-pre.js"></script> > +<script src="http://127.0.0.1:8000/resources/slow-script.pl?delay=100"></script> > +<script> > + shouldBeTrue('internals.isPreloaded("resources/testFrame.html");'); > +</script> > +</head> > +<body> > +<p>Tests that iframes are preloaded.</p> > +<iframe src="resources/testFrame.html"></iframe> > +</body> > +</html> This test verifies that the iframe is preloaded but not that the preloaded resources is actually used (and doesn't just load again). Could you change it so that it verifies this too? This should happen even if the resource is uncacheable (as it often is).
Chris Dumez
Comment 5 2015-10-14 11:36:29 PDT
Created attachment 263091 [details] Patch
Chris Dumez
Comment 6 2015-10-14 11:37:41 PDT
Comment on attachment 263091 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=263091&action=review > LayoutTests/fast/preloader/frame-src-expected.txt:2 > +testFrame.html has MIME type text/html I now dump the responses so we see the frame is only requested once. Is this sufficient or did you have something else in mind?
Antti Koivisto
Comment 7 2015-10-15 06:48:51 PDT
Comment on attachment 263091 [details] Patch It would be also good to have the test ensure that this works when testFrame is served no-store (like it often is).
Chris Dumez
Comment 8 2015-10-15 10:09:40 PDT
Created attachment 263167 [details] Patch
Chris Dumez
Comment 9 2015-10-15 10:10:03 PDT
(In reply to comment #7) > Comment on attachment 263091 [details] > Patch > > It would be also good to have the test ensure that this works when testFrame > is served no-store (like it often is). Added such test in the patch.
Build Bot
Comment 10 2015-10-15 10:47:39 PDT
Comment on attachment 263167 [details] Patch Attachment 263167 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.webkit.org/results/292013 New failing tests: fast/preloader/frame-src.html
Build Bot
Comment 11 2015-10-15 10:47:43 PDT
Created attachment 263172 [details] Archive of layout-test-results from ews106 for mac-mavericks-wk2 The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews. Bot: ews106 Port: mac-mavericks-wk2 Platform: Mac OS X 10.9.5
Chris Dumez
Comment 12 2015-10-16 09:20:57 PDT
Created attachment 263273 [details] Patch
WebKit Commit Bot
Comment 13 2015-10-16 10:06:43 PDT
Comment on attachment 263273 [details] Patch Clearing flags on attachment: 263273 Committed r191180: <http://trac.webkit.org/changeset/191180>
WebKit Commit Bot
Comment 14 2015-10-16 10:06:50 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.