RESOLVED FIXED 178073
Don't use intrinsic width if our container's width is zero
https://bugs.webkit.org/show_bug.cgi?id=178073
Summary Don't use intrinsic width if our container's width is zero
Tim Horton
Reported 2017-10-09 01:13:36 PDT
Steps to Reproduce: 1. Using the attached test case, resize your browser window to the minimum width. Expected: The SVG line should not be visible. Actual: The SVG line discontinuously jumps to overlap the text in the next grid column when the width reaches 0. Notes: Chrome and Firefox both behave more sensibly (though they do not agree about the correct behavior).
Attachments
testcase (604 bytes, text/html)
2017-10-09 01:13 PDT, Tim Horton
no flags
screenshot of safari (bad case) (27.34 KB, image/png)
2017-10-09 01:14 PDT, Tim Horton
no flags
Another example to reproduce the issue (458 bytes, text/html)
2017-10-09 06:58 PDT, Manuel Rego Casasnovas
no flags
Simplified test case (251 bytes, text/html)
2017-10-09 07:03 PDT, Manuel Rego Casasnovas
no flags
Patch (294.73 KB, patch)
2017-10-13 06:02 PDT, Manuel Rego Casasnovas
no flags
Archive of layout-test-results from ews107 for mac-elcapitan-wk2 (2.03 MB, application/zip)
2017-10-13 06:59 PDT, Build Bot
no flags
Archive of layout-test-results from ews115 for mac-elcapitan (2.30 MB, application/zip)
2017-10-13 07:12 PDT, Build Bot
no flags
Archive of layout-test-results from ews121 for ios-simulator-wk2 (1.26 MB, application/zip)
2017-10-13 07:35 PDT, Build Bot
no flags
Patch (667.45 KB, patch)
2017-10-13 09:21 PDT, Manuel Rego Casasnovas
no flags
Archive of layout-test-results from ews115 for mac-elcapitan (1.90 MB, application/zip)
2017-10-13 10:52 PDT, Build Bot
no flags
Patch for landing (668.22 KB, patch)
2017-10-16 04:23 PDT, Manuel Rego Casasnovas
no flags
Tim Horton
Comment 1 2017-10-09 01:13:48 PDT
Created attachment 323160 [details] testcase
Tim Horton
Comment 2 2017-10-09 01:14:23 PDT
Created attachment 323161 [details] screenshot of safari (bad case)
Manuel Rego Casasnovas
Comment 3 2017-10-09 06:58:32 PDT
Created attachment 323172 [details] Another example to reproduce the issue The problem is easily reproducible with an image too. Check the new attached example, the size of the image depends on the "1fr" column. When it's very close to 0, the image is size 100% of the column breadth, but when it goes to 0 then the image uses the intrisic size for width (100px in the example). In Chromium and Firefox the image just disppears when the column size is 0px. We need to investigate what's going on with that image, and why it's not using the 0px breadth of the column to resolve its 100% width. The differences between Firefox and Chromium in the first example might be related to the intrinsic size of the SVG or something like that.
Manuel Rego Casasnovas
Comment 4 2017-10-09 07:03:20 PDT
Created attachment 323173 [details] Simplified test case This last example is even more simple, just a grid with a column of 0px. Then the image has a 100% width and is inside a grid item on the first column. The width of the image should be 0px but it's 100px.
Manuel Rego Casasnovas
Comment 5 2017-10-13 04:52:36 PDT
Ok so the issue is not related to Grid Layout but is a generic issue for replaced elements. It has been fixed a while ago in Blink: https://codereview.chromium.org/253743008/ But it seems the patch was never imported into WebKit. The patch is quite small so I guess we can easily import it.
Manuel Rego Casasnovas
Comment 6 2017-10-13 06:02:02 PDT
Manuel Rego Casasnovas
Comment 7 2017-10-13 06:03:16 PDT
The patch is missing the Mac baselines for width100percent-image.html but I'll get them from the EWS.
Build Bot
Comment 8 2017-10-13 06:59:16 PDT
Comment on attachment 323664 [details] Patch Attachment 323664 [details] did not pass mac-wk2-ews (mac-wk2): Output: http://webkit-queues.webkit.org/results/4844499 New failing tests: fast/replaced/width100percent-image.html
Build Bot
Comment 9 2017-10-13 06:59:17 PDT
Created attachment 323665 [details] Archive of layout-test-results from ews107 for mac-elcapitan-wk2 The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews. Bot: ews107 Port: mac-elcapitan-wk2 Platform: Mac OS X 10.11.6
Build Bot
Comment 10 2017-10-13 07:11:58 PDT
Comment on attachment 323664 [details] Patch Attachment 323664 [details] did not pass mac-debug-ews (mac): Output: http://webkit-queues.webkit.org/results/4844481 New failing tests: fast/replaced/width100percent-image.html
Build Bot
Comment 11 2017-10-13 07:12:00 PDT
Created attachment 323666 [details] Archive of layout-test-results from ews115 for mac-elcapitan The attached test failures were seen while running run-webkit-tests on the mac-debug-ews. Bot: ews115 Port: mac-elcapitan Platform: Mac OS X 10.11.6
Build Bot
Comment 12 2017-10-13 07:35:12 PDT
Comment on attachment 323664 [details] Patch Attachment 323664 [details] did not pass ios-sim-ews (ios-simulator-wk2): Output: http://webkit-queues.webkit.org/results/4844561 New failing tests: fast/replaced/width100percent-image.html
Build Bot
Comment 13 2017-10-13 07:35:13 PDT
Created attachment 323669 [details] Archive of layout-test-results from ews121 for ios-simulator-wk2 The attached test failures were seen while running run-webkit-tests on the ios-sim-ews. Bot: ews121 Port: ios-simulator-wk2 Platform: Mac OS X 10.12.6
Manuel Rego Casasnovas
Comment 14 2017-10-13 09:21:14 PDT
Build Bot
Comment 15 2017-10-13 10:52:12 PDT
Comment on attachment 323680 [details] Patch Attachment 323680 [details] did not pass mac-debug-ews (mac): Output: http://webkit-queues.webkit.org/results/4846322 New failing tests: http/tests/loading/basic-credentials-sent-automatically.html
Build Bot
Comment 16 2017-10-13 10:52:14 PDT
Created attachment 323702 [details] Archive of layout-test-results from ews115 for mac-elcapitan The attached test failures were seen while running run-webkit-tests on the mac-debug-ews. Bot: ews115 Port: mac-elcapitan Platform: Mac OS X 10.11.6
Sergio Villar Senin
Comment 17 2017-10-16 03:38:02 PDT
Comment on attachment 323680 [details] Patch Makes sense. Bonus points for interoperability
Manuel Rego Casasnovas
Comment 18 2017-10-16 04:23:38 PDT
Created attachment 323889 [details] Patch for landing
WebKit Commit Bot
Comment 19 2017-10-16 05:10:54 PDT
Comment on attachment 323889 [details] Patch for landing Clearing flags on attachment: 323889 Committed r223389: <https://trac.webkit.org/changeset/223389>
WebKit Commit Bot
Comment 20 2017-10-16 05:10:56 PDT
All reviewed patches have been landed. Closing bug.
Radar WebKit Bug Importer
Comment 21 2017-10-16 05:12:58 PDT
Note You need to log in before you can comment on or make changes to this bug.