CSS border-image-slice longhand not overridden by default value in border-image shorthand
Reported by
andersg...@gmail.com,
Dec 29
|
|||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/71.0.3578.98 Safari/537.36 Steps to reproduce the problem: 1. Visit https://jsfiddle.net/pt8cdm7k/ (or see the attached HTML test page) 2. See the result 3. Inspect the rendered result using the Chrome Developer Tools What is the expected behavior? Test case: The browser should render a <button> element surrounded by four dots, one dot at each corner. Details: A `border-image` shorthand without explicitly declaring a `border-image-slice` value should use the default `border-image-slice` value of '100%'. What went wrong? Test case: The <button> is rendered with a full border. Details: It seems Chrome doesn't properly override/revert the already applied `border-image-slice` value when the `border-image` shorthand is declared without a `border-image-slice` value. Or the fullhand value wrongly cascades into the shorthand. Using the Developer Tools, the `border-image-slice:1` ruleset appears overridden as it's struck through. In the computed values, however, the <button> still has a `border-image-slice` value of '1'. Clicking this value to see where in the CSS it's declared, points at the `border-image` shorthand, and expands this ruleset with the default value of '100%'. Did this work before? Yes 67 Does this work in other browsers? N/A Chrome version: 71.0.3578.98 Channel: stable OS Version: 10.0 Flash Version: Chrome up until Chrome 67 displays the expected behavior, as well as other browsers like Firefox 64 and Edge 17 (which is based on Chrome 64). Chrome 68, 69, 70 and 71 appears to be affected by this bug, as well as Opera 57 (which is based on Chrome 70).
,
Jan 1
Able to reproduce the issue on the reported chrome version #71.0.3578.98 and latest chrome #73.0.3656.0 using Windows 10, Ubuntu 17.10 and Mac OS 10.13.6 by following steps as per comment#0. The following is the bisect information. Bisect Information: =================== Good Build: 68.0.3401.0 Bad Build: 68.0.3402.0 You are probably looking for a change made after 552305 (known good), but no later than 552306 (first known bad). CHANGELOG URL:https://chromium.googlesource.com/chromium/src/+log/0d41b261e10706d8eed652f198778864401b5ae0..fa1a40ae240b768f36d3a98ae028f4a35acc01e1 Reviewed-on: https://chromium-review.googlesource.com/1013806 As we are not able to assign to the author of the CL "rob.buis@samsung.com", hence assigning to the reviewer "timloh@chromium.org". @Timothy Loh: Please help us in reassigning the issue if it is not related to your change. Thanks.!
,
Jan 2
Rune, can you triage this? It looks like BorderImageSlice::InitialValue() doesn't match the types that BorderImageSlice::ParseSingleValue() creates and it gets silently dropped by CSSToStyleMap::MapNinePieceImageSlice(). It isn't obvious to me whether the early return there and elsewhere in CSSToStyleMap are still relevant.
,
Jan 4
,
Jan 4
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/9153af5e85eca2980b7522886a30368f630c2853 commit 9153af5e85eca2980b7522886a30368f630c2853 Author: Rune Lillesveen <futhark@chromium.org> Date: Fri Jan 04 15:04:39 2019 Use expected CSSValue type for initial border-image-slice. The apply code early returns if the type is not a CSSBorderImageSliceValue. Create an initial value which matches the expected type. Bug: 918243 Change-Id: I3228d2410789306d915db892ad54e918631b8864 Reviewed-on: https://chromium-review.googlesource.com/c/1396118 Reviewed-by: Xida Chen <xidachen@chromium.org> Commit-Queue: Rune Lillesveen <futhark@chromium.org> Cr-Commit-Position: refs/heads/master@{#619951} [modify] https://crrev.com/9153af5e85eca2980b7522886a30368f630c2853/third_party/blink/renderer/core/css/properties/longhands/border_image_slice_custom.cc [add] https://crrev.com/9153af5e85eca2980b7522886a30368f630c2853/third_party/blink/web_tests/external/wpt/css/css-backgrounds/border-image-slice-shorthand-reset.html
,
Jan 4
|
|||||
►
Sign in to add a comment |
|||||
Comment 1 by swarnasree.mukkala@chromium.org
, Dec 31Labels: Triaged-ET Needs-Triage-M71 Needs-Bisect
Status: Untriaged (was: Unconfirmed)