SVG's <use xlink:href="#ид"> does not understand non-ANSI (for example, cyrillic) characters
Reported by
cool...@gmail.com,
Oct 29 2017
|
||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:56.0) Gecko/20100101 Firefox/56.0 Steps to reproduce the problem: navigate to https://jsfiddle.net/CoolCmd/dhg71Lp2/ What is the expected behavior? at the right side, 3 circles must be shown: red, yellow and green What went wrong? only one circle is shown: red those lines fail: <use xlink:href="#red"/> <use xlink:href="#желтый"/> Did this work before? Yes 62 Does this work in other browsers? Yes Chrome version: 63.0.3239.18 (Официальная сборка) beta (64 бит) (cohort: Beta) Channel: beta OS Version: 6.1 (Windows 7, Windows Server 2008 R2) Flash Version: the latest versions of Firefox and Edge are worked as expected
,
Oct 30 2017
Apparently my FF 58 disagree with yours, because it shows all three circle =). Also, the way I'm reading the spec, the fragment referencing the yellow circle would get percent encoded, and thus given percent decoding when using it, it would still point to the expected element?
,
Oct 30 2017
Firefox 58 shows three circles in jsfiddle, but only two circles when loading a local file:///R:/test.html As for the spec, I don't know if the browser should translate #желтый into %-encoded ASCII automatically.
,
Oct 30 2017
Ok, that makes sense I suppose - presumably adding a <meta charset="utf-8"> or similar will result in all three circles being rendered in the file: case.
,
Oct 30 2017
i updated the testcase. according to the comment #2, the link should not work in the Chrome 63, but it works. i think <a href> and <use xlink:href> behavior must be identical. and desirable with automatic encoding.
,
Oct 30 2017
Tested the issue on chrome Stable #62.0.3202.75,Beta #63.0.3239.18 & Canary 64.0.3253.0 on Windows-7 and was able to reproduce the issue. This is a Non-Regression issue since seeing this from M50 #50.0.2624.0, Note : Observed different behaviors from M-50 to M-64 and able to reproduce the issue in MAC 10.12.6 and Linux Ubuntu 14.04.Please find the screen shot for reference. Thank you.
,
Oct 30 2017
,
Oct 30 2017
#6, yep, with <meta charset="utf-8"> all three circles are seen in FF58. I've removed my comments #2 and #3 as no longer pertinent.
,
Oct 31 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/7fed5aeb041c935425adb69333bdf610bef896ee commit 7fed5aeb041c935425adb69333bdf610bef896ee Author: Fredrik Söderquist <fs@opera.com> Date: Tue Oct 31 19:07:17 2017 Percent-decode fragments before using them for SVG references We were using the "raw" form of the fragment, meaning that any percent- encoded fragments would not resolve correctly. Bug: 779420 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2 Change-Id: Ic091e775589fd59d25ad1878115f2ad2593bc733 Reviewed-on: https://chromium-review.googlesource.com/746885 Reviewed-by: Stephen Chenney <schenney@chromium.org> Commit-Queue: Fredrik Söderquist <fs@opera.com> Cr-Commit-Position: refs/heads/master@{#512895} [add] https://crrev.com/7fed5aeb041c935425adb69333bdf610bef896ee/third_party/WebKit/LayoutTests/svg/custom/fill-url-percent-encoding-expected.html [add] https://crrev.com/7fed5aeb041c935425adb69333bdf610bef896ee/third_party/WebKit/LayoutTests/svg/custom/fill-url-percent-encoding.html [add] https://crrev.com/7fed5aeb041c935425adb69333bdf610bef896ee/third_party/WebKit/LayoutTests/svg/custom/use-href-percent-encoding-expected.html [add] https://crrev.com/7fed5aeb041c935425adb69333bdf610bef896ee/third_party/WebKit/LayoutTests/svg/custom/use-href-percent-encoding.html [modify] https://crrev.com/7fed5aeb041c935425adb69333bdf610bef896ee/third_party/WebKit/Source/core/svg/SVGURIReference.cpp [modify] https://crrev.com/7fed5aeb041c935425adb69333bdf610bef896ee/third_party/WebKit/Source/core/svg/SVGUseElement.cpp
,
Nov 1 2017
,
Dec 8 2017
Issue 792848 has been merged into this issue.
,
Dec 8 2017
should we consider this for a re-roll of M63? Code change looks trivial enough and had sufficient baketime
,
Dec 8 2017
This bug requires manual review: Request affecting a post-stable build Please contact the milestone owner if you have questions. Owners: cmasso@(Android), cmasso@(iOS), gkihumba@(ChromeOS), govind@(Desktop) For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
,
Dec 12 2017
M63 version 63.0.3239.84 has been out since 12/06/17. Do we have any report this affects real websites? If not, we WON'T be taking this change in for next M63 respin (if any).
,
Dec 12 2017
None of the reports we've had has mentioned a real website. (If such an example exist, now would be a good time to mention it.)
,
Dec 13 2017
Hi, I've reported the issue https://bugs.chromium.org/p/chromium/issues/detail?id=792848 which has been merged into this issue. I'm one of the developer of www.combeenation.com, we provide a system with which our customers (mostly companies) are able to create configurators for their customizable products. Those configurators have sometimes 2D graphics, which we generate via SVG based on their input values. And since some of our customers are based in germany/austria, they use a lot of umlauts, therefore we had a lot of troubles when the new Chrome 63 rolled out. Our frontend dev deployed a hotfix where basically all umlauts are replaced via RegEx, but we don't really see this as a long term solution. An example of a configurator which is using umlauts in SVG (Hotfix is enabled, therefore the graphic is working): https://portal.combeenation.com/Cfgr/pieno/DOORS So yes we are a real website.
,
Dec 13 2017
Tested the issue on Windows-7, Ubuntu 14.04 and Mac OS 10.12.6 using chrome latest Canary M65-65.0.3293.0 & 65.0.3287.0(Chrome OS) by following steps mentioned in the original comment. Observed that images are displaying as expected. Hence adding TE-Verified label. Please find the snap shot for reference. Note: For Linux tested in 65.0.3292.0 due to 65.0.3293.0 build failure. Thank you!
,
Dec 13 2017
The decision has been made to not merge this into M-63 given the point we're at in the release cycle, the riskiness of the change, and the existence of a (admittedly not great) workaround. Look for it in Chrome 64.
,
Dec 13 2017
Rejecting merge to M63 based on comment #20.
,
Jan 26 2018
Issue 806177 has been merged into this issue. |
||||||||||
►
Sign in to add a comment |
||||||||||
Comment 1 by cool...@gmail.com
, Oct 29 2017