240279 2022-05-10 08:38:43 -0700 Traditional Mongolian script characters incorrectly oriented 2023-12-22 10:28:31 -0800 1 1 1 Unclassified WebKit Layout and Rendering Safari 15 Unspecified Unspecified RESOLVED FIXED https://bugs.webkit.org/show_bug.cgi?id=112488 InRadar P2 Normal --- 1 ishida webkit-unassigned ahmad.saleem792 bfulgham fantasai.bugs ishida mmaxfield simon.fraser webkit-bug-importer xidorn-webkit zalan oldest_to_newest 1867936 0 ishida 2022-05-10 08:38:43 -0700 Lines of text in the Traditional Mongolian script run vertically from top to bottom of the page. Lines flow from left to right. The script is cursive, meaning that the characters join along the vertical centre line. Here is an example: https://user-images.githubusercontent.com/4839211/167632640-cb805b08-59a1-4c73-a28f-b47f7048f153.png Note the orientation of the glyphs. Specs: css-writing-modes-4 provides the writing-modes property. The vertical-lr value is designed to display Traditional Mongolian text as expected. Tests & results: i18n test suite, Vertical text, Vertical-LR, Glyphs (https://w3c.github.io/i18n-tests/results/writing-mode-vertical#lr_glyphs) Results Gecko: ✅ Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:100.0) Gecko/20100101 Firefox/100.0 Blink: ✅ Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/101.0.4951.54 Safari/537.36 WebKit: ❌ Rotates lines appropriately, but doesn't display characters in the correct orientation and doesn't join them up (see the left-hand line below). Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.4 Safari/605.1.15 https://user-images.githubusercontent.com/4839211/105961907-09663500-6077-11eb-9034-13afb495de68.png To get this to work in WebKit you need to add text-orientation: sideways to the CSS, which should not be necessary, and is not needed for Gecko and Blink browsers. This issue is being tracked by the W3C at https://w3c.github.io/mlreq/gap-analysis/#issue39_vertical_text Another (long) bug was raised in 2013 and ground to a halt in 2014 (8 years ago!). I'm starting a new bug because i want to just focus on making Mongolian look correct and getting WebKit to be interoperable with other browsers from a content author perspective. (See https://bugs.webkit.org/show_bug.cgi?id=112488) 1867962 1 mmaxfield 2022-05-10 10:10:16 -0700 This has the same root cause as https://bugs.webkit.org/show_bug.cgi?id=212128 1869694 2 webkit-bug-importer 2022-05-17 08:39:11 -0700 <rdar://problem/93426525> 1879500 3 bfulgham 2022-06-30 13:29:59 -0700 Marking as Dupe of Bug 212128 to match Radar state. *** This bug has been marked as a duplicate of bug 212128 *** 1879690 4 ishida 2022-07-01 01:42:14 -0700 I had checked bug 212128 before raising this, and concluded that bug 212128 may involve a fix for a small problem that is not necessarily the fix needed for this much more generalised problem. Certainly, for CJK, rotation of punctuation is handled independently from correctly rotating ordinary text. It may be that Mongolian is a special case, but then i still think that bug 212128 should be rather classed as superceded by this issue, and some text should be added to the bug report to indicate that this is a much bigger problem than just fixing parentheses and brackets. 1880896 5 mmaxfield 2022-07-05 21:38:21 -0700 I think we would probably fix all of these bugs in a single patch (or single patch series). Under the assumption that we probably only need a single bug to track these incorrectly rotated characters, do you think we should reverse the duping relationship and dup that bug to this one instead? 1880897 6 ishida 2022-07-05 21:41:05 -0700 Yeah, that's what i was thinking, since it seems to be a superset that includes the other bug. I agree that it would seem to make sense to fix all in a single patch. 1880910 7 mmaxfield 2022-07-05 23:16:07 -0700 Reversing dup direction. 1880912 8 mmaxfield 2022-07-05 23:16:39 -0700 *** Bug 212128 has been marked as a duplicate of this bug. *** 1990283 9 ahmad.saleem792 2023-11-06 13:48:40 -0800 Blink merge: https://chromium.googlesource.com/chromium/blink/+/e3ba3532accd84e6a0253705595b97556ae7ca67 In my local testing by adding above in WebKit, it progress both test cases mentioned in Comment 0 and also one of WPT: https://wpt.fyi/results/css/css-writing-modes/mongolian-orientation-002.html?label=master&label=experimental&aligned&q=Mongolian 1990284 10 ahmad.saleem792 2023-11-06 13:49:45 -0800 (In reply to Ahmad Saleem from comment #9) > Blink merge: > https://chromium.googlesource.com/chromium/blink/+/ > e3ba3532accd84e6a0253705595b97556ae7ca67 > > In my local testing by adding above in WebKit, it progress both test cases > mentioned in Comment 0 and also one of WPT: > > https://wpt.fyi/results/css/css-writing-modes/mongolian-orientation-002. > html?label=master&label=experimental&aligned&q=Mongolian WebKit Source: https://searchfox.org/wubkat/source/Source/WebCore/platform/graphics/FontCascadeFonts.cpp#224 Line 235 & Line 270. 1990286 11 ahmad.saleem792 2023-11-06 13:52:03 -0800 (In reply to Ahmad Saleem from comment #10) > (In reply to Ahmad Saleem from comment #9) > > Blink merge: > > https://chromium.googlesource.com/chromium/blink/+/ > > e3ba3532accd84e6a0253705595b97556ae7ca67 > > > > In my local testing by adding above in WebKit, it progress both test cases > > mentioned in Comment 0 and also one of WPT: > > > > https://wpt.fyi/results/css/css-writing-modes/mongolian-orientation-002. > > html?label=master&label=experimental&aligned&q=Mongolian > > WebKit Source: > https://searchfox.org/wubkat/source/Source/WebCore/platform/graphics/ > FontCascadeFonts.cpp#224 > > Line 235 & Line 270. Note - it fixes this as well from other bug: http://www.mongolfont.com/test/webkit/div.html 1990997 12 ahmad.saleem792 2023-11-08 16:16:10 -0800 Attempt PR - https://github.com/WebKit/WebKit/pull/20195 2001609 13 ews-feeder 2023-12-22 10:28:28 -0800 Committed 272454@main (6c53862898c9): <https://commits.webkit.org/272454@main> Reviewed commits have been landed. Closing PR #22186 and removing active labels.