[Webkit-unassigned] [Bug 270508] Combining character temporarily underlines rest of paragraph

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Mar 5 09:13:18 PST 2024


https://bugs.webkit.org/show_bug.cgi?id=270508

Jean Abou Samra <jean at abou-samra.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jean at abou-samra.fr

--- Comment #2 from Jean Abou Samra <jean at abou-samra.fr> ---
> What's the procedure for typing a compose character?

On a standard French keyboard layout (AZERTY), the characters â, ê, î, … are typed by pressing the key with a "^" symbol, then the a, e, i, … key. (To get an ASCII caret "^", you press that key twice.) It's the same principle with "¨".

I actually don't know if speaking of a "combining character" was really correct. With the help of a Python REPL, I can see that what my keyboard inserts when I press "^" then "a" is a single character, U+0XE2 LATIN SMALL LETTER A WITH CIRCUMFLEX (this: "â"), and not the sequence with a combining character that you get if you convert that to Unicode NFD form, namely U+0061 LATIN SMALL LETTER A followed by U+0302 COMBINING CIRCUMFLEX ACCENT (this: "â").

So, from the user point of view, the "^" key is bound to a combining circumflex accent, but technically it is inserting precomposed characters. I don't know if this is the case "from the start" or if there is some layer doing NFC normalization or such. I confess I'm ignorant of how keyboards work under Wayland.

> Is it a regression? (If I can bisect it, then it's much more likely to be resolved.)

As far as I remember, it has always been like this on Fedora 39. I can't recall how it worked in the more distant past, sorry. My version of the webkit2gtk4.1 is 2.42.5 (the latest). I just did "sudo dnf downgrade webkit2gtk4.1" and it gave me webkit2gtk4.1 2.42.1, but that didn't fix it.

What I find the most puzzling is that Evolution's Markdown mode doesn't have this issue. @Milan Crha, maybe you have insights on what it configures differently in WebKit compared to the HTML mode or plain text mode?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20240305/761576ce/attachment.htm>


More information about the webkit-unassigned mailing list