Re: [LyX/master] Add support for IM attribute specification over multiple segments

Scott Kostyshak <[email protected]> Sun, 14 Jun 2026 15:25:46 +0200
Newsgroups gmane.editors.lyx.devel
Message-ID <ai6rkTC5HSkB5DO0@jimmy>
On Sun, Jun 14, 2026 at 03:11:48PM +0200, Kornel Benko wrote:
> Am Sun, 14 Jun 2026 12:59:44 +0200
> schrieb Scott Kostyshak <[email protected]>:
> 
> > On Sun, Jun 14, 2026 at 11:13:31AM +0200, Kornel Benko wrote:
> > > Am Sun, 14 Jun 2026 05:04:06 +0000
> > > schrieb Koji Yokota <[email protected]>:
> > >   
> > > > commit f198a3847c685e29565ffb952caf7870ba45bc60
> > > > Author: Koji Yokota <[email protected]>
> > > > Date:   Sun Jun 14 13:57:45 2026 +0900
> > > > 
> > > >     Add support for IM attribute specification over multiple segments
> > > >     
> > > >     Although QInputMethodEvent documents that there should be at most one
> > > >     text format attribute specified on each segment, the de-facto standard
> > > >     seems different. LyX allowed duplicate attribute specification segment
> > > >     by segment so far. This fix extends it to allow overwrapped
> > > >     specification over segments.
> > > >     
> > > >     Eg. existing attribute  [   seg   ] [charFormat for a segment] [seg]
> > > >         incoming attribute  [ <------ charFormat over segments ------> ]
> > > > ---
> > > >  src/frontends/qt/GuiInputMethod.cpp | 192 ++++++++++++++++++++++++++++--------
> > > >  src/frontends/qt/GuiInputMethod.h   |   3 +-
> > > >  src/frontends/qt/GuiPainter.cpp     |   4 +-
> > > >  3 files changed, 155 insertions(+), 44 deletions(-)  
> > > ...
> > > 
> > > Now I get the attached.  
> > 
> > Interesting, I can't reproduce. Maybe a difference in versions?
> > 
> > $ g++ --version
> > g++ (Ubuntu 15-20250404-0ubuntu1) 15.0.1 20250404 (experimental) [master
> > r15-9193-g08e803aa9be]
> > 
> > $ clang++ --version
> > Ubuntu clang version 20.1.2 (0ubuntu1)
> > 
> > Scott
> 
> Mine is
> $ /usr/lib/ccache/g++-14 --version
> g++-14 (Ubuntu 14.2.0-4ubuntu2~24.04.1) 14.2.0
> Copyright (C) 2024 Free Software Foundation, Inc.

Might in the end depend on different optimization flags we use, who
knows.

If I had more time and motivation I might be curious to look into this
but not much time these days.

Scott

-- 
lyx-devel mailing list
[email protected]
https://lists.lyx.org/mailman/listinfo/lyx-devel
signature.asc (application/pgp-signature, 488 B)
-----BEGIN PGP SIGNATURE-----

iQEzBAEBCgAdFiEEMfkICpsJzYuNuGlOA4ZnarZHC+sFAmouq9oACgkQA4ZnarZH
C+s29Qf/acMtJ8rEIXLdOmIZeCubVU4nwGe8nozX9ehcmpNDXwqAAwZgf/nHEvFn
/weBjZyk2ejfSoqJepqHAapgidi1CIsvERn/70gsrPF53yuWE81ZaL3pcWYUo7Me
sl1h9hMVdixXcPrIoeG1Uzy0G65XPz/CYjdD4YymPEKVKnJgp98mFiE3+k1cu/FM
X0zvXD7q5UvZTTgXbQ6X30BhwJRfQRhTAZjCQaFftbB/vbyazEn5VdEHx2YI0Ui7
n26yXr/CFKZCcXsjXD/inpilSCKYlHQ0kZFwwCemGzIQl+ksIRQaPnCG+3E3Vk6a
yMZK4QSa4orWT1/DFBmXGruVHbcwRA==
=Zn6r
-----END PGP SIGNATURE-----