bug#44466: 27.1; quail input fails at read-only boundary

[Eli Zaretskii]

> From: lg.zevlg@gmail.com
> Date: Sun, 8 Nov 2020 23:50:54 +0400
> Cc: dick.r.chiang@gmail.com, 44466@debbugs.gnu.org
> 
> 
> > 8 нояб. 2020 г., в 20:18, Eli Zaretskii <eliz@gnu.org> написал(а):
> > 
> > 
> >> 
> >> From: lg.zevlg@gmail.com
> >> Date: Sun, 8 Nov 2020 20:07:46 +0400
> >> Cc: dick.r.chiang@gmail.com, 44466@debbugs.gnu.org
> >> 
> >>> Any reason why pressing a key on a button should disregard the active
> >>> input method?  What if the button needs the user to type the character
> >>> which the input method would produce?  AFAIU, the patch we installed a
> >>> year ago makes this impossible.
> >> 
> >> The reason is the same as for read-only buffers, making single char 
> >> bindings work. For example if you change input method in image-mode and 
> >> press “q” key this will kill buffer, because correct command is executed.
> > 
> > If this is a feature, it should be documented somewhere.  Is it?
> 
> I guess so, this feature was in Emacs for years. Usually people marks whole 
> buffer as read-only and then mark special places to inhibit read-only, so 
> quail translation inhibits for read-only parts and does the translation in 
> writable parts. In telega.el project we do opposit - insert read-only parts 
> in writable buffer and expect quail translation not to apply in that 
> read-only parts of the buffer. We might change this behaviour, so patch to 
> quail could be reverted, however I think it is better to fix this in quail, 
> to reflect the logic in internal checks for region writability

Thanks, I've now fixed this on the emacs-27 branch.