"Neil Hodgson" <[EMAIL PROTECTED]> wrote:
> 
>    After some discussion, Peufeu has implemented this and the
> implementation can be downloaded from the SourceForge tracker. An
> example of current and modified behaviour is in some animated GIFs:
> http://peufeu.free.fr/scite/normal-scite.gif
> http://peufeu.free.fr/scite/patched-scite.gif
> 
>    If people like this I'll probably include something similar as an
> option. It would be optional as it wouldn't work in all cases and will
> make tabs larger.

Having tabs lining up so well in variable width fonts is pretty spiffy,
but I would consider going a bit farther and have all spaces be the same
width and the tab width being a function of spaces per tab and space
width.

I personally have a fairly visually pleasing fixed-width font, and
generally prefer having all of my columns line up, so while I wouldn't
use the feature, it seems like a worthwhile addition.

 - Josiah

P.S. I have been looking into a method that would allow >1 byte of
styling information per character, but it would seem to necessitate a
fairly substantial rewriting of cellbuffer, a possible re-architecting
of it to locate style separate from character content, and a replacement
of styling information assuming char returns for StyleAt(), etc.

Such a modification would further necessitate changes to basically every
application that currently uses the Scintilla editing component and at
least a dozen of the lexers. While many of these changes would generally
be on the order of replacing 'char' with 'int' and understanding that
one could get a style > 255 (this is assuming that 2 style bytes are
desireable generally, and/or 4 would be desireable for certain features
(think 16 or 24 indicators for SubEthaEdit-like group programming,
marking, etc.).

With a little work, it wouldn't be terribly difficult to offer
programmatic changing of the number of bytes for styling information to
1, 2, or 4, but ultimately it is a question of: would having the option
of 1, 2, or 4 styling bytes be a desireable feature? And would such a
feature be desireable enough to necessitate changing Scintilla-derived
programs?


_______________________________________________
Scintilla-interest mailing list
[email protected]
http://mailman.lyra.org/mailman/listinfo/scintilla-interest

Reply via email to