emacs-pretest-bug
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: before- and after-string issues


From: martin rudalics
Subject: Re: before- and after-string issues
Date: Wed, 18 May 2005 14:51:06 +0200
User-agent: Mozilla Thunderbird 1.0 (Windows/20041206)

> So perhaps this
>
>        1. All after-strings come in front of before-strings, except
>        when they come from the same overlay.
>
> should be changed to this:
>
>        1. The overall order is as follows:
>
>        All after-strings from overlays whose end-markers don't
>        advance, and text properties that are not rear-sticky,
>
>        All before-strings from overlays whose beginning-markers don't
>        advance, and text properties that are front-sticky,
>
>        All after-strings from overlays whose end-markers do advance,
>        and text properties that are rear-sticky,
>
>        All before-strings from overlays whose beginning-markers don't
>        advance, and text properties that are not front-sticky,
>

In the last entry you probably mean

        All before-strings from overlays whose beginning-markers do
        advance, and text properties that are not front-sticky,

>
> At least, it seems to be right in all the cases as far as I can see.
>

As long as there are no empty overlays: Consider two non-empty overlays
with the same start and end position.  One has a before-string the other
an after-string.  Now delete all characters covered by the overlay.
When you delete the last one the strings exchange positions.  Your
proposal wouldn't change this.  If the after-string overlay has an
end-marker that advances, inserting a character will now re-exchange the
position of the strings.  Confusing, at least.

But I also fail to understand why you included text properties in the
discrimination.  before- and after-strings are overlay properties,
stickyness is a text property.  The face of a before-/after-string would
be affected by stickyness iff you allowed inheritance of face properties
from ordinary characters to adjacent before-/after-strings.  This
(inheritance, yes/no) would have to be decided in the first place.  The
relative order of such strings during display should not be affected by
stickyness properties.  Well, maybe I'm too silly to understand this.






reply via email to

[Prev in Thread] Current Thread [Next in Thread]