Re: transient-mark-mode and region highlighting

[Milan Zamazal]

>>>>> "RMS" == Richard Stallman <address@hidden> writes:

    RMS> These two features naturally go together.  The highlighted
    RMS> region tells you that commands such as replace-string will
    RMS> operate on the region.  

This is an inconvenient way of signaling for me, I prefer more moderate
indicator (that I can set up myself, e.g. a flag in the mode line) if I
need one.

    RMS> So I don't expect many users to want to enable the second and
    RMS> not the first.

    RMS> Do you see a reason to think many users want this?

I don't think too many users want this, but those who want don't have
any means to set up it in their ~/.emacs.

>>>>> "PJ" == Pavel Janík <address@hidden> writes:

    PJ> I do not use it myself, because instead of this "region
    PJ> sensitivity" of transient-mark-mode, I (*now*) prefer to narrow
    PJ> to region, do what I want, and widen again. But I understand
    PJ> that it may be a lot faster for me to turn the highlighting off
    PJ> and use transient-mark-mode for this specific reason.

One of the especially annoying side effects of the current
transient-mark-mode is region highlighting when one performs I-search
and then returns back with `C-x C-x'.

Please note that region narrowing as suggested by Pavel, is not a
solution because commands like comment-dwim check for
transient-mark-mode and an active region.  Also, simply customizing the
`region' face to make it invisible doesn't help either, because this
makes mouse selections invisible.  Since this is the only case when the
region highlighting is useful to me, another solution could be to put
mouse selections under a different face.  I'd be happy with it, but
introducing some transient-mark-mode highlighting flag seems to be more
logical to me.

Regards,

Milan Zamazal

-- 
http://www.zamazal.org