Re: follow-link not on mouse-face

[David Kastrup]

"Richard M. Stallman" <address@hidden> writes:

>     I think we should simply remove all special properties from the context
>     lines: they're just context and do not really mean anything wrt the
>     match itself.  When you click on a context line, point moves to the
>     match line in the buffer, not to the line you clicked in the Occur
>     buffer, which doesn't make much sense.
>
> I agree, as regards those context lines.
>
>     With the following patch, mouse actions and highlighting are put on the
>     match line only, which solves the problem.
>
> The change you described about context lines does not solve the
> problem, as I understand the problem.  I thought the problem was about
> the line numbers on match lines.
>
> Your patch has the effect of turning off Mouse-1 on the line
> numbers of match lines.  Whether that is desirable is the question
> we were discussing, and we had not reached a conclusion yet.

Double mouse clicks already invoke mouse-2, and it was not deemed
desirable to have whole lines flash in green, and it was not deemed
desirable to use underlining for links sice they should be
conspicuous.

In other GUIs, double-clicking on something for following links was
pretty common.  We implemented the double-clicking stuff so that one
could use mouse-1, too, even where follow-link was not active.  It was
supposed to placate the people with button-challenged mouses.

I don't see how the issue has suddenly become open again, as Drew
contends, just because the implementation does not yet seem to do what
we agreed on, namely doing mouse-highlighting and follow-link with
mouse-1 just on line numbers.  I don't remember exactly whether there
was agreement about whether it would be still a good idea to let
mouse-2 and double-mouse-1 (but not mouse-1) on the rest of the line
act as a non-highlighted link.

I can't accept the argument that people would expect to have mouse-1
and mouse-2 always do the same thing.  It does not do that elsewhere
in Emacs.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum