On 9/13/06, Philip Ganchev <[EMAIL PROTECTED]> wrote:
> On 9/11/06, Axel Liljencrantz <[EMAIL PROTECTED]> wrote:
> > On 9/9/06, Martin Bähr <[EMAIL PROTECTED]> wrote:
> [...]
> > > when highlighting a matching ([{" the cursor gets hidden, and every time
> > > it happens i am getting lost as to where the cursor is, and where the
> > > next insert or delete will happen. i am trying to guess, but i am almost
> > > always guessing wrong.
> [...]
> > That's not good. I didn't know about this problem since the terminals
> > I use don't exhibit this behaviour, the background color is always
> > different under the cursor. But it's not hard to disable background
> > color changes under the cursor. I'll add that feature.
>
> Yes, with Fish 1.21.10 and Xterm, the cursor background color is
> always different than the normal background color. I originally
> reported the dissapearing cursor for Fish 1.21.8.
Ok.
>
> But I would say it is still hard to tell which of the two colored
> backgrounds represents the cursor. For example, suppose normal color
> is black on white, 'fish_color_match' is 'yellow' and the cursor is
> over a parenthesis. Then the parenthesis is bold yellow on grey and
> the matching parenthesis is white on yellow (actually on brown).
> White on brown is more visually distinct, so it is tempting to guess
> that it is the cursor.
>
> Also, it is not quite obvious that the characters match, because their
> colors do not match -- although they are similar in that they both
> have colored backgrounds. It would be more obvoius if they were both
> (for example) bright yellow, one on normal background, the other on
> cursor background.
>
> You can make the cursor always be the same background color, but then
> that must be a reserved color that can never be a foreground.
The actual cursor color is unknown and close to impossible to even
make a qualified guess on in a reliable way. The same thing is true
for the foreground and background colors.
>
> Instead, you can make it invert its usual background only when over a
> character of the same color. But there may still be visual problems.
> For example, on common terminals like Xterm it is hard to read blue on
> black, purple on red, yellow (brown) on red or purple, cyan on green,
> bold blue on purple, white on cyan or green, magenta on yellow, cyan
> on "white" (grey), and others.
>
> This is solved if the cursor is always inverted. This may be more
> consistent than inverting only sometimes as above. But then, to make
> it clear where the cursor is, perhaps normal text should never be
> inverted.
Fish has no control over if the cursor inverts colors, keeps the
original colors or always uses the same color. Nor can fish control
what happens to the cursor when over a non-default background color.
>
> Ideally, the following conditions would be satisfied (at least with
> default colors):
> 1. all text on the command line is easily legible
> 2. it is always obvious where is the cursor
> 3. matching pairs of characters are visible
Conditions 1 and 2 are satisfied by the new scheme, which is to drop
all color information on the character under the cursor. Condition
three is not fulfilled, but I don't see how we can reliably fulfill it
without jeopardizing conditions 1 and 2. A command line program simply
does not have enough control over the terminal to do the right thing
in these situations. :(
--
Axel
-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Fish-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/fish-users
