strangely-placed cursor in summary buffer

strangely-placed cursor in summary buffer

[Bill White]

I've recently redesigned my summary buffer and noticed that the
default position of the cursor is strange - it starts out in column
23.  I guess I'd like it in column 1.  Any tips?

See <url:http://members.wri.com/billw/cursor-position.jpg>

Cheers -

bw
-- 
Bill White . billw@wolfram.com . http://members.wri.com/billw
"No ma'am, we're musicians."

Re: strangely-placed cursor in summary buffer

[Bill White]

On Fri Jul 27 2001 at 08:12, Bill White <billw@wolfram.com> said:

    > I've recently redesigned my summary buffer and noticed that the
    > default position of the cursor is strange - it starts out in
    > column 23.  I guess I'd like it in column 1.  Any tips?
    > 
    > See <url:http://members.wri.com/billw/cursor-position.jpg>

Sorry - forgot version info:

Oort Gnus v0.04 (nightly build), GNU Emacs 21.0.104.1
(i586-pc-linux-gnu, X toolkit) of 2001-07-13 on g.wolfram.com

Cheers -

bw
-- 
Bill White . billw@wolfram.com . http://members.wri.com/billw
"No ma'am, we're musicians."

Re: strangely-placed cursor in summary buffer

[Karl Kleinpaste]

Bill White <billw@wolfram.com> writes:
> I've recently redesigned my summary buffer and noticed that the
> default position of the cursor is strange - it starts out in column
> 23.  I guess I'd like it in column 1.

The quality in question isn't "column 23"; it's "column just after the
1st `:' in the line."  This is based on the idea of the default
summary line which contains "LineCount `:' Author", so the cursor
sits right beside the author name.

R   > [  13: Bill White          ] strangely-placed cursor in summary buffer
            ^
This is to say, you've invalidated an (arguably bad) assumption on
Gnus' part, regarding what it should find in summary lines.

Maybe Gnus should have a %C cursor position specifier, in both *Group*
and *Summary*.

Re: strangely-placed cursor in summary buffer

[ShengHuo ZHU]

Bill White <billw@wolfram.com> writes:

> I've recently redesigned my summary buffer and noticed that the
> default position of the cursor is strange - it starts out in column
> 23.  I guess I'd like it in column 1.  Any tips?
> 
> See <url:http://members.wri.com/billw/cursor-position.jpg>

Since you customized gnus-summary-line-format, why not customize
gnus-summary-position-point (a function) too?

ShengHuo

Re: strangely-placed cursor in summary buffer

[Bill White]

On Fri Jul 27 2001 at 11:25, ShengHuo ZHU <zsh@cs.rochester.edu> said:

    > Bill White <billw@wolfram.com> writes:
    > 
>> I've recently redesigned my summary buffer and noticed that the
>> default position of the cursor is strange - it starts out in column
>> 23.  I guess I'd like it in column 1.  Any tips?
>> 
>> See <url:http://members.wri.com/billw/cursor-position.jpg>
    > 
    > Since you customized gnus-summary-line-format, why not customize
    > gnus-summary-position-point (a function) too?

Well that's pretty dang cool.  Thanks for pointing it out!

Cheers -

bw
-- 
Bill White . billw@wolfram.com . http://members.wri.com/billw
"No ma'am, we're musicians."

Re: strangely-placed cursor in summary buffer

[Kai Großjohann]

On Fri, 27 Jul 2001, Karl Kleinpaste wrote:

> Maybe Gnus should have a %C cursor position specifier, in both
> *Group* and *Summary*.

Whoo-hoo!  *clap clap clap*

kai
-- 
~/.signature: No such file or directory

Re: strangely-placed cursor in summary buffer

[Lars Magne Ingebrigtsen]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

>> Maybe Gnus should have a %C cursor position specifier, in both
>> *Group* and *Summary*.
>
> Whoo-hoo!  *clap clap clap*

Sounds like an idea.  Hm...  What would be an efficient
implementation?  It has to know where to go on every single line -- I
think that means keeping an alist of article numbers/character
positions updated.

And the line insertion functions would have to treat %C as a break
point, which means that (potentially) the summary buffer generation
would be slower.  And keeping that alist updated would certainly not
be pleasant.

Just having the cursor go to a pre-defined character is so much
easier.  :-)

Of course, %C could be implemented as a character of the users'
choice, and then `gnus-summary-position-point' (and friends) would
just skip to that character instead.  But that would severly limit
which characters the user could choose for the %C marker.

No-break space would be a nice, invisible choice, I guess.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: strangely-placed cursor in summary buffer

[Lars Magne Ingebrigtsen]

Oh my.  It's so odd seeing my entire name in the summary buffer.  I'd
gotten so used to being named "Lars Magne Ingebrigt".  That's been my
name since, er, 1991 or something.  :-)

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: strangely-placed cursor in summary buffer

[Kai Großjohann]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Of course, %C could be implemented as a character of the users'
> choice, and then `gnus-summary-position-point' (and friends) would
> just skip to that character instead.  But that would severly limit
> which characters the user could choose for the %C marker.

You could put a text property on the spot where the cursor should go.
Then putting the cursor in the right spot can be done by putting it on
the beginning of the line and then searching for the next change of
that property.

How about that?

(Is it possible to put a text property on a region which contains zero
characters?  If all else fails, you can still put the text property on
the character just before the %C, or just after the %C.)

kai
-- 
~/.signature: No such file or directory

Re: strangely-placed cursor in summary buffer

[Lars Magne Ingebrigtsen]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> You could put a text property on the spot where the cursor should go.
> Then putting the cursor in the right spot can be done by putting it on
> the beginning of the line and then searching for the next change of
> that property.
>
> How about that?

Text properties makes things go slower in general.  Back when I used
to fiddle with such stuff, it seemed like there was some exponential
delay in buffer updating with the number of text props there were in
the buffer.

However, things are better now, aren't they?  Perhaps text properties
are the right solution here...  Hm.

I think they are the best solution; yes.  It does mean that they'll be
sync points when generating the summary lines.  That is, where there
is now a (format "lotsastuff"), there will be (format "somestuff"),
put text prop, (format "somemorestuff").  This performance hit will
only, er, hit the people who actually use the %C spec, which is
something I like.

Ok.  %C as a text prop it is...

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: strangely-placed cursor in summary buffer

[Kai Großjohann]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> I think they are the best solution; yes.  It does mean that they'll be
> sync points when generating the summary lines.  That is, where there
> is now a (format "lotsastuff"), there will be (format "somestuff"),
> put text prop, (format "somemorestuff").  This performance hit will
> only, er, hit the people who actually use the %C spec, which is
> something I like.

Stupid question: doesn't Gnus use text properties in the summary
buffer already?  For %< and %> or something?

Hm.  Btw, people have been asking how to make the names come out green
and the subjects come out yellow in the summary buffer.  While you're
at it...

kai
-- 
~/.signature: No such file or directory

Re: strangely-placed cursor in summary buffer

[Colin Marquardt]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Hm.  Btw, people have been asking how to make the names come out green
> and the subjects come out yellow in the summary buffer.  While you're
> at it...

I have that at home, after it was posted to... somewhere. In case
that code is wanted, I could get it.

Colin

Re: strangely-placed cursor in summary buffer

[Lars Magne Ingebrigtsen]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Stupid question: doesn't Gnus use text properties in the summary
> buffer already?  For %< and %> or something?

Yes, it does.  More text props is more text props, though.  But this
is probably not an issue any more.  It's been years since I did
benchmarking on this...

> Hm.  Btw, people have been asking how to make the names come out green
> and the subjects come out yellow in the summary buffer.  While you're
> at it...

Isn't that simple?  You just specify the font specs.  (This didn't use
to be possible, but I think it was fixed early this year...)

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: strangely-placed cursor in summary buffer

[Colin Marquardt]

Colin Marquardt <colin.marquardt@usa.alcatel.com> writes:

> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
>
>> Hm.  Btw, people have been asking how to make the names come out green
>> and the subjects come out yellow in the summary buffer.  While you're
>> at it...
>
> I have that at home, after it was posted to... somewhere. In case
> that code is wanted, I could get it.

It's really simple:

(copy-face 'default 'face-1)
(set-face-foreground 'face-1 "steelblue")
(setq gnus-face-1 'face-1)
(copy-face 'default 'face-2)
(set-face-foreground 'face-2 "lime green")
(setq gnus-face-2 'face-2)
(copy-face 'default 'face-3)
(set-face-foreground 'face-3 "pink")
(setq gnus-face-3 'face-3)
(setq gnus-summary-line-format "%U%R%z: %1{%-15,15n%} %2{%d%} %3{%[%4L%]%}  %(%I%S%) \n")

Cheers,
  Colin