Make button face appear magenta on a Linux console

Make button face appear magenta on a Linux console

[Nick Roberts]

The button face uses underlining but the Linux console doesn't have that
capability.  So although the [back] button in a help buffer suggests a link,
it's not apparent that you can press <RET>/click on the the word `customize'
(which is unquoted in the help buffer) to customize a variable.  I see that in
an xterm that doesn't have underlining, the button face has a magenta
foreground (and in one that does, it keeps it's default colour - how clever is
that!).  Can someone who knows how this all works give the button face on a
Linux console a magenta foreground please? (not directly though, but via the
underline mechanism).

-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: Make button face appear magenta on a Linux console

[Andreas Schwab]

Nick Roberts <nickrob@snap.net.nz> writes:

> Can someone who knows how this all works give the button face on a
> Linux console a magenta foreground please? (not directly though, but via the
> underline mechanism).

This is a limitation of the linux console.  The underline and dim video
modes cannot be used with color (see the ncv capability).

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Make button face appear magenta on a Linux console

[Nick Roberts]

 > > Can someone who knows how this all works give the button face on a Linux
 > > console a magenta foreground please? (not directly though, but via the
 > > underline mechanism).
 > 
 > This is a limitation of the linux console.  The underline and dim video
 > modes cannot be used with color (see the ncv capability).

I think you're saying Xterm does the work, not Emacs.  OK, how about giving
button face a magenta foreground directly then?  Is it possible to specify
the console and not an xterm in defface, i.e, not using min-colors?


-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: Make button face appear magenta on a Linux console

[Andreas Schwab]

Nick Roberts <nickrob@snap.net.nz> writes:

>  > > Can someone who knows how this all works give the button face on a Linux
>  > > console a magenta foreground please? (not directly though, but via the
>  > > underline mechanism).
>  > 
>  > This is a limitation of the linux console.  The underline and dim video
>  > modes cannot be used with color (see the ncv capability).
>
> I think you're saying Xterm does the work, not Emacs.

I don't understand what work you mean.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Make button face appear magenta on a Linux console

[Nick Roberts]

 > >  > > Can someone who knows how this all works give the button face on a
 > >  > > Linux console a magenta foreground please? (not directly though, but
 > >  > > via the underline mechanism).
 > >  > 
 > >  > This is a limitation of the linux console.  The underline and dim video
 > >  > modes cannot be used with color (see the ncv capability).
 > >
 > > I think you're saying Xterm does the work, not Emacs.
 > 
 > I don't understand what work you mean.

If I do "man ls" in the xterm without underlining, words that are meant to be
displayed with an underline are displayed with a magenta foreground (likewise
bold is displayed as blue).

If a console can't do this, then presumably Emacs could by specifying the
button face as magenta.

Another possibility is to make the button face inherit from link, or info-xref
face, then, by default it gets displayed as cyan.  A long time ago I had a
patch (which I've lost now) which kept track of destinations visited in the
help buffer and used info-xref-visited for them.

-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: Make button face appear magenta on a Linux console

[Andreas Schwab]

Nick Roberts <nickrob@snap.net.nz> writes:

> If I do "man ls" in the xterm without underlining,

What do you mean with "xterm without underlining"?

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Make button face appear magenta on a Linux console

[Nick Roberts]

 > > If I do "man ls" in the xterm without underlining,
 > 
 > What do you mean with "xterm without underlining"?

Clearly I don't know exactly what I mean, or I wouldn't need to ask, but I
don't think it needs much imagination and if if you keep replying to my
question with another question clearly I'm not going to get anywhere.

I mean if I have an nroff file that starts:

  .\" DO NOT MODIFY THIS FILE!  It was generated by help2man 1.28.
  .TH LS "1" "July 2002" "ls (fileutils) 4.1.11" "User Commands"
  .SH NAME
  ls \- list directory contents
  .SH SYNOPSIS
  .B ls
  [\fIOPTION\fR]... [\fIFILE\fR]...
  .SH DESCRIPTION
  ...

then [\fIOPTION\fR]... [\fIFILE\fR]... displays as underlined text in an xterm
`with underlining' and with a magenta foreground in one `without underlining'.

-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: Make button face appear magenta on a Linux console

[Eli Zaretskii]

> From: Nick Roberts <nickrob@snap.net.nz>
> Date: Mon, 23 Apr 2007 08:56:04 +1200
> Cc: emacs-devel@gnu.org
> 
> If I do "man ls" in the xterm without underlining, words that are meant to be
> displayed with an underline are displayed with a magenta foreground (likewise
> bold is displayed as blue).
> 
> If a console can't do this, then presumably Emacs could by specifying the
> button face as magenta.

We lack infrastructure for such features.  What we need is a way to
form conditionals inside defface forms that look at display
capabilities, so that we could, for example, define an alternative for
the underline attribute on displays that don't support underline.
What we have now allows to form conditionals on _some_ of the
capabilities, in a patchy and ad-hoc way: we can ask about color and
grayscale support, about the number of available distinct colors, and
about the value of window-system, but that's it.

(I hope it is clear that such changes are for after the release.  As a
matter of fact, even changing the default faces should not be done
now.)

Re: Make button face appear magenta on a Linux console

[Johan Bockgård]

Eli Zaretskii <eliz@gnu.org> writes:

> What we need is a way to form conditionals inside defface forms that
> look at display capabilities, so that we could, for example, define
> an alternative for the underline attribute on displays that don't
> support underline.

You mean like (info "(elisp)Defining Faces")

    `supports'
          Whether or not the frame can display the face attributes
          given in VALUE... (*note Face Attributes::). See the
          documentation for the function
          `display-supports-face-attributes-p' for more information on
          exactly how this testing is done. *Note Display Face
          Attribute Testing::.

?


The `underline' face is defined like this in faces.el

    (defface underline
      '((((supports :underline t))
         :underline t)
        (((supports :weight bold))
         :weight bold)
        (t :underline t))
      "Basic underlined face."
      :group 'basic-faces)

And on my console the `underline' face is indeed bold.

-- 
Johan Bockgård

Re: Make button face appear magenta on a Linux console

[Eli Zaretskii]

> From: bojohan+news@dd.chalmers.se (Johan =?utf-8?Q?Bockg=C3=A5rd?=)
> Date: Mon, 23 Apr 2007 15:37:59 +0200
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> > What we need is a way to form conditionals inside defface forms that
> > look at display capabilities, so that we could, for example, define
> > an alternative for the underline attribute on displays that don't
> > support underline.
> 
> You mean like (info "(elisp)Defining Faces")
> 
>     `supports'

Sounds like it got implemented while I was sleeping.

Re: Make button face appear magenta on a Linux console

[Glenn Morris]

Nick Roberts wrote:

> OK, how about giving button face a magenta foreground directly then?
> Is it possible to specify the console and not an xterm in defface,
> i.e, not using min-colors?

I made it blue, same as ms-dos uses. Change it to magenta if you feel
passionate about it.

Re: Make button face appear magenta on a Linux console

[Nick Roberts]

 > > OK, how about giving button face a magenta foreground directly then?
 > > Is it possible to specify the console and not an xterm in defface,
 > > i.e, not using min-colors?
 > 
 > I made it blue, same as ms-dos uses. Change it to magenta if you feel
 > passionate about it.

Well, it doesn't really go with my tie... No, actually it's not quite what I
was looking for because it loses underline when available.

I've changed it to inherit from link on a tty.  This means:

1) It's the same as your change on a Linux console (cyan).
2) X based terminals with underline still display links with an underline.
3) X based terminals now generally display links in blue.  I recall that some
   people didn't like this when links were dense, but this is only on a
   terminal and their apperance on a graphical display is unchanged.

-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: Make button face appear magenta on a Linux console

[Richard Stallman]

      I see that in
    an xterm that doesn't have underlining, the button face has a magenta
    foreground (and in one that does, it keeps it's default colour - how clever is
    that!).

I suspect this cleverness is done by xterm, not by Emacs.  There is
nothing in xterm.el which would do it.  It could be that the terminfo
entry for `xterm' specifies a command for underlining, and then
`xterm' implements that as magenta when it can't implement that as
underlining.

You can probably see what's going on if you make a termscript file for
the same Emacs command sequence in both kinds of xterm.  Are they
different?