GNU bug report logs - #77774
stipples as fill-column indicator

Previous Next

To reply to this bug, email your comments to 77774 AT debbugs.gnu.org.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Johann Höchtl <johann.hoechtl <at> gmail.com>
To: bug-gnu-emacs <at> gnu.org
Subject: stipples as fill-column indicator
Date: Sun, 13 Apr 2025 09:09:34 +0200

[Message part 1 (text/plain, inline)]

I follow
https://emacs.stackexchange.com/questions/81278/how-to-have-a-solid-fill-column-indicator-when-line-spacing-is-greater-than-zero
to set stipples as fill-column-indicator

The result is

[image: grafik.png]

I get this result on Windows using Emacs HEAD as well as on Linux using
Emacs 30.1 pgtk-build (Arch repo)

While the screenshot has display-line-numbers-mode enabled, it is not
necessary to expose the IMHO bug.

To reproduce:

emacs -q --load fc.el fc.el

fc.el attached.

To my understanding according to
https://www.gnu.org/software/emacs/manual/html_node/elisp/Face-Attributes.html
:

> Alternatively, the value can specify the bitmap directly, with a list of
the form (width height data). Here, width and height specify the size in
pixels, and data is a string containing the raw bits of the bitmap, row by
row.

so :stipple '(7 1 " ") has a bit mask of  "0010 0000" and to my
understanding should result in a single black horizontal line. Alas it
produces two, which I consider a bug.

This, however, is not consistent. On Linux, when I add more lines to the
file so that display-line-numbers-mode would result in a shift of the
'viewport', sometimes only one line is displayed, as expected. On Windows
the behaviour remains consistent is so far as always two lines are display

[Message part 2 (text/html, inline)]

[grafik.png (image/png, inline)]

[fc.el (application/octet-stream, attachment)]

Message #8 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Johann Höchtl <johann.hoechtl <at> gmail.com>
To: 77774 <at> debbugs.gnu.org
Date: Sun, 13 Apr 2025 09:20:41 +0200

[Message part 1 (text/plain, inline)]

I did notice that when I dynamically increase the font width using
text-scale-mode, the number of lines increases even further. This may be in
line with the comment I cited

> the stipple width should be set to window-font-width

So this may not be a reliable method of setting one horizontal line anyway.

[Message part 2 (text/html, inline)]

Message #11 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Johann Höchtl <johann.hoechtl <at> gmail.com>
Cc: 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Sun, 13 Apr 2025 13:32:20 +0300

> From: Johann Höchtl <johann.hoechtl <at> gmail.com>
> Date: Sun, 13 Apr 2025 09:09:34 +0200
> 
> I follow
> https://emacs.stackexchange.com/questions/81278/how-to-have-a-solid-fill-column-indicator-when-line-spacing-is-greater-than-zero
> to set stipples as fill-column-indicator
> 
> The result is
> 
> grafik.png
> 
> I get this result on Windows using Emacs HEAD as well as on Linux using Emacs 30.1 pgtk-build (Arch repo)
> 
> While the screenshot has display-line-numbers-mode enabled, it is not necessary to expose the IMHO bug.
> 
> To reproduce:
> 
> emacs -q --load fc.el fc.el
> 
> fc.el attached.
> 
> To my understanding according to
> https://www.gnu.org/software/emacs/manual/html_node/elisp/Face-Attributes.html:
> 
> > Alternatively, the value can specify the bitmap directly, with a list of the form (width height data). Here,
> width and height specify the size in pixels, and data is a string containing the raw bits of the bitmap, row by
> row. 
> 
> so :stipple '(7 1 " ") has a bit mask of  "0010 0000" and to my understanding should result in a single black
> horizontal line. Alas it produces two, which I consider a bug.
> 
> This, however, is not consistent. On Linux, when I add more lines to the file so that display-line-numbers-mode
> would result in a shift of the 'viewport', sometimes only one line is displayed, as expected. On Windows the
> behaviour remains consistent is so far as always two lines are display 

FWIW, I cannot reproduce the problem on MS-Windows: I get a single
vertical line, as you expect.  It doesn't matter whether I turn
display-line-numbers-mode on or off, the result is consistent with
your expectations.

Message #14 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Johann Höchtl <johann.hoechtl <at> gmail.com>
Cc: 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Sun, 13 Apr 2025 13:41:22 +0300

> From: Johann Höchtl <johann.hoechtl <at> gmail.com>
> Date: Sun, 13 Apr 2025 09:20:41 +0200
> 
> I did notice that when I dynamically increase the font width using text-scale-mode, the number of lines
> increases even further. This may be in line with the comment I cited
> 
> > the stipple width should be set to window-font-width 

Then maybe this is your problem, and you should solve it accordingly?

Message #17 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Johann Höchtl <johann.hoechtl <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Mon, 14 Apr 2025 10:08:14 +0200

[Message part 1 (text/plain, inline)]

I do still believe it is a bug.

I changed the definition of to

(set-face-attribute 'fill-column-indicator nil :background nil 
:foreground "white" :stipple '(10 1 " \x0\x0\x0\x0\x0\x0"))

one blank representing a single set bit, followed by quite a few null 
string literals as I do not know what happens if I dynamically change 
font sizes. Does this influence the necessity to change the bitmap 
literal of a stipple?

Nonetheless, at Linux at scale level 0 a single line is shown, at some 
increases of the scale level  I see two or more lines, at higher scale 
levels just one again.

Also, display-line-numbers-mode influences the position of the stipples. 
Which IMHO shouldn't be the case.




On 4/13/25 12:41, Eli Zaretskii wrote:
>> From: Johann Höchtl <johann.hoechtl <at> gmail.com>
>> Date: Sun, 13 Apr 2025 09:20:41 +0200
>>
>> I did notice that when I dynamically increase the font width using text-scale-mode, the number of lines
>> increases even further. This may be in line with the comment I cited
>>
>>> the stipple width should be set to window-font-width
> Then maybe this is your problem, and you should solve it accordingly?

[nodisplaylinenumbers.png (image/png, attachment)]

[withdisplaylinenumbers.png (image/png, attachment)]

Message #20 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Johann Höchtl <johann.hoechtl <at> gmail.com>,
 Po Lu <luangruo <at> yahoo.com>
Cc: 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Mon, 14 Apr 2025 11:55:15 +0300

> Date: Mon, 14 Apr 2025 10:08:14 +0200
> Cc: 77774 <at> debbugs.gnu.org
> From: Johann Höchtl <johann.hoechtl <at> gmail.com>
> 
> I do still believe it is a bug.

Please explain why.

> I changed the definition of to
> 
> (set-face-attribute 'fill-column-indicator nil :background nil 
> :foreground "white" :stipple '(10 1 " \x0\x0\x0\x0\x0\x0"))
> 
> one blank representing a single set bit, followed by quite a few null 
> string literals as I do not know what happens if I dynamically change 
> font sizes. Does this influence the necessity to change the bitmap 
> literal of a stipple?

I don't think stipple bitmaps are automatically extended as you change
the text scale.  Po Lu, is that supposed to work automagically
somehow?

> Nonetheless, at Linux at scale level 0 a single line is shown, at some 
> increases of the scale level  I see two or more lines, at higher scale 
> levels just one again.

What happens if you enlarge the bitmap even more?

> Also, display-line-numbers-mode influences the position of the stipples. 
> Which IMHO shouldn't be the case.

True.  But that is a different problem.

Message #23 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Johann Höchtl <johann.hoechtl <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: Po Lu <luangruo <at> yahoo.com>, 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Tue, 22 Apr 2025 17:42:16 +0200

[Message part 1 (text/plain, inline)]

Am Mo., 14. Apr. 2025 um 10:55 Uhr schrieb Eli Zaretskii <eliz <at> gnu.org>:

> > Date: Mon, 14 Apr 2025 10:08:14 +0200
> > Cc: 77774 <at> debbugs.gnu.org
> > From: Johann Höchtl <johann.hoechtl <at> gmail.com>
> >
> > I do still believe it is a bug.
>
> Please explain why.
>
> > I changed the definition of to
> >
> > (set-face-attribute 'fill-column-indicator nil :background nil
> > :foreground "white" :stipple '(10 1 " \x0\x0\x0\x0\x0\x0"))
> >
> > one blank representing a single set bit, followed by quite a few null
> > string literals as I do not know what happens if I dynamically change
> > font sizes. Does this influence the necessity to change the bitmap
> > literal of a stipple?
>
> I don't think stipple bitmaps are automatically extended as you change
> the text scale.  Po Lu, is that supposed to work automagically
> somehow?
>
> > Nonetheless, at Linux at scale level 0 a single line is shown, at some
> > increases of the scale level  I see two or more lines, at higher scale
> > levels just one again.
>
> What happens if you enlarge the bitmap even more?
>

I did and made an "absurdly" large  1 by <large> bitmap. It consistently
becomes two lines at scale level +5.

For the sake of simplicity and ruling out other factors I set display scale
level to 100%.

When starting at OS scale level 150%, I see already two stipple lines at
emacs scale mode +0 (no scale)

So there might be a common culprit to the misbehaviour.


> > Also, display-line-numbers-mode influences the position of the stipples.
> > Which IMHO shouldn't be the case.
>
> True.  But that is a different problem.
>

[Message part 2 (text/html, inline)]

Message #26 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Po Lu <luangruo <at> yahoo.com>
To: Johann Höchtl <johann.hoechtl <at> gmail.com>
Cc: Eli Zaretskii <eliz <at> gnu.org>, 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Wed, 23 Apr 2025 09:47:57 +0800

Johann Höchtl <johann.hoechtl <at> gmail.com> writes:

>  I don't think stipple bitmaps are automatically extended as you change
>  the text scale.  Po Lu, is that supposed to work automagically
>  somehow?

No, they're not.  Stipples are not designed for displaying lines or bars
after all, just repeating patterns that produce something of a dithering
effect.

Message #29 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Johann Höchtl <johann.hoechtl <at> gmail.com>
To: Po Lu <luangruo <at> yahoo.com>
Cc: Eli Zaretskii <eliz <at> gnu.org>, 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Wed, 23 Apr 2025 18:13:30 +0200

[Message part 1 (text/plain, inline)]

Am Mi., 23. Apr. 2025 um 03:48 Uhr schrieb Po Lu <luangruo <at> yahoo.com>:

> Johann Höchtl <johann.hoechtl <at> gmail.com> writes:
>
> >  I don't think stipple bitmaps are automatically extended as you change
> >  the text scale.  Po Lu, is that supposed to work automagically
> >  somehow?
>
> No, they're not.  Stipples are not designed for displaying lines or bars
> after all, just repeating patterns that produce something of a dithering
> effect.
>
> I use indent-bars mode https://github.com/jdtsmith/indent-bars to a great
extend. Indent bars-mode uses stipples to render straight lines for ...
indent-bars.

This package manages to display straight lines irrespective of operating
system screen scale value or Emacs scale-mode value other than 0.

As long as indent-bars mode doesn't use other drawing primitives than what
(display-fill-column-indicator-mode) is using, I think the stipple
behaviour of (display-fill-column-indicator-mode) contains a bug.

[Message part 2 (text/html, inline)]

Message #32 received at 77774 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Johann Höchtl <johann.hoechtl <at> gmail.com>
Cc: luangruo <at> yahoo.com, 77774 <at> debbugs.gnu.org
Subject: Re: bug#77774: stipples as fill-column indicator
Date: Thu, 24 Apr 2025 08:17:35 +0300

severity 77774 wishlist
thanks

> From: Johann Höchtl <johann.hoechtl <at> gmail.com>
> Date: Wed, 23 Apr 2025 18:13:30 +0200
> Cc: Eli Zaretskii <eliz <at> gnu.org>, 77774 <at> debbugs.gnu.org
> 
> Am Mi., 23. Apr. 2025 um 03:48 Uhr schrieb Po Lu <luangruo <at> yahoo.com>:
> 
>  Johann Höchtl <johann.hoechtl <at> gmail.com> writes:
> 
>  >  I don't think stipple bitmaps are automatically extended as you change
>  >  the text scale.  Po Lu, is that supposed to work automagically
>  >  somehow?
> 
>  No, they're not.  Stipples are not designed for displaying lines or bars
>  after all, just repeating patterns that produce something of a dithering
>  effect.
> 
> I use indent-bars mode https://github.com/jdtsmith/indent-bars to a great extend. Indent bars-mode uses
> stipples to render straight lines for ... indent-bars.
> 
> This package manages to display straight lines irrespective of operating system screen scale value or
> Emacs scale-mode value other than 0.
> 
> As long as indent-bars mode doesn't use other drawing primitives than what
> (display-fill-column-indicator-mode) is using, I think the stipple behaviour of
> (display-fill-column-indicator-mode) contains a bug.

AFAIU, it isn't a bug, it's perhaps a missing feature.  To have
stipples scale correctly with text-scale, someone has to write code to
scale the bit patterns used for the stipples.  Patches to that effect
will be welcome.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.