Gnus development mailing list
 help / color / mirror / Atom feed
* Recent change on Emacs master: aggressive word-wrap of header lines in article display
@ 2021-12-11 16:32 Adam Sjøgren
  2021-12-11 16:58 ` Russ Allbery
                   ` (2 more replies)
  0 siblings, 3 replies; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 16:32 UTC (permalink / raw)
  To: ding

[-- Attachment #1: Type: text/plain, Size: 122 bytes --]

A recent change seems to have changed the display of long header lines
to word wrap waaaay before the end of the window:


[-- Attachment #2: word-wrap.png --]
[-- Type: image/png, Size: 63825 bytes --]

[-- Attachment #3: Type: text/plain, Size: 336 bytes --]


Was that intentional and/or can it be tuned?

This is with Emacs from master@8f3f785c55 and 8c50016b10 (updated right
now).


  Best regards,

    Adam

-- 
 "Witness the most amazing accomplishment of                Adam Sjøgren
  NASA -- managing to turn the exploration             asjo@koldfront.dk
  of space into a huge snore."

^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 16:32 Recent change on Emacs master: aggressive word-wrap of header lines in article display Adam Sjøgren
@ 2021-12-11 16:58 ` Russ Allbery
  2021-12-11 17:41   ` Andreas Schwab
  2021-12-11 17:57   ` Adam Sjøgren
  2021-12-11 17:31 ` Eric Abrahamsen
  2021-12-11 19:57 ` Adam Sjøgren
  2 siblings, 2 replies; 21+ messages in thread
From: Russ Allbery @ 2021-12-11 16:58 UTC (permalink / raw)
  To: ding

Adam Sjøgren <asjo@koldfront.dk> writes:

> A recent change seems to have changed the display of long header lines
> to word wrap waaaay before the end of the window:

I've noticed this happening for a while, and when I've investigated, it's
been because the header in the original is RFC-2047-encoded and the
wrapping makes sense for the encoded text but not the decoded text.  My
theory was that the header was wrapped before instead of after decoding.

If you check with C-u g, does that fit the behavior you're seeing?

If so, that does seem like a bug, and like it would be better to decode
the header first and then wrap it.

-- 
Russ Allbery (eagle@eyrie.org)             <https://www.eyrie.org/~eagle/>


^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 16:32 Recent change on Emacs master: aggressive word-wrap of header lines in article display Adam Sjøgren
  2021-12-11 16:58 ` Russ Allbery
@ 2021-12-11 17:31 ` Eric Abrahamsen
  2021-12-11 18:01   ` Adam Sjøgren
  2021-12-11 19:57 ` Adam Sjøgren
  2 siblings, 1 reply; 21+ messages in thread
From: Eric Abrahamsen @ 2021-12-11 17:31 UTC (permalink / raw)
  To: ding

Adam Sjøgren <asjo@koldfront.dk> writes:

> A recent change seems to have changed the display of long header lines
> to word wrap waaaay before the end of the window:
>
>
>
>
> Was that intentional and/or can it be tuned?
>
> This is with Emacs from master@8f3f785c55 and 8c50016b10 (updated right
> now).

Does changing `gnus-article-normalized-header-length' change that wrap?
Though that option hasn't changed anytime recently.

For my part, I'm going to figure out how to turn off variable-width
fonts in the headers, that is fugly...



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 16:58 ` Russ Allbery
@ 2021-12-11 17:41   ` Andreas Schwab
  2021-12-11 17:57     ` Adam Sjøgren
  2021-12-11 18:49     ` Russ Allbery
  2021-12-11 17:57   ` Adam Sjøgren
  1 sibling, 2 replies; 21+ messages in thread
From: Andreas Schwab @ 2021-12-11 17:41 UTC (permalink / raw)
  To: Russ Allbery; +Cc: ding

On Dez 11 2021, Russ Allbery wrote:

> If so, that does seem like a bug, and like it would be better to decode
> the header first and then wrap it.

I don't think Gnus does any wrapping on its own, it just keeps the line
breaks of the original header.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510  2552 DF73 E780 A9DA AEC1
"And now for something completely different."


^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 16:58 ` Russ Allbery
  2021-12-11 17:41   ` Andreas Schwab
@ 2021-12-11 17:57   ` Adam Sjøgren
  1 sibling, 0 replies; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 17:57 UTC (permalink / raw)
  To: ding

Russ writes:

> Adam Sjøgren <asjo@koldfront.dk> writes:
>
>> A recent change seems to have changed the display of long header lines
>> to word wrap waaaay before the end of the window:
>
> I've noticed this happening for a while, and when I've investigated, it's
> been because the header in the original is RFC-2047-encoded and the
> wrapping makes sense for the encoded text but not the decoded text.  My
> theory was that the header was wrapped before instead of after decoding.
>
> If you check with C-u g, does that fit the behavior you're seeing?

Not quite, here is the raw header:

    Subject: =?UTF-8?Q?Susanne_Bier_modtager_i_aften_k=C3=A6mpe_filmpris=3A_=27Jeg_tro?=
     =?UTF-8?Q?ede=2C_det_var_en_fejl=27?=

The linefeed in the header would result in:

    Susanne Bier modtager i aften kæmpe filmpris: 'Jeg tro
    ede, det var en fejl'

But what is shown is:

    Susanne Bier modtager i aften kæmpe filmpris: 'Jeg troede, det var
    en fejl'

I'm suspecting it might be side-effect related to the change to default
to variable width font, but I haven't bisected it [yet].


  Best regards,

    Adam

-- 
 "I need someone to believe in me                           Adam Sjøgren
  I need someone to throw blue sky at me"              asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 17:41   ` Andreas Schwab
@ 2021-12-11 17:57     ` Adam Sjøgren
  2021-12-11 19:27       ` Emanuel Berg
  2021-12-11 18:49     ` Russ Allbery
  1 sibling, 1 reply; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 17:57 UTC (permalink / raw)
  To: ding

Andreas writes:

> On Dez 11 2021, Russ Allbery wrote:
>
>> If so, that does seem like a bug, and like it would be better to decode
>> the header first and then wrap it.
>
> I don't think Gnus does any wrapping on its own, it just keeps the line
> breaks of the original header.

Wrong.


-- 
 "I'm always interested in avoiding learning new            Adam Sjøgren
  things."                                             asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 17:31 ` Eric Abrahamsen
@ 2021-12-11 18:01   ` Adam Sjøgren
  2021-12-11 18:45     ` Jose A. Ortega Ruiz
  0 siblings, 1 reply; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 18:01 UTC (permalink / raw)
  To: ding

Eric writes:

> Does changing `gnus-article-normalized-header-length' change that wrap?

The default value seems to be 40. I tried setting it to 200. The article
is rendered the same.

> For my part, I'm going to figure out how to turn off variable-width
> fonts in the headers, that is fugly...

I've been using that for years; I like it - different strokes for
different folks, I guess :-)


  Best regards,

    Adam

-- 
 "Soon we'll have spent a whole month at sea,               Adam Sjøgren
  splitting atoms for no apparent reason"              asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 18:01   ` Adam Sjøgren
@ 2021-12-11 18:45     ` Jose A. Ortega Ruiz
  2021-12-11 20:01       ` Adam Sjøgren
  0 siblings, 1 reply; 21+ messages in thread
From: Jose A. Ortega Ruiz @ 2021-12-11 18:45 UTC (permalink / raw)
  To: ding

On Sat, Dec 11 2021, Adam Sjøgren wrote:

> Eric writes:
>
>> Does changing `gnus-article-normalized-header-length' change that wrap?
>
> The default value seems to be 40. I tried setting it to 200. The article
> is rendered the same.
>
>> For my part, I'm going to figure out how to turn off variable-width
>> fonts in the headers, that is fugly...
>
> I've been using that for years; I like it - different strokes for
> different folks, I guess :-)

I noticed it a week or so ago and managed to fix it with:

   (setq gnus-treat-fold-headers nil)

I suspect its default value (or how it is used) was changed with the
move to variable pitch (which i totally side-step everywhere by
customizing variable-pitch and another new one if forgot to a fixed font
family (Hack in my case)).

Hope this helps,
jao
-- 
If languages had true garbage collection, most programs would
delete themselves upon execution.
    - Robert Sewell, paraphrased



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 17:41   ` Andreas Schwab
  2021-12-11 17:57     ` Adam Sjøgren
@ 2021-12-11 18:49     ` Russ Allbery
  1 sibling, 0 replies; 21+ messages in thread
From: Russ Allbery @ 2021-12-11 18:49 UTC (permalink / raw)
  To: ding

Andreas Schwab <schwab@linux-m68k.org> writes:
> On Dez 11 2021, Russ Allbery wrote:

>> If so, that does seem like a bug, and like it would be better to decode
>> the header first and then wrap it.

> I don't think Gnus does any wrapping on its own, it just keeps the line
> breaks of the original header.

Oh!  Yes, that makes sense, and that would explain it.

-- 
Russ Allbery (eagle@eyrie.org)             <https://www.eyrie.org/~eagle/>


^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 17:57     ` Adam Sjøgren
@ 2021-12-11 19:27       ` Emanuel Berg
  0 siblings, 0 replies; 21+ messages in thread
From: Emanuel Berg @ 2021-12-11 19:27 UTC (permalink / raw)
  To: ding

Adam Sjøgren wrote:

>>> If so, that does seem like a bug, and like it would be
>>> better to decode the header first and then wrap it.
>>
>> I don't think Gnus does any wrapping on its own, it just
>> keeps the line breaks of the original header.
>
> Wrong.

For the subject that was in the message I know reply to, in
`gnus-article-mode' that is displayed as

  Subject: Re: Recent change on Emacs master: aggressive word-wrap of header line
  s in article display

with a backslash where the line break happens.

In `message-mode', where I'm now, the line isn't broken

  Subject: Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display

I must admit I never gave any case any thought until reading
this thread, but FWIW I would think it optimal to have it
displayed broken in both cases, only make the break happen at
a natural place, e.g. a whitespace.

-- 
underground experts united
https://dataswamp.org/~incal



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 16:32 Recent change on Emacs master: aggressive word-wrap of header lines in article display Adam Sjøgren
  2021-12-11 16:58 ` Russ Allbery
  2021-12-11 17:31 ` Eric Abrahamsen
@ 2021-12-11 19:57 ` Adam Sjøgren
  2021-12-11 20:12   ` Adam Sjøgren
  2 siblings, 1 reply; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 19:57 UTC (permalink / raw)
  To: ding

Here is the result of bisecting the new line wrapping of headers:

    3bc0768e9b82feaf5ce1665227055e97e72af164 is the first bad commit
    commit 3bc0768e9b82feaf5ce1665227055e97e72af164
    Author: Lars Ingebrigtsen <larsi@gnus.org>
    Date:   Wed Dec 1 14:10:19 2021 +0100

        Enable gnus-treat-fold-headers by default

        * lisp/gnus/gnus-art.el (gnus-treat-fold-headers): Change default
        to `head'.

     etc/NEWS              | 3 +++
     lisp/gnus/gnus-art.el | 4 ++--
     2 files changed, 5 insertions(+), 2 deletions(-)


  Best regards,

    Adam

-- 
 "Clear? Huh! Why a four-year-old child could               Adam Sjøgren
  understand this report. Run out and find me a        asjo@koldfront.dk
  four-year-old child. I can't make head or tail out
  of it."



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 18:45     ` Jose A. Ortega Ruiz
@ 2021-12-11 20:01       ` Adam Sjøgren
  0 siblings, 0 replies; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 20:01 UTC (permalink / raw)
  To: ding

Jose A. writes:

> I noticed it a week or so ago and managed to fix it with:
>
>    (setq gnus-treat-fold-headers nil)

Indeed, the commit that changed this (3bc0768e9b) changes the default
value to 'head from nil.


  Best regards,

    Adam

-- 
 "I refuse to be victimized by notions of virtuous          Adam Sjøgren
  behaviour"                                           asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 19:57 ` Adam Sjøgren
@ 2021-12-11 20:12   ` Adam Sjøgren
  2021-12-11 20:23     ` Adam Sjøgren
  0 siblings, 1 reply; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 20:12 UTC (permalink / raw)
  To: ding

Adam writes:

> Here is the result of bisecting the new line wrapping of headers:

>         Enable gnus-treat-fold-headers by default

This following commit is supposed to change gnus-treat-fold-headers to
fold variable pitch fonts at the window width, it looks like:

    commit 2d60566c8787b63b5e4e2958e9ec9f55fbcf9e87
    Author: Lars Ingebrigtsen <larsi@gnus.org>
    Date:   Wed Dec 1 06:30:32 2021 +0100

        Use proportional fonts in the Gnus headers by default

        * lisp/gnus/gnus-art.el (gnus-header): Inherit from `variable-pitch'.
        (gnus--variable-pitch-p): New helper function.
        (gnus-article-treat-fold-headers): Fill using pixel filling.

I.e. the code here:

    (defun gnus-article-treat-fold-headers ()
      ...
            (if (not (gnus--variable-pitch-p (get-text-property (point) 'face)))
                (mail-header-fold-field)
              (forward-char 1)
              (pixel-fill-region (point) (point-max) (pixel-fill-width)))
      ...

But it doesn't on my screen, as seen in the screenshot at the top of
this thread.

I should step through that and see what actually happens.


  Best regards,

    Adam

-- 
 "I refuse to be victimized by notions of virtuous          Adam Sjøgren
  behaviour"                                           asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 20:12   ` Adam Sjøgren
@ 2021-12-11 20:23     ` Adam Sjøgren
  2021-12-11 20:29       ` Adam Sjøgren
  0 siblings, 1 reply; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 20:23 UTC (permalink / raw)
  To: ding

Adam writes:

>     (defun gnus-article-treat-fold-headers ()
>       ...
>             (if (not (gnus--variable-pitch-p (get-text-property (point) 'face)))
>                 (mail-header-fold-field)
>               (forward-char 1)
>               (pixel-fill-region (point) (point-max) (pixel-fill-width)))

> I should step through that and see what actually happens.

gnus--variable-pitch-p returns nil on gnus-header-name, so
mail-header-fold-field gets called!

M-x describe-face on gnus-header-name gives this:

    Face: gnus-header-name (sample) (customize this face)

    Documentation:
    Face used for displaying header names.

    Defined in ‘gnus-art.el’.

               Family: DejaVu Sans Condensed
              Foundry: unspecified
                Width: unspecified
               Height: unspecified
               Weight: unspecified
                Slant: unspecified
           Foreground: maroon
    DistantForeground: unspecified
           Background: unspecified
            Underline: unspecified
             Overline: unspecified
       Strike-through: unspecified
                  Box: unspecified
              Inverse: unspecified
              Stipple: unspecified
                 Font: unspecified
              Fontset: unspecified
               Extend: unspecified
              Inherit: unspecified

Hm, why doesn't Emacs think DejaVu Sans Condensed is variable pitch?

Maybe I should step through gnus--variable-pitch-p as well.

  (gnus--variable-pitch-p 'gnus-header-name) → nil

Ok, so the reason is that the definition of a variable pitch face is
whether it is, or inherits from, the face variable-pitch.

I just used customize to set the font of gnus-header-name.


  Best regards,

    Adam

-- 
 "Så ruller domino-effekten.                                Adam Sjøgren
  Du kender det selv!"                                 asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 20:23     ` Adam Sjøgren
@ 2021-12-11 20:29       ` Adam Sjøgren
  2021-12-11 20:40         ` Eric Abrahamsen
  0 siblings, 1 reply; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 20:29 UTC (permalink / raw)
  To: ding

[-- Attachment #1: Type: text/plain, Size: 528 bytes --]

Adam writes:

> Ok, so the reason is that the definition of a variable pitch face is
> whether it is, or inherits from, the face variable-pitch.
>
> I just used customize to set the font of gnus-header-name.

So my fix is to change this line in the custom-set-faces call in my
~/.emacs.d/init.el:

     '(gnus-header-name ((t (:foreground "maroon" :family "DejaVu Sans Condensed"))))

to:

     '(gnus-header-name ((t (:foreground "maroon" :inherit variable-pitch :family "DejaVu Sans Condensed"))))

And Bob's my uncle again:


[-- Attachment #2: fixed.png --]
[-- Type: image/png, Size: 48496 bytes --]

[-- Attachment #3: Type: text/plain, Size: 193 bytes --]



  Best regards,

    Adam

-- 
 "This is either madness... or brilliance."                 Adam Sjøgren
 "It's remarkable how often those two traits           asjo@koldfront.dk
  coincide."

^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 20:29       ` Adam Sjøgren
@ 2021-12-11 20:40         ` Eric Abrahamsen
  2021-12-11 20:51           ` Adam Sjøgren
  2021-12-11 23:21           ` Stephen Berman
  0 siblings, 2 replies; 21+ messages in thread
From: Eric Abrahamsen @ 2021-12-11 20:40 UTC (permalink / raw)
  To: ding

Adam Sjøgren <asjo@koldfront.dk> writes:

> Adam writes:
>
>> Ok, so the reason is that the definition of a variable pitch face is
>> whether it is, or inherits from, the face variable-pitch.
>>
>> I just used customize to set the font of gnus-header-name.
>
> So my fix is to change this line in the custom-set-faces call in my
> ~/.emacs.d/init.el:
>
>      '(gnus-header-name ((t (:foreground "maroon" :family "DejaVu Sans Condensed"))))
>
> to:
>
>      '(gnus-header-name ((t (:foreground "maroon" :inherit variable-pitch :family "DejaVu Sans Condensed"))))
>
> And Bob's my uncle again:

I didn't quite follow Bob's progress here, but do you think there's a
bug to report? It seems weird that header wrapping would depend on
variable-pitch or not...

I realized I mostly don't like the variable pitch because it seems to
display about 20% larger than the fixed-width fonts, which looks
terrible (I've seen people complain about this elsewhere, can't remember
what the issue was).



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 20:40         ` Eric Abrahamsen
@ 2021-12-11 20:51           ` Adam Sjøgren
  2021-12-11 20:54             ` Eric Abrahamsen
  2021-12-11 23:21           ` Stephen Berman
  1 sibling, 1 reply; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 20:51 UTC (permalink / raw)
  To: ding

Eric writes:

> I didn't quite follow Bob's progress here, but do you think there's a
> bug to report?

Not really - if anything should be improved it's how Gnus detects that a
face is variable pitch.

I was assuming it was looking at the font, and it isn't.

> It seems weird that header wrapping would depend on variable-pitch or
> not...

It uses the new pixel-fill-region stuff to wrap at the window width. It
makes sense, I think? It's better than at a fixed number of characters,
which looks haphazard for a variable pitch font.

> I realized I mostly don't like the variable pitch because it seems to
> display about 20% larger than the fixed-width fonts, which looks
> terrible (I've seen people complain about this elsewhere, can't remember
> what the issue was).

I guess I have just configured/customized the sizes to match at some
point.

When Lars switched the modeline to variable pitch I switched my
echo-area/minibuffer to variable pitch as well; I think it looks nice
(but it was a bit of a faff to configure:

    ; echo-area face:
    (defface echo-area
      '((t (:family "DejaVu Sans Condensed")))
      "Face for the echo-area")

    ;;; And in custom-set-faces:
    ;;;  '(minibuffer-prompt ((t (:inherit echo-area :foreground "medium blue"))))

    ; This to get variable width font in the line below the modeline:
    (dolist
        (buf (list " *Minibuf-0*" " *Minibuf-1*" " *Echo Area 0*" " *Echo Area 1*"))
      (when (get-buffer buf)
        (with-current-buffer buf
          (setq-local face-remapping-alist '((default (:inherit echo-area)))))))

    ; This as well - and minibuffer-prompt face above!
    (defun my-minibuffer-setup-hook ()
      (buffer-face-set 'echo-area))
    (add-hook 'minibuffer-setup-hook 'my-minibuffer-setup-hook)

gathered from searches/various places :-))


  Best regards,

    Adam

-- 
 "My feet are so tired, my brain is so wired                Adam Sjøgren
  And the clouds are weeping"                          asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 20:51           ` Adam Sjøgren
@ 2021-12-11 20:54             ` Eric Abrahamsen
  2021-12-11 21:04               ` Adam Sjøgren
  0 siblings, 1 reply; 21+ messages in thread
From: Eric Abrahamsen @ 2021-12-11 20:54 UTC (permalink / raw)
  To: ding

Adam Sjøgren <asjo@koldfront.dk> writes:

> Eric writes:
>
>> I didn't quite follow Bob's progress here, but do you think there's a
>> bug to report?
>
> Not really - if anything should be improved it's how Gnus detects that a
> face is variable pitch.
>
> I was assuming it was looking at the font, and it isn't.
>
>> It seems weird that header wrapping would depend on variable-pitch or
>> not...
>
> It uses the new pixel-fill-region stuff to wrap at the window width. It
> makes sense, I think? It's better than at a fixed number of characters,
> which looks haphazard for a variable pitch font.

Okay, cool.

>> I realized I mostly don't like the variable pitch because it seems to
>> display about 20% larger than the fixed-width fonts, which looks
>> terrible (I've seen people complain about this elsewhere, can't remember
>> what the issue was).
>
> I guess I have just configured/customized the sizes to match at some
> point.
>
> When Lars switched the modeline to variable pitch I switched my
> echo-area/minibuffer to variable pitch as well; I think it looks nice
> (but it was a bit of a faff to configure:
>
>     ; echo-area face:
>     (defface echo-area
>       '((t (:family "DejaVu Sans Condensed")))
>       "Face for the echo-area")
>
>     ;;; And in custom-set-faces:
>     ;;;  '(minibuffer-prompt ((t (:inherit echo-area :foreground "medium blue"))))
>
>     ; This to get variable width font in the line below the modeline:
>     (dolist
>         (buf (list " *Minibuf-0*" " *Minibuf-1*" " *Echo Area 0*" " *Echo Area 1*"))
>       (when (get-buffer buf)
>         (with-current-buffer buf
>           (setq-local face-remapping-alist '((default (:inherit echo-area)))))))
>
>     ; This as well - and minibuffer-prompt face above!
>     (defun my-minibuffer-setup-hook ()
>       (buffer-face-set 'echo-area))
>     (add-hook 'minibuffer-setup-hook 'my-minibuffer-setup-hook)
>
> gathered from searches/various places :-))

I've tried so hard to pretend that face configuration doesn't exist in
Emacs, and that's worked out perfectly until this change! I guess that's
the only reason I'm grumpy. I don't want to do all that config.



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 20:54             ` Eric Abrahamsen
@ 2021-12-11 21:04               ` Adam Sjøgren
  0 siblings, 0 replies; 21+ messages in thread
From: Adam Sjøgren @ 2021-12-11 21:04 UTC (permalink / raw)
  To: ding

Eric writes:

> I've tried so hard to pretend that face configuration doesn't exist in
> Emacs, and that's worked out perfectly until this change! I guess that's
> the only reason I'm grumpy. I don't want to do all that config.

I only venture there when it gets too annoying, so once a decade or so.


  Best regards,

    Adam

-- 
 "I said to Hank Williams: How lonely does it get?          Adam Sjøgren
  Hank Williams hasn't answered yet"                   asjo@koldfront.dk



^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 20:40         ` Eric Abrahamsen
  2021-12-11 20:51           ` Adam Sjøgren
@ 2021-12-11 23:21           ` Stephen Berman
  2021-12-12  0:15             ` Eric Abrahamsen
  1 sibling, 1 reply; 21+ messages in thread
From: Stephen Berman @ 2021-12-11 23:21 UTC (permalink / raw)
  To: Eric Abrahamsen; +Cc: ding

On Sat, 11 Dec 2021 12:40:03 -0800 Eric Abrahamsen <eric@ericabrahamsen.net> wrote:

> I realized I mostly don't like the variable pitch because it seems to
> display about 20% larger than the fixed-width fonts, which looks
> terrible (I've seen people complain about this elsewhere, can't remember
> what the issue was).

Are you referring to the change in the Gnus header faces?  This is due
to making them inherit from the new variable-pitch-text face, which
inherits from variable-pitch face but scales the height by 1.1.  I
didn't like that either and after customizing it to removing the
scaling, it looks fine to me.  (This is with DejaVu Sans for variable
pitch and DejaVu Sans Mono for fixed pitch.)

Steve Berman


^ permalink raw reply	[flat|nested] 21+ messages in thread

* Re: Recent change on Emacs master: aggressive word-wrap of header lines in article display
  2021-12-11 23:21           ` Stephen Berman
@ 2021-12-12  0:15             ` Eric Abrahamsen
  0 siblings, 0 replies; 21+ messages in thread
From: Eric Abrahamsen @ 2021-12-12  0:15 UTC (permalink / raw)
  To: ding

Stephen Berman <stephen.berman@gmx.net> writes:

> On Sat, 11 Dec 2021 12:40:03 -0800 Eric Abrahamsen
> <eric@ericabrahamsen.net> wrote:
>
>> I realized I mostly don't like the variable pitch because it seems
>> to
>> display about 20% larger than the fixed-width fonts, which looks
>> terrible (I've seen people complain about this elsewhere, can't
>> remember
>> what the issue was).
>
> Are you referring to the change in the Gnus header faces? This is due
> to making them inherit from the new variable-pitch-text face, which
> inherits from variable-pitch face but scales the height by 1.1.  I
> didn't like that either and after customizing it to removing the
> scaling, it looks fine to me. (This is with DejaVu Sans for variable
> pitch and DejaVu Sans Mono for fixed pitch.)

Oh, that's better! Thanks for that suggestion, maybe I'll keep this.



^ permalink raw reply	[flat|nested] 21+ messages in thread

end of thread, other threads:[~2021-12-12  0:16 UTC | newest]

Thread overview: 21+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-11 16:32 Recent change on Emacs master: aggressive word-wrap of header lines in article display Adam Sjøgren
2021-12-11 16:58 ` Russ Allbery
2021-12-11 17:41   ` Andreas Schwab
2021-12-11 17:57     ` Adam Sjøgren
2021-12-11 19:27       ` Emanuel Berg
2021-12-11 18:49     ` Russ Allbery
2021-12-11 17:57   ` Adam Sjøgren
2021-12-11 17:31 ` Eric Abrahamsen
2021-12-11 18:01   ` Adam Sjøgren
2021-12-11 18:45     ` Jose A. Ortega Ruiz
2021-12-11 20:01       ` Adam Sjøgren
2021-12-11 19:57 ` Adam Sjøgren
2021-12-11 20:12   ` Adam Sjøgren
2021-12-11 20:23     ` Adam Sjøgren
2021-12-11 20:29       ` Adam Sjøgren
2021-12-11 20:40         ` Eric Abrahamsen
2021-12-11 20:51           ` Adam Sjøgren
2021-12-11 20:54             ` Eric Abrahamsen
2021-12-11 21:04               ` Adam Sjøgren
2021-12-11 23:21           ` Stephen Berman
2021-12-12  0:15             ` Eric Abrahamsen

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).