Re: The 3-faces problem

[Alejandro Colomar (man-pages)]

Hi, Branded!

On 11/2/21 19:45, G. Branden Robinson wrote:
> Hi, Alex!
>
> At 2021-11-02T19:07:05+0100, Alejandro Colomar (man-pages) wrote:
> > On 11/2/21 18:49, G. Branden Robinson wrote:
> > > printf '_\\fB_\\fI_\\f(BI_\n' | nroff | cat -s
> >
> > Yes, I see 4 different glyphs.
>
> Okay, so that much is sane in our respective environments.  :)
>
> > I was using xface termincal, but I tried on xterm after your email and
> > see the same result.  IF you have a big enough font size and print
> > them almost together, you'll be able to see the difference:
> >
> > $ cat underscore.7test
> > .TH UNDERSCORE 7TEST today
> > .SH UNDERSCORE
> > .RB [ U ] INT \fIN\fP _WI_DTH RO_MAN
> >
> > There are 3 different underscores: italics, bold, and roman,
> > respectiely.  At least with the less pager.  more doesn't even
> > highlight, so I can't test with a different pager.
> >
> > I used `man ./underscore.7`.
>
> I get _visually_ identical output[1] dumping straight to the terminal
> with "nroff -man", piping through "less -R", piping through "more", and
> running "man" on the file.

For me, there are two different outputs:

A:
        $ nroff -man ./underscore.7
        $ nroff -man ./underscore.7 | more
        $ man -P more ./underscore.7

B:
        $ nroff -man ./underscore.7 | less
        $ nroff -man ./underscore.7 | less -R
        $ man ./underscore.7

In case A, I see plain text, as if the pager was cat (no highlighting at 
all).

In case B, I see 3 different underscores, as I previously said.  It 
seems to me that the bug that was reported to Debian may responsible for 
that, since it only happens if there's a leading underscore in the word.

I'll send you a following email with a PNG of what I see.

> However, when I further pipe each of these through xxd, I get the same
> results for the first three--but "man" produces different output.

With xxd I have the same results as you.  man is different.

>  I
> suppose this has something to do with the GROFF_SGR (N.B., not
> "GROFF_NO_SGR" [see grotty(1)]) feature that man-db man added in
> response to complaints from people who were nostalgic for the wire
> protocol of Western Electric Teletypes and simultaneously harassing
> Thomas Dickey for 24-bit direct color support in ncurses so they could
> "skin" Midnight Commander.[2][3]
>
> Anyway, I think what is happening with man is that it is forcing grotty
> into overstriking mode, calling less as the pager, and less is
> reconstructing SGR escape sequences based on its heuristic
> interpretation of the overstriking sequences.  Mark Wooding did warn
> us[2] that this was theoretically possible.
>
> I can't get less(1) to reveal to me what SGR escape sequences it's
> constructing; when I try, apparently it calls isatty() on its output
> conceals them if it returns false.  But it's absolutely munging its
> input because the overstriking sequences are not present.
>
> Can you confirm that you're getting the same output I shared earlier
> from running "nroff -man" directly on the file, with no pager?  It's
> important for the troubleshooting process that we determine if our
> grotty(1)s are behaving the same.  From there we can try to determine if
> a pager or man(1) is somehow corrupting things.

See above.  If you need me to try any other commands, tell me.

> Oh, also, I'm running a pretty new less(1).  I compiled less 581.2 from
> source so I could test the new OSC 8 stuff in grotty(1)[4].  It could
> easily be behaving differently from your less(1); sid seems to have
> 551?[5]

Yes, 551.

> Regards,
> Branden
>
> [1] But like the rest of me, my eyes aren't as young as they used to be.
> [2] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=963490
> [3] https://invisible-island.net/ncurses/ncurses-slang.html
> [4] https://lists.gnu.org/archive/html/groff/2021-10/msg00000.html
> [5] http://ftp.au.debian.org/debian/pool/main/l/less/

Regards,

Alex

--
Alejandro Colomar
Linux man-pages comaintainer; https://www.kernel.org/doc/man-pages/
http://www.alejandro-colomar.es/