From: "Alejandro Colomar (man-pages)" <alx.manpages@gmail.com>
To: "G. Branden Robinson" <g.branden.robinson@gmail.com>
Cc: linux-man@vger.kernel.org
Subject: Re: [PATCH v2] mount_setattr.2: New manual page documenting the mount_setattr() system call
Date: Sun, 1 Aug 2021 12:38:36 +0200 [thread overview]
Message-ID: <1cde7dff-b62b-4697-1229-dd02529b3110@gmail.com> (raw)
In-Reply-To: <20210801100234.mcgwwxr42wxwe7gf@localhost.localdomain>
Hi Branden!
On 8/1/21 12:02 PM, G. Branden Robinson wrote:
> [CC list brutally trimmed]
>
> Hi, Alex!
>
> At 2021-07-31T13:20:59+0200, Alejandro Colomar (man-pages) wrote:
>> On 7/31/21 3:42 AM, G. Branden Robinson wrote:
> [...]
>>> I recommend:
>>>
>>> .BI MOUNT_ATTR_SIZE_VER number\c
>>> \&.
>>
>> I also prefer your way (at least in cases like this one (maybe in the
>> synopsis \f would be more appropriate)). It is more consistent with
>> our current style of placing each identifier in a line of its own, and
>> normal text separately (punctuation is placed wherever it's simpler,
>> but in this case I think it's simpler in a separate line).
>
> Another benefit of using the font escapes that I was reminded of today
> while doing a big revision pass on groff_man_style(7)[1] is that you get
> italic corrections for free with them. This was already true to some
> extent in groff 1.22.4, but I refactored the implementation of the
> macros earlier this year to be brutally consistent[2].
>
> I would say that I like being able to tell man page authors that they
> need not learn the italic correction escapes, but I'd probably hear from
> many of them that they had no intention of doing so in the first place.
> I might get threatened with defections to RST and Sphinx just for
> bringing it up. 🤣
What are "font escapes" and "italic correction escapes"? I don't know
those technical terms.
>
> [...]
>>> groff -man -rCHECKSTYLE=n
> [...]
>> I'll try it. Thanks!
>
> Cool!
>
>>
>>>>> +.BR open_tree (2)
>>>>> +with the
>>>>> +.I OPEN_TREE_CLONE
>>>>> +flag and it must not already have been visible in the filesystem.
>>>>> +.RE
>>>>> +.IP
>>>>
>>>> .IP here doesn't mean anything, if I'm not mistaken.
>>>
>>> It certainly _should_--it should cause the insertion of vertical
>>> space. (It would also cause a break, but .RE already did that.)
>>>
>>> The interaction of .IP and .RS/.RE is tricky and can probably be
>>> blamed, back in 2017, for irritating me to the point that I became a
>>> groff developer. I've documented it as extensively as I am able in
>>> groff_man_style(7)[1].
>>
>> Yes, indeed there are 2 consecutive blank lines, which is incorrect
>> most likely.
>
> That sounds like a bug in your man(7) renderer, and it sounds badly
> violative of the semantics of these macros in _any_ implementation.
> (You can't stack adjacent paragraph macros to make additional blank
> space; you just get the one.[3]) This stuff has been stable since 1979.
>
> I do not get _2_ blank lines after the paragraph ending "visible in the
> filesystem." Just one. None of groff 1.22.4 (Debian), groff Git HEAD,
> nor mandoc 1.14.4 misrender for me in that way. What's your tool set?
> If it's groff, I'm intensely curious to know how it's screwing this up.
> I can likely help you root-cause it.
Ahh, I used mgroff(1) (mental groff). I should debug my mental parser.
So, as I suspected, that .IP is ignored, if my mental groff is working
correctly now.
>
>> Probably a glitch of copying and pasting without really understanding
>> what each macro does (not to blame Christian, but that the
>> groff_man(7) language (or dialect actually) is not something familiar
>> to programmers, and most of them legitimately don't have time to learn
>> it well).
>
> There is a wealth of terrible examples to follow, which make the
> language seem harder than it is. A large part of my work on groff's man
> pages has been to make them good examples to follow--but as, at my last
> count, groff's ~60 man pages produce 364 pages of type on U.S. letter
> paper, this is a process that is taking some time.
>
> I acknowledge that you and Michael are wrestling an even bigger bear. :)
Yup. Since almost when I started here, I'm trying to have the man-pages
be consistent across all of the pages, both in terms of source code and
rendered output. Not an easy thing...
I hope that when I finish this (if it can ever be finished), reading and
writing man-pages is a simpler task for newbies :)
Regards,
Alex
--
Alejandro Colomar
Linux man-pages comaintainer; https://www.kernel.org/doc/man-pages/
http://www.alejandro-colomar.es/
next prev parent reply other threads:[~2021-08-01 10:38 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-30 9:41 [PATCH v2] mount_setattr.2: New manual page documenting the mount_setattr() system call Christian Brauner
2021-07-30 18:15 ` Alejandro Colomar (man-pages)
2021-07-31 1:42 ` G. Branden Robinson
2021-07-31 11:20 ` Alejandro Colomar (man-pages)
2021-08-01 10:02 ` G. Branden Robinson
2021-08-01 10:38 ` Alejandro Colomar (man-pages) [this message]
2021-08-01 11:36 ` G. Branden Robinson
2021-07-31 9:43 ` Christian Brauner
2021-07-31 12:30 ` Alejandro Colomar (man-pages)
2021-08-01 10:51 ` G. Branden Robinson
2021-08-01 11:50 ` Alejandro Colomar (man-pages)
2021-08-03 11:59 ` groff_man(7), man(7), and man-pages(7) (was: [PATCH v2] mount_setattr.2: ...) G. Branden Robinson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1cde7dff-b62b-4697-1229-dd02529b3110@gmail.com \
--to=alx.manpages@gmail.com \
--cc=g.branden.robinson@gmail.com \
--cc=linux-man@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox