[9front] usb(3) patch

[9front] usb(3) patch

[igor]

The print version of usb(3) contains a verbatim `delim $$` at
the top of the page. Remove it as it looks like a mishap that
occurred during a bulk import (see `git/log /sys/man/3/usb`.

term% git/diff /sys/man/3/usb
--- //.git/fs/object/775a4bea4386c12067057de0e56dd8baa34f43ec/tree//sys/man/3/usb
+++ /sys/man/3/usb
@@ -1,7 +1,4 @@
 .TH USB 3
-.EQ
-delim $$
-.EN
 .SH NAME
 usb \- USB Host Controller Interface
 .SH SYNOPSIS

Re: [9front] usb(3) patch

[Alex Musolino]

> The print version of usb(3) contains a verbatim `delim $$` at
> the top of the page. Remove it as it looks like a mishap that
> occurred during a bulk import (see `git/log /sys/man/3/usb`.

No, this is required.  It is setting the delimiter for eqn(1).  Have a
look at /sys/man/3/usb:369.  Without setting the delimiter this part
is ignored by eqn(1) and rendered incorrectly when viewed with page
(e.g. `man -P 3 usb`).

Re: [9front] usb(3) patch

[igor]

Quoth Alex Musolino <alex@musolino.id.au>:
> > The print version of usb(3) contains a verbatim `delim $$` at
> > the top of the page. Remove it as it looks like a mishap that
> > occurred during a bulk import (see `git/log /sys/man/3/usb`.
> 
> No, this is required.  It is setting the delimiter for eqn(1).  Have a
> look at /sys/man/3/usb:369.  Without setting the delimiter this part
> is ignored by eqn(1) and rendered incorrectly when viewed with page
> (e.g. `man -P 3 usb`).

Ah, yeah. Thanks for the explanation.

I realise I did not accurately state where I saw the issue, namely
with the printed version of the manuals. There, at the top of the
page there is a verbatim `delim $$` printed, here a picture:

• https://9lab.org/img/plan9/random/usb3.man.book.jpg

Need to repro the typesetting steps of the book and see what
happened there…

[9front] [PATCH] manual: preprocess usb(3) with eqn for print output

[Anthony Martin]

igor@9lab.org reported that the eqn directive
"delim $$" was appearing at the top of the
printed manual page for usb(3). This was due
to a lack of preprocessing for this specific
file.

All other files are accounted for:

	% cd /sys/man
	% for(i in TS EQ PS G1) echo $i `{grep -l '^\.'$i ?/[0-9a-z]*}
	TS 1/tbl
	EQ 1/eqn 3/usb
	PS 1/pic
	G1 1/grap
	%

Note that 1/pic does not currently need to
be preprocessed because the examples were
inadvertently elided when removing the tpic
command. That should be corrected in a later
patch.

---
diff 065cf0b708be75e84e568261a044d66286ef6e3d fc0732aa11564503810cd76e84f267d18ee97220
--- a/sys/man/mkfile	Sun Jun 20 20:56:14 2021
+++ b/sys/man/mkfile	Mon Jun 21 01:23:43 2021
@@ -24,7 +24,7 @@
 				switch($i/$j){
 				case 1/tbl
 					tbl $j
-				case 1/eqn 3/realtime 6/auth
+				case 1/eqn 3/usb
 					eqn $j
 				case 1/pic
 					pic $j

Re: [9front] usb(3) patch

[Alex Musolino]

> I realise I did not accurately state where I saw the issue, namely
> with the printed version of the manuals. There, at the top of the
> page there is a verbatim `delim $$` printed, here a picture:
> 
> • https://9lab.org/img/plan9/random/usb3.man.book.jpg
> 
> Need to repro the typesetting steps of the book and see what
> happened there…

We're just not running eqn(1) for usb(3) in /sys/man/mkfile.  I've
pushed a couple of changes to fix it.