Re: GNUism in groff tests, was: pic anomalies

2020-01-04 Thread Tadziu Hoffmann


>  A debugging build I made in December is about 20k lines
>  before the preprocessor runs and about 30k afterwards.
>  The formal verification code, written in Isabelle/HOL[2],
>  is about an order of magnitude greater than that.

Just out of curiosity (since I'm unfamiliar with the concepts),
what is the code complexity like?  If the verification code is
larger, is it also more complex?  If yes, can the verification
code somehow be checked against possible errors as well?





Re: How to use refer macros-agnostic way?

2020-01-04 Thread Tadziu Hoffmann

> You should run refer directly on your example file and
> examine how it's changed it.

I was about to suggest the same.

Based on this (and just for fun), I quickly hacked together
the attached macros.  They are by no means sufficient for a
real application, but they should give you some idea on how
you might go about if you want to write your own macros
instead of including one of the "refer-m*.tmac" macro files
provided with groff.

By the way, unless the bibliography file is automatically
generated and you have no influence over it, I would suggest
to get this into a consistent format first (for example, with
author first names either all abbreviated or all written out),
so you don't have to do extra processing in the macros (e.g.,
abbreviating names) to achieve consistency in the final output.

Also, on my machine I've had to use a space between the
opening brackets:

  and another to
  .[ [
  Large Imagery
  .]],

otherwise refer would glue the reference to the preceding word.
(For the same reason, the first argument of bracket-label
also has a leading space.)


.\" 
.\" start of references
.de ]<
.sp 2
.ft B
References
.ft
.sp 1
.nh
.ad l
..
.\" 
.\" end of references
.de ]>
.hy
.ad b
..
.\" 
.\" start of bibliography entry
.de ]-
.\" clear all registers in preparation for a new item
.ds [A
.ds [C
.ds [D
.ds [I
.ds [T
.ds [W
..
.\" 
.\" end of bibliography entry
.de ][
.in +2n
.ti -2n
.DT \\*([D\" extract year from date
.ref-\\$2
.in -2n
.sp .5
..
.\" 
.de DT
.ds [D \\$\\n(.$\" year is assumed to be the last argument
..
.\" 
.de ref-article-in-book
\\*([A:
.ft I
\\*([T\c
\&.
.ft
In:
.ft I
\\*([B\c
.ft
.tr -\(en
.if !'\\*([P'', p.\~\\*([P\c
.tr --
\&.
.if !'\\*([I''\\*([I,
\\*([C
(\\*([D)
..
.\" 
.de ref-book
\\*([A:
.ft I
\\*([T\c
.ft
\&.
.if !'\\*([I''\\*([I,
\\*([C
(\\*([D)
..
.\" 


Thanks, Dave Kemper for helping with the archives and bug list!

2020-01-04 Thread G. Branden Robinson
Hi folks,

I'd like to take a moment to acknowledge and thank Dave Kemper ()
for his work in scouring the archives of this list and lifting bug
reports out of them and into the Savannah bug tracker.

This gives us some stuff to look at and work on for the eventual next
release of groff, and helps us be more responsive to our user base--it's
better to address a bug report after a long time than not at all.

Dave has also been noting the ages of some of these bugs and whether
they're still present in recent git revisions.  This kind of work will
help us to write regression tests and annotate the next set of release
notes with useful information that other GNU projects are sharing.  (See
info-gnu for the recent GNU grep release announcement for an example.)

This is work that I've had vague intentions of doing but didn't get
around to.  I appreciate Dave taking care of it.

Thanks, Dave!

Regards,
Branden


signature.asc
Description: PGP signature