Re: Possible regression with `groff -C`

[G. Branden Robinson]

At 2020-04-12T12:40:38+1000, John Gardner wrote:
> > Any lurking experts have any suggestions?
> 
> From a glance, it looks like the `.C` register is being set twice: once by
> the calling tty.tmac, and again by unicode.tmac. If I comment out a few
> lines in the latter:
> 
> .\" unicode.tmac
> .\"
> .\" .nr _C \n(.C
> .cp 0
> .char - \[hy]
> .char ` \[oq]
> .char ' \[cq]
> .\" .cp \n[_C]
> 
> 
> ... it suddenly works.

Aha!  Thanks for taking that glance.  I think I see what needs to be
done here.  First, what should not be done:

1. "Just revert the commit."
2. Actually comment all this out.

Instead, I think what we need is:

A. To comment this file warning people like me off from trying to
save/restore .C in the future.  This file is effectively an internal of
tty.tmac and people should look there for its context.

B. To remove the save/restore.

C. To prefix these .char requests with .do, so they're executed with
full groff powers even in a compatibility context.

It may seem kind of weird to render a legacy document on a UTF-8
device, and there may be other barriers to doing so, but this should
not be one of them.

I'd code this up and write a regression test for it.

Thanks again.

Regards,
Branden

signature.asc
Description: PGP signature