Follow-up Comment #15, bug #68064 (group groff): At 2026-03-05T18:19:35-0500, Dave wrote: > Follow-up Comment #14: > > [comment #13 comment #13:] >> I erred by piping nroff's output through "cat -s", > > If that is indeed an error, it's originally mine. > >> which in a POSIX shell conceals any erroneous exit status from nroff. > > But we're not testing the exit status of nroff, we're testing the > value of the systat register. In fact, all tests in this ticket have > ignored the exit status of nroff. > >> $ cat ATTIC/bad-sy.groff >> .sy exit 2 >> .tm command exited with status \n[systat] > > As comment #2 points out, using .tm obscures the bug.
Ah, okay. I am utterly stumped as to how or why it could be doing that.
But, with the following exhibit, I return to my position of comment #5;
I was unaware of this misbehavior and inadvertently fixed it with commit
24e314b94c, 12 November 2024.
$ cat ATTIC/bad-sy-2.groff
.sy exit 2
\n[systat]
$ ~/groff-1.22.3/bin/groff -aU ATTIC/bad-sy-2.groff
<beginning of page>
0
$ ~/groff-1.22.4/bin/groff -aU ATTIC/bad-sy-2.groff
<beginning of page>
0
$ ~/groff-1.23.0/bin/groff -aU ATTIC/bad-sy-2.groff
<beginning of page>
0
$ ~/groff-1.24.0/bin/groff -aU ATTIC/bad-sy-2.groff
<beginning of page>
512
Can you reproduce _this_?
If so, then I think we have enough to resolve the bug as fixed, and to
set the "planned release" field to 1.24.0.
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?68064>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
signature.asc
Description: PGP signature
