Follow-up Comment #11, bug #64300 (group groff):
My contemplated is even less disruptive a difference from AT&T _troff_ than I
thought.
The difference I was seeing is due to AT&T's increased eagerness to start the
output, meaning to write the _trout_ header (and eventually trailer). Those
implementations routinely do so even if the input document is nothing but a
`tm` request.
But even on them, invoking any of the requests in question with the no-break
control character does *not* advance the drawing position onto the page.
Observe.
(Also, I arsed myself over to a Solaris 11 machine to check it.)
Last login: Sat Jan 3 11:02:45 2026 from 99-41-225-135.l
Oracle Corporation SunOS 5.11 11.3 June 2018
Please make sure to read /etc/SETUP for general info.
Terminal type? [xterm]
-bash-4.4$ for req in br brp ce cf fi fl in nf rj sp; do echo '***' $req;
printf '.c2 @\n.@%s\n.tm nl=\\n(nl\n' "$req" | troff >/dev/null; done
*** br
nl=-1
*** brp
nl=-1
*** ce
nl=-1
*** cf
nl=-1
*** fi
nl=-1
*** fl
nl=-1
*** in
nl=-1
*** nf
nl=-1
*** rj
nl=-1
*** sp
nl=-1
And locally, with DWB 3.3, Heirloom Doctools, and Plan 9 from User Space...
$ for req in br brp ce cf fi fl in nf rj sp; do echo '***' $req; printf '.c2
@\n.@%s\n.tm nl=\\n(nl\n' "$req" | dwb troff >/dev/null; done
*** br
nl=-1
*** brp
nl=-1
*** ce
nl=-1
*** cf
nl=-1
*** fi
nl=-1
*** fl
nl=-1
*** in
nl=-1
*** nf
nl=-1
*** rj
nl=-1
*** sp
nl=-1
$ for req in br brp ce cf fi fl in nf rj sp; do echo '***' $req; printf '.c2
@\n.@%s\n.tm nl=\\n(nl\n' "$req" | heirloom troff >/dev/null; done
*** br
nl=-1
*** brp
nl=-1
*** ce
nl=-1
*** cf
nl=-1
*** fi
nl=-1
*** fl
nl=-1
*** in
nl=-1
*** nf
nl=-1
*** rj
nl=-1
*** sp
nl=-1
$ for req in br brp ce cf fi fl in nf rj sp; do echo '***' $req; printf '.c2
@\n.@%s\n.tm nl=\\n(nl\n' "$req" | 9 troff >/dev/null; done
*** br
nl=-1
*** brp
nl=-1
*** ce
nl=-1
*** cf
nl=-1
*** fi
nl=-1
*** fl
nl=-1
*** in
nl=-1
*** nf
nl=-1
*** rj
nl=-1
*** sp
nl=-1
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?64300>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
