> > . nr env-count 0 1
> > .
> > . de CI
> > . ev e-\\n+[env-count]
> > . nop \&\\$*
> > . nr have-int \\n[.int]
> > . ev
> > . fam C
> > . ie \\n[have-int] \{\
> > . I \&\\$*
> > . fam
> > . \}
> > . el \
> > . I \&\\$*\F[]
> > ..
>
> This seems to screw up for very long arguments. I guess you need to
> evaluate the nop inside a diversion, otherwise if the argument
> results in more than one output line only the last remaining stuff
> remains in the current partial line of the environment, the
> completed lines get printed.
Yep.
> But how about this:
>
> .de CI
> .ds xx \\n[.fam]
> .fam C
> .I [EMAIL PROTECTED]
> .fam
> ..
Excellent! This is *very* nice. BTW, attaching something directly to
\$@ doesn't work: Saying [EMAIL PROTECTED]' as a macro argument is equivalent to
`\$@ foo'. So this is probably the final version:
.de CI
. ds xx \\n[.fam]
. fam C
. I \\$@ \F[]\F[\\*[xx]]
. fam
..
As a consequence, I'm not going to implement the removal of
environments :-)
Werner
_______________________________________________
Groff mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/groff
