On Sun, Nov 26, 2017 at 03:04:29PM -0500, Peter Schaffter wrote:
> Ulrich --
>
> On Sun, Nov 26, 2017, Ulrich Lauther wrote:
> > 2) if ROMAN or roman is used without a trailing number, e.g., ROMAN instead
> > of ROMAN7,
> > the command ".tm1 "[mom]: You must append a number to the \\$1 argument
> > to \\$0."
> > should be invoced, but is not.
> >
> > The small patch appended introduces
> > LIST VARIABLE <width of largest enunerator>
> > and
> > ITEM <enumerator>
>
> Entering the widest glyph literally rather than its width, which
> would have to be calculated, makes things simpler. Furthermore, if
> the glyph is in double-quotes, users can add space to or subtract
> it from the gutter between the enumerators and the text with \*[FWD dist].
in my tests I used \w'<largest enumerator>', but yes, your solution is much
more
convenient and elegant.
As I am not a versatile groff programmer, I was expecting that some polishing
of my code was called for.
>
> Apply the attached, revised patch and test.
Yes, my tests, both VARIABLE list and error message related to ROMAN,
work nicely now.
Thanks a lot,
ulrich
PS:
why do we have to use "ROMAN<n>" and not "ROMAN <n>", which would be easier to
process?
Historical reasons?
