Re: [Groff] Re: groff and keep query

[Steve Izma]

> The use of the macro is illustrated here:
> 
>   .KF
>   .PSPIC fig3.2.eps 5.0i
>   .ps 10
>   .vs 12
>   .Fc "L" "3.2" "The starting model ...  'true' model to crete 'observed 
> data'."
>   .ps 12
>   .vs 20
>   .KE
> 
> The macro is defined as:
> 
>   .de Fc
>   .       \"Figure caption $1 is C or L for justification
>   .       \"              $2 is Number, $3 is text all on one line
>   .vs 14
>   .if '\\$1'C' \{\
>   .ce 1 \
>   \}
>   Fig. \\$2. \\$3
>   .br
>   .vs 20
>   .sp
>   .tm FIG:.sp
>   .tm FIG:\\$2\tT{
>   .tm FIG:\\$3
>   .tm FIG:T}\t\\n(PN
>   ..
>   

Since .KF uses a diversion to capture the floating object, you
can delay the interpretation of \\n(PN with something like:


.de fig_toc
.ie '\\n[.z]'' \{\
.       tm FIG:.sp
.       tm FIG:\\$1\tT{
.       tm FIG:\\$2
.       tm FIG:T}\t\\n(PN
.\}
.el \!.fig_toc \\$*
..

and call fig_toc in your Fc macro rather than immediate calls to
tm. You need to give fig_toc the 2nd and 3rd arguments to Fc

Why does tmac.ms use PN rather than %? I would think calling \\n%
would be safer here.

-- 
Steve Izma,                               (519) 884-0710 ext. 6125
    Wilfrid Laurier University Press      FAX: (519) 725-1399
    Waterloo, Ont., Canada N2L 3C5        address@hidden