🚀 go-pugleaf

In article <705@hslrswi.UUCP>, graham@hslrswi.UUCP (Graham Tritt) writes:
> >> I am wondering if there
> >> is a preprocessor for the whole process that will allow me to say
> >> 	#define .sqroot. @what ever the eqn code is
> >> 	#define .ECOLI.  \fIE. coli\fP
> >> 	#include ...
> >
> >A possible solution to this is to postprocess your document
> >with sed or awk or even global changes with ed.  (sed is preferred).
> >
> >Set up your sed script to look like:
> >s/.ECOLI./\\fIE. coli\\fP/   (syntax not guaranteed)
> >etc...
> >etc...
> >
> >All you have to do is set up your "defines" with sed syntax,
> >and then you t/nroff run a command which looks like:
> >
> >eqn file | troff -macro | sed ... | lp
> >
> >Mark T. Dornfeld
> >utgpu!romwa
> 
> What you need is a pre-processor rather than a post-processor.
> Or, for instance, your line lengths and justification set by
> troff (in our case nroff) will be all screwed up.
> Tell me if I am wrong.
No, you're right, I was wrong.  Somehow my brain got switched
off when I wrote that reply.  I preprocess troff all the time.
Nothing like a little public humiliation to put one in one's
place :-).
> 
> We use several extra pre(pre)processors to make such substitutions.
> For instance, for words in a file "listofstrings", to insert \fI \fR
> bracketing, which is handy for manual pages containing lots of
> command names.  Similarly for the expansion of abbreviations.
> We find that sed is enough, and can be wrapped up in a friendly
> (but slow) script.  Some of these are used once only followed
> by manual checking. Some are automatically called everytime
> (if needed, like lbl, tbl or col) in the script which intelligently
> decides what has to be done and calls nroff in the appropriate pipe.