Responding to Richard Goerwitz: in general, I agree. General purpose
languages are more suitable for the stripping task than a proprietary
word-processor macro system. I do it this way because I already
use Word anyway, and I therefore have to use Word's macros if
I wish to convert to a system of ASCII markup, whether it is SGML
or OUP's printers (for History and Computing). I have the same
problem when I wish to convert the SCRIPT files I used when writing
my thesis into Word. (I suppose I could export from Word to RTF
and then write a program to convert the RTF, but I've never investigated
And I don't have to invest a lot of time doing it. A macro simply
to strip out markup would take about five minutes to write. One
to do conversions would take rather more time! But they're simple
because they intended to be single-purpose, not to be grand
solutions for every problem. I use macros because they suit my purpose and I
find them easy because I have previous programming experience.
I'm currently using such conversions to produce a software guide.
The data is stored in Paradox. Because Paradox's report facilities
are limited, I use them simply to generate a report with simple
tags such as <title>, </title>, <emph>, etc. I then load the file
into Word, where a macro package replaces the markup with Word
formatting, all applied through stylesheets, to retain flexibility.
The resulting laser-printed copy will eventually be given to a
printer as camera-ready copy.
I'm still tinkering with the macros to make the process as automatic
as possible, but it works. To repeat, this is a specialised application
which brings together two software packages I'm using for other
reasons. It is not intended to be a model for everyone.