[mdb-dev] some more "utilities"

[<kn...@gr...>]

I wanted to take a minute to describe what I'm trying to do with this
software and describe another couple utilities that I've generated today.

Our group receives a bunch of C header files from another group that is
derived from the data which that group maintain in the Access database.  We
have to parse these header files and do some data massaging to generate a
whole bunch of internal structures that we use.  A royal pain.  We've
recently rewritten that software to use lex / yacc to improve parsing
performance by two orders of magnitude (yes - a factor of 100 improvement),
but it still has quirks.  We wanted to get our data from the source.

I have generated a little utility that basically generates C header files
that mimic the structure of the database and another one that takes a CSV
(Comma Separated Values) dump from the Access database and creates the
arrays of data.  We're not really interested in creating a database to
update, just having a read-only copy of the data on the UNIX side of the
house, as opposed to having to work on the PC in some Windows environment.
Then we can do the data massaging upon data that is more of a "first class
citizen" than upon derived data.

Eventually, I hope to be able to generate both files from the .MDB file, but
for the moment, getting the header files automatically and having to FTP
over the CSV files and convert them isn't so bad (we get files about once a
month or so).

I'm going to shake this software out with some older databases (I've only
worked with the latest release so far) before submitting it to be included
among the tools.

-- Karl --