[Tfug] Managing data

[Claude Rubinson]

Hi all,

I'm trying to figure out how best to manage the data for my
dissertation, and I was wondering if anyone on the list might have
some ideas.  Here's the deal: I like keeping my data as plain text for
a variety of reasons (primarily that it's easy to maintain and that I
can keep it under version control).  The structure of the data,
however, is fairly complex so I keep the data normalized (i.e., spread
across a number of different tables) and the ability to join tables in
a SQL-like fashion would be great (i.e., inner and outer joins, rather
than having to code it up in awk, which is my current strategy).

I'm thinking that I'm just going to write a script that reads the
various tables and creates SQL statements from them to (re)create the
data.  Then I can just hook that script into Emacs such that when I
modify a text file and save it, the data in the database gets
automatically updated.  The database would be only for reading and all
data maintenance would be done on the text files.

Writing such a script really isn't complex at all, and it'll probably
serve my needs just fine.  But I was thinking that maybe there's
already a solution out there that I'm unaware of.  Or maybe somebody
on the list has another idea or solution.

(Incidentally, I definitely don't want to use the database for data
maintenance.  I've really fallen in love with Emacs' orgtbl mode,
which makes working with tabular data a breeze.)

Claude