database

[Michael Somos]

Sloane wrote :

> it is just one large plain vanilla file!

Incredible! I had overlooked the obvious choice. It had seemed to me
that this was too simple and might have some unknown drawbacks as well.

> i edit it using vim

Fantanstic! I use vim myself since it is the "vi" replacement on
Red Hat. I did not think the performance would be adequate for that.
Also, for version control I thought at least CVS might be involved.
A full-feature database might have transaction logs also.

> and search it with gre

I recall the name of that but could not locate the source for download
when I looked previously. It is like trying to find info on Java the
island nowadays.

> all on a large unix machine

I imagine that the file must be around 32MB uncompressed. That seems
to require a big machine to handle it all in-memory.

Plouffe wrote :

> as with Neil Sloane, i keep a copy of the EIS database
> on disk and I use fgrep, look and Maple.

It had not occurred to me that "look" could use a a file other than the
/usr/dict/words, but sure enough the "man" page says it is so.

> For converting the sequences from the plain vanilla
> text file (Neil's own original file), I use a unix script
> that transforms it into a  Maple ready file. 

I am not sure what this means since I don't use Maple, but I believe
that a UNIX script could do practically anything given the right commands.
Shalom, Michael

-- 
Michael Somos <somos at grail.cba.csuohio.edu>     Cleveland State University
http://grail.cba.csuohio.edu/~somos/            Cleveland, Ohio, USA 44115