R-alpha: Re: read.dbf

Paul Gilbert pgilbert@bank-banque-canada.ca
Wed, 24 Sep 1997 10:27:38 -0400


>> If you're going to get into reading databases let me know. I've
already
>> done a fair amount of work on this and have a working interface to
Fame
>> time series databases (and any other database which can be made to
>> support a generic protocol I've been working on).

>I probably won't have any time for this is the near future, but perhaps

>someone else will fill in ...
>Anyway, if there is more info about the `generic protocol' you
mentioned
>that you can easily make available, the please do so.

Kurt

There is a fair amount of information and example code available at
http://www.bank-banque-canada.ca/pgilbert
Follow the link to TSPADI.

There is a working server front end for the Fame time series database
software as well as a very simplistic file based "database" we made for
test purposes. There is also a client side interface for Splus and for a
"generic client" which is intended as an example and also for test
purposes. The version of the code on the web page needs a few minor
changes for R. I have done that and will soon make it available.

Please note that this is not a database protocol like SQL which allows
queries. The client must provide an identifier for an object and the
server simply returns the object.

The protocol uses RPC and has only been tested on TCP/IP  networks,
although I've been told RPC is available with some other networks. It
does work across the Internet. (It can also be run locally without a
network using loopback.)

An interesting project would be to write a wrapper (eg. in Perl) for
some of the data being provided as text formatted pages. Then a server
could be setup to provide the data with the PADI protocol so that it
could be used directly by application programs like R.  If anyone has
some bright student looking for something to do I can provide more
details.

Paul Gilbert

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-