python + sapdb = (-:

[Andrew Dalke]

Daniel Dittmar:
> This is the column type LONG: LONG ASCII and LONG UNICODE for text and
LONG
> BYTE for binary data. The python interface returns a stream object for
these
> columns so that they can be read without loading the whole data into
memory.
> The network protocol allows to skip parts of the stream, but this isn't
> available in the Python interface yet.

Yup, right in the manual I see now "A LONG column is a data type that
contains a
sequence of characters of any length to which no functions can be applied."
with a
link which says 'any length' means 'no greater than 2GB.'

I would like to read only a given sub-part of that long string.  Any comment
on when skipping will be added to the Python interface?

> User defined functions and the ability to create an index using them is
> currently being implemented.

That's not a near-term requirement for me, but there are some interesting
possibilities I would like to try out in, oh, a year or so.  I'll keep a
watch
out for newer versions.

Thanks,

                    Andrew
                    dalke at dalkescientific.com