csv-import

[Bernhard Reiter]

On Mon, Nov 29, 2004 at 11:15:31PM +0100, Moritz Lennert wrote:
> On Mon, November 29, 2004 11:44, Moritz Lennert said:

> > But I guess we should agree on one implementation of a
> > MemoryShapestore. This is up to you developers to decide, as I
> > don't know/understand the code enough to decide this.
> 
> The MemoryShapeStore in Ole's Code is in data_fake.py. To my innocent eyes it
> looks quite complete and self-contained (the import from punkt is actually not
> needed).
> 
> But again: it is up to you to decide on one single implementation.
> And I would need this decision in order to continue working on a
> new csv import extension.

Look at this froma distance it might make sense to separate the issues
to not stop anybody from improving the situation and getting more
experiences.

The problem with a memory only store of geographic objects
probably is the interface and secondly the implementation.
So you could just use any implementation to get started
and later we change it as long as the interface stays the same.

And as I believe it will take us a while to get to a really nice interface
we can use the experiences made by working code for this.


> I've looked at csv.py in more detail and I like it a lot, especially because
> of the apprarently very flexible automatic file format recognition which
> should allow reading most csv formats.
> 
> I could imagine a first step of just parsing a csv file and writing a new
> MemoryShape. This should be quite easy with an existing MemoryShapeStore and
> csv.py.
> 
> Then in a second step we could write an import wizard for fine-tuning the
> import if necessary.

Sounds good to me.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://www.intevation.de/pipermail/thuban-devel/attachments/20041130/f533d36d/attachment.bin