Jesse,
I've been out of town and didn't see your earlier posting
until this morning. After reading your note, I repeated my
test procedure. It appears that the transaction is opened as soon
as the data is plotted and kept open until the program is
terminated (it may close out if just the layer is closed, but
it didn't occur to me to test that hypothesis until it
was too late). In my test this morning, I never did anything
to interact with the data on the screen (in the earlier tests,
I did use the info button to retrieve data on individual objects).
So I guess it depends on how you look at it.
If just viewing the data constitutes starting an edit, then
your description is accurate. I can certainly see an argument
for taking that approach, but it would be nice if there were
some way to moderate uDig's impact on Postgres in these
circumstances.
Gary
Jesse wrote:
> Judging from Gary's analysis a transaction is not opened until an
> edit starts. But it seems that it is not closed correctly on
> commit.... Unless he didn't commit? So maybe I was jumping to
> conclusions earlier.
---
Gary W. Lucas, Senior Software Engineer
Sonalysts, Inc
215 Parkway North
Waterford, CT 06320
(860) 326-3682
_______________________________________________
User-friendly Desktop Internet GIS (uDig)
http://udig.refractions.net
http://lists.refractions.net/mailman/listinfo/udig-devel
