Request to use parameter arrays in executemany() about pyodbc HOT 5 CLOSED

We ran into this problem while inserting thousands of ids into a table. Our workaround was to break the id list into groups of 250 (**see below) and send a bunch of insert statements at once joined with a semi-colon along with the list of ids as parameters.

**We noticed if we'd send more than 500 statements, the function would return without complaining, but missing several inserts.

Fixing insert many to work as the ceODBC library does would be a great help.

from pyodbc.

Yes, please! There's a thread on this on the (now closed?) Google Code page (https://code.google.com/p/pyodbc/issues/detail?id=250).

I am using ceodbc for all of my insert and update statements, and pyODBC for my select statements. It would be nice to just use a single connection and cursor for everything!

from pyodbc.

It doesn't look like that could be sensibly merged. It seems there are many changes not related to parameter arrays.

Also, is it really needed to add an extra kwarg to the connect method? Can pyodbc not check support for parameters arrays and do the right thing automatically?

from pyodbc.

The recently-merged pull request #256 may help with this.

from pyodbc.

Feature added in 4.0.19. See fast_executemany for details.

from pyodbc.