[Sqlalchemy-users] mssql + adodbapi DateTime conversion patch

[Peter L. Buschman]

adodbapi (or perhaps SQL Server in general) doesn't seem to like the
isoformat() output
when used as follows:
start_datetime = datetime.datetime.now() -
datetime.timedelta(days=7)
debug(start_datetime.strftime('%Y-%m-%d %H:%M:%S'))
query = select([table.c.somedatetime], table.c.somedatetime >=
start_datetime)
query.execute()
produced the following exception.
com_error: (-2147352567, 'Exception occurred.', (0, 'Microsoft OLE DB
Provider for SQL Server', 'Syntax error converting datetime from
character string.', None, 0, -2147217913), None) 
Changing the following method in MSDateTime resolved the error but it
meant eliminating the millisecond resolution of the datetime field,
therefore
I'm not sure if this is a good change or if there is a better way to do
this without losing granularity.  Since my data is only granular to
seconds, this
works for
me...                                                    

# BEFORE
    def convert_bind_param(self, value, engine):
        if hasattr(value,
"isoformat"):
            return
value.isoformat(' ')
        else:
            return
value
# AFTER
    def convert_bind_param(self, value, engine):
        if hasattr(value,
"isoformat"):
            return
value.strftime('%Y-%m-%d %H:%M:%S')
        else:
            return
value
Thoughts?
--PLB