Re: Ok... I got most of it working but . . .

I am using arrays of double precision numbers in certain columns in my
table. When I format the CString object (analogous to Glib::ustring
objects in the linux world) to update this column it looks like this:

"{0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00}"

However, when the odbc driver tries to update the new added record, I
get the following error message server side (and at my application):

ERROR: invalid input syntax for type double precision:
"{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0}"

Evidently, the odbc driver is deciding to optimize my query perhaps?
What am I doing wrong this time?

Thanks,

Bob Caryl

Is that supposed to be an SQL string, like
"INSERT INTO my_table (field1, field2, field3, field3) VALUES
(0.00,0.00,0.00)"

parentheis -- not curly braces?

Bob Caryl wrote:
> I am using arrays of double precision numbers in certain columns in my
> table. When I format the CString object (analogous to Glib::ustring
> objects in the linux world) to update this column it looks like this:
>
>
> "{0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00}"
>
>
> However, when the odbc driver tries to update the new added record, I
> get the following error message server side (and at my application):
>
> ERROR: invalid input syntax for type double precision:
> "{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0}"
>
> Evidently, the odbc driver is deciding to optimize my query perhaps?
> What am I doing wrong this time?
>
> Thanks,
>
> Bob Caryl
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 9: In versions below 8.0, the planner will ignore your desire to
> choose an index scan if your joining column's datatypes do not
> match

Greg Campbell wrote:

> Is that supposed to be an SQL string, like
> "INSERT INTO my_table (field1, field2, field3, field3) VALUES
> (0.00,0.00,0.00)"
>
> parentheis -- not curly braces?
>
>
>
> Bob Caryl wrote:
>
>> I am using arrays of double precision numbers in certain columns in
>> my table. When I format the CString object (analogous to
>> Glib::ustring objects in the linux world) to update this column it
>> looks like this:
>>
>>
>> "{0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00}"
>>
>>
>> However, when the odbc driver tries to update the new added record, I
>> get the following error message server side (and at my application):
>>
>> ERROR: invalid input syntax for type double precision:
>> "{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0}"
>>
>> Evidently, the odbc driver is deciding to optimize my query
>> perhaps? What am I doing wrong this time?
>>
>> Thanks,
>>
>> Bob Caryl
>>
>>
>>
>> ---------------------------(end of broadcast)---------------------------
>> TIP 9: In versions below 8.0, the planner will ignore your desire to
>> choose an index scan if your joining column's datatypes do not
>> match
>
To quote the documentation on the use of arrays in postgresql tables:

"To INSERT values into an array, you enclose all the array elements in
single quotes and braces ({}) and separate multipl elements with a comma."

Hence, if I have an array of double precision numbers called
misctendamt, and I wanted to insert 16 zero values into that array the
INSERT statement would look like this:

"INSERT INTO my_table (misctendamt) VALUES
'{0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00}'"

Unfortunately when using a derived MFC CRecordset object, I do not
actually write queries. I have to use class members of my derived
CRecordset class that are actually variables corresponding to the
columns of my table. So, for the array of double precision numbers, I
populate the CString variable with the
"{0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00,0.00}"
data, and then the odbc driver is apparently optimizing it to read
"{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0}" and it is that the postgresql
server is choking on. The server wants to see those decimal points!

Bob Caryl