Sometime in August Horst Herb assaulted keyboard and produced...
|On Monday 06 August 2001 06:55, you wrote:
|
|> this is what i'm doing:
|> # i get the new value like this:
|> select nextval('testtable_the_key_seq');
|
|you are not assigning it to any variable?
|
|> #and then i preforme the insert like this:
|> insert into testtable (an_id, timestamp) values (2, 2);
|
|I assume that an_id is not your primary key?
|
|> say the last value of the sequence is 'n-1' so the select nextval
|> statement above will return 'n' but when i preform the insert the value of
|> testtable.the_key is actually 'n+1'. should i wrap the insert up into a
|> transaction? any help would be most appreciated.
|>
|> i dont know if this is relevent but in perl i'm preparing the statement by
|> using (?,?) and place holders for the (2,2).
|
|Why do you call nextval? If your key attribute is "serial", postgres calls
|nextval for you autiomatically. Thus, you call it once and postgres calls it
|once, hence an increment of 2
|
|Horst
this is what i want to do:
i want to insert data into a table that has a serialized key. then i want
the value of the key for the row i just inserted. from the stuff i read
on the net i though i was supposed to run the nextval command first then
preform the insert. referring specifically to the url below:
http://www.mail-archive.com/dbi-users@isc.org/msg15723.html
i was under the impression that nextval would give what the nextval would
be. i didnt think it was suppose to increment it. after reading your
response and someone else's it seems i need to get the next value of the
key and insert it explicitly is that correct?
for what it's worth the sql to create the table:
CREATE TABLE testtable(
the_key serial,
an_id int4,
timestamp int4,
PRIMARY KEY(the_key));
thanks
john
