There was a question last week about the behavior of this function, which is inconsistently (and inaccurately) documented in the manual.
A little black box testing shows the following:
1) mysql_insert_id successfully retrieves the last auto_inserted id used by a connection.
2) multiple processes, each with their own connection will obtain the correct last-inserted-id for that connection. This is not table specific! It is the last autoinsert to take place.
3) the function will return the same auto_incremented id until a table with an auto_increment column is updated. ie: select queries and do NOT reset the id.
4) in the case of updates, the following behavior, in contrast to the documentation, is observed:
A. Inserting a row into a table will reset mysql_insert_id
whether you make use of auto_increment functionality
or not. mysql_insert_id will be set to the new value
even if you forced a new value in via sql. ie: insert into test (id, desc) values (NULL, 'test');
mysql_insert_id will = 32 (or whatever the next value was). insert into test (id, desc) values (77, 'test;);
mysql_insert_id will = 77. B. Updating a row with an auto_insert column will reset
mysql_insert_id to 0.Notes:
Version: 4.0.15 standard Platform: Linux x86
-Blunt Jackson
-- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
