Am 15.04.2011 17:59, schrieb Gary: > Michael, thank you for your reply > > ""Might I suggest, instead of the 2 part juggling act, you drop the > auto-increment property on your second table, and just use the value > derived from the first as the joining key in the second. Then there > is only one sequence to worry about with nothing to sync against > "" > > There is only one AI into the main page. This is the insert code, I have > probably left more in than you need to see. > > What I also did was to add some duplicate columns in the two tables (email, > ip, timestamp) so in the event I need to manually to in I would be able to > decifer who goes where. > > On second look, it would appear I am NOT using a join, but two inserts.... I > don't recall why I did it that way
this code is unreadable for me because of its coding-style and if i see "addslashes" for database inserts i start to fear and run away you are using two inserts so what do you do there and where can be anything out of sync on the database-level? where is the magic in your code without using mysql_insert_id() or LAST_INSERT_ID() - what should this code do? * insert in main table * fetch mysql_insert_id() what is thread-safe * use that value in the second table ____________ and please do not use such ugly hacks as in the begin of your code addslashes() has no useable security for user-input even mysql_escape_string() has not -> mysql_real_escape_string()
signature.asc
Description: OpenPGP digital signature
