in reply to Using localtime() as a unique identifier

You should use a key that will allow you to identify each row uniquely. If you insert multiple records into the database "at the same time", chances are some of them will be inserted within the same second. localtime's smallest unit of measurement is a second, so it probably is not going to garuantee uniqueness very well ... 
$ perl -le"print localtime for 0..9"
55161926610362061
55161926610362061
55161926610362061
55161926610362061
55161926610362061
55161926610362061
55161926610362061
55161926610362061
55161926610362061
55161926610362061

[download]
Time::HiRes, however: 
$ perl -MTime::HiRes=time -le"print time for 0..9"
1059265047.22228
1059265047.22246
1059265047.22248
1059265047.22251
1059265047.22253
1059265047.22255
1059265047.22257
1059265047.22259
1059265047.22261
1059265047.22264

[download]
may work very well for your needs. But, keep in mind that a simple integer works well if you remember to keep it incremented. This is one reason why i prefer a database management system instead. (but i wouldn't be surprised if a CPAN module exists that auto increments a key for you.)

UPDATE: i ran the second Perl one-liner again and redirected to a file. Here's the contents of that file: 

1059265476.27685
1059265476.27695
1059265476.27696
1059265476.27696
1059265476.27697
1059265476.27697
1059265476.27698
1059265476.27698
1059265476.27699
1059265476.27699

[download]
Like i said, an auto incremented id makes uniqueness a lot easier to maintain.

jeffa

L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)

In Section Seekers of Perl Wisdom