Stories
Slash Boxes
Comments
NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.

All the Perl that's Practical to Extract and Report

use Perl Log In

Log In

[ Create a new account ]

cwest (1514)

cwest
  (email not shown publicly)
http://caseywest.com/
AOL IM: caseyrwest (Add Buddy, Send Message)

Perl geek.

Journal of cwest (1514)

Monday July 28, 2003
03:49 PM

Date Module(s) Benchmarks

[ #13739 ]

In corrispondance with the author of Date::Tie, a few benchmarks were passed around (by him). The conclusion is that when compared to DateTime, Date::Tie is considerably faster. My suspicions were correct when it was discovered that Time::Piece was considerably faster than Date::Tie.
  Even so, way to go Date::Tie for beating DateTime. I fell victim to the default expectation that anything tied is slow. While that holds true, tied interfaces aren't always the slowest, so do your benchmarking.

Posted from caseywest.com, comment here.

The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More | Login | Reply
Loading... please wait.
  • Funny because Time::Piece 2.00 (not yet out of beta) ISA DateTime.

    I'm not sure that performance is that critical - it's probably in the thousands of operations per second region. And DateTime buys you so much - a unified base class with support for decimal second values and dates outside the 1970-2038 region.
  • Does Date::Tie handle leap seconds? Does it do _anything_ at all useful with time zones? How flexible is its date math? Does it handle dates outside the epoch?

    There are a number of things DateTime does that literally no other Perl date/time module can do. If you need any of them, then you probably are stuck with DateTime ;)

    Note that the author of Date::Tie is also a major contributor the DateTime project as well.