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 ]

quidity (1296)

quidity
  (email not shown publicly)
http://the.earth.li/~alex/
Jabber: quidity@jabber.earth.li

A relapsing reformed physicist.

Journal of quidity (1296)

Sunday December 02, 2001
10:25 PM

Wrap Around

[ #1470 ]

I've been having far too many late nights and late mornings, so my time zone has moved from Oxford to somewhere on the west coast of America without my following it. Hopefully I can rectify this. The root cause of all lateness is that I had three Christmas dinners in November, now that Christmas is actually approaching I can probably avoid any more dinners.

In perl-space, I've mostly been hiding away and making parrot more how it should be. The test suite is looking more and more sane, with less and less of it being generated as the tests are run. This should mean that it will soon be possible to remove perl from the test suite, although it will be a much tougher job removing it from the generation of code (we don't even have an assembler in C yet...). I've also started poking around with the internals and think I've finally understood how vtables are meant to work, which should make it easier to fix the bugs the tests are showing up.

I've also now released Parallel::MPI::Simple, but still haven't got it working on some clustered machines. I've left it alone for a couple of weeks and will spend a bit of time thinking about it over the next week to see if I can solve the problems. I'd also like to get a couple of other things working, especially process launching but in reverse, so a large MPI job written mostly in C or Fortran can spawn a single perl node to deal with IO or configuration file parsing.

I had hoped for a bit of fun today solving the puzzle in New Scientist, but it was too open to a brute force approach this week with only 3**9 possible solutions to try. In the end the tricks I discovered to avoid writing deeply nested loops where more fun than the actual solving of the problem.

Oh yes, and ruby is nice, but don't tell anyone I said that.