lowjoe

Recent Actions

• Commented on Faster Readonly variables with Const::XS
  You should focus the benchmarks only on usage, not on creation. No user who cares about performance is defining constants in a tight loop. And don't forget about neilb's excellent roundup of constant modules: https://neilb.org/reviews/constants.html...
• Commented on Base conversion
  You only mentioned Math::GMPz in passing as related to bigints, but it's also a very fast XS-based module that does base conversions, and supports a higher base, 62 vs ntheory's 31. When I was looking for something to decode bitcoin's...
• Commented on Hash of Arrays Deathmatch : Native Perl vs. DBM::Deep vs. Redis
  calling rand() a lot is slow, so you might want to minimize its impact by using this technique: https://stackoverflow.com/a/63178400 you can then also "rotate" each string for a further reduction of a factor of length($string)....
• Commented on An objective criteria for deprecating community platforms
  You're the same guy advocating the Perl Facebook group, right? The group is private, inside a walled garden, not indexed by search engines: https://twitter.com/PerlDean/status/1594018066911764482...
• Commented on Perl performance evolution over the last decade
  Tracking times in smaller things would also help, since they probably contribute to the changes in the larger, real-world examples. I noticed that the performance of the lc() function changes drastically among perl versions. Just between 5.34 and 5.36 the...
• Commented on Integrated Inconsistencies.
  (The first paragraph of my previous reply was supposed to blockquoted, but guess my html tags weren't applied)...
• Commented on Integrated Inconsistencies.
  So I then took over Algorithm::Line::Bresenham, and found the original code appeared to take the y-coordinate before the x-coordinate...blasphemy And you changed it, breaking everybody's code who was using it, without even a mention in the docs nor changelog (which...
• Commented on Perl performance on Apple M1
  sounds like the "bad state" is preferring the efficiency cores instead of the performance cores....
• Commented on Alternatives to rand()
  Math::Random::MTwist is now only available from Backpan. It disappeared without any notice....
• Commented on Perl7 is a fork of values
  preaction, if you want uncertainty and to break the language and the community, fork it/off....
• Commented on I started to write "Perl Philosophy" lately
  that's an idiom meaning "it's all a foreign language to me". you wrote a post in english pointing english-speaking users to a site that had no english....
• Commented on JSON::Transform - transform JSON-able data structures without code
  looks like an alternative to the wonderful jq tool but with a more inscrutable DSL...
• Commented on I start creating static perl
  When I saw the title saying "static perl", my first thought was a statically-linked build of perl....
• Commented on Article: Installation of Image::Magick under Debian
  Alien::ImageMagick still requires it's own copy of ImageMagick to be compiled, and worse yet, it might require a separate build of perl. It's pod also only mentions that version of Image::Magick on cpan isn't "cpanm compatible". If somebody from the...
• Commented on Article: Installation of Image::Magick under Debian
  It would be nice if you would mention why it's even necessary to do all this instead of just aptitude install ImageMagic and then cpanm Image::Magick. The cpan test page for Image::Magick (PerlMagick) is full of install failures and unknowns...

Comment Threads

• Mike B commented on An objective criteria for deprecating community platforms

  Fun fact: public-inbox is written in Perl!

• Dean commented on An objective criteria for deprecating community platforms

  Thanks for the link Dragonflydigest https://www.dragonflydigest.com/2022/12/24/lazy-in-other-bsds-reading-for-end-of-2022/

• Robert Acock commented on Faster Readonly variables with Const::XS

  My benchmark measures both instantiation and access. If I were to benchmark only access, Const::Fast and Const::XS would show very similar performance. I include both in the benchmark because it's relevant in scripting scenarios, where the process isn't long-running, and variables/constants are instantiated each run. And yes, I know we're talking about fractions of a second but those add up.
  

• Tom Wyant commented on Faster Readonly variables with Const::XS

  Note that Readonly appears to make use of Readonly::XS if that is available. Readonly::XS does not declare a Perl version dependency that I can see, but it is installed on my Perl 5.8.9.

• Sebastian Schleussner commented on Faster Readonly variables with Const::XS

  Line 38ff:
  "Const::XS::PP::const" should be "Const::PP::const".