Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2014-04-03

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:03 ronsavage Marpa::R2 V 2.084
00:03 ronsavage Counts: Tests: 542. Modules: 8. Passes: 8. Fails: 0
00:03 ronsavage Duration: 1 minute and 39 seconds
02:45 jeffreykegler joined #marpa
02:45 jeffreykegler ronsavage: re http://irclog.perlgeek.de/​marpa/2014-04-03#i_8533275 -- thanks!
03:19 jeffreykegler Reini
03:21 jeffreykegler ... Urban has a new blog post on Perl's each keyword: http://blogs.perl.org/users/rurb​an/2014/04/do-not-use-each.html
03:21 jeffreykegler I've been avoiding each and looking at Marpa'
03:23 jeffreykegler s code, I see only 4 uses of it -- two of them in the test suite.  By the time of Marpa's next release, they'll probably be gone.
04:20 ronsavage I did: ack 'each %' */lib where */lib is all my Perl modules' lib dirs, and found several harmless cases, and 1 bug.
04:22 ronsavage There has been an outstanding CPAN Tester report of failures in Data::Session, but I could never reproduce it. And it only started getting reported after the hash key code change in Perl recently, re randomization.
04:22 ronsavage I'm pretty sure this explains those reports.
11:49 LLamaRider joined #marpa
15:30 jeffreykegler joined #marpa
15:32 jeffreykegler ronsavage: re http://irclog.perlgeek.de/​marpa/2014-04-03#i_8533663 -- it'll be interesting to see if this issue goes away with the elimination of each.  Btw, it's also possible to set Perl to not use the new hash randomization.
15:35 jeffreykegler Libmarpa's C library does not use hashing at all --just AVL's based on Ben Pfaff's code.  I think that AVL's are underutilized.
16:15 jeffreykegler1 joined #marpa
18:02 yxhuvud joined #marpa
18:30 jeffreykegler1 A question for the channel -- I read some time ago something on the web (a blog post?) that talked about the special difficulties of C programming ...
18:30 jeffreykegler1 that is, if you already are good at Perl, Python, etc., etc., what special challenges are there about C?
18:31 jeffreykegler1 And it listed memory management, the necessity to make sure the memory that your pointers point at is still there, and 3 or 4 other things.
18:32 jeffreykegler1 A very nice compact, helpful list I thought, and I'd liked to find the link again because it's perfect for other folks just getting into C.
18:35 jeffreykegler1 The basic idea is that with C, you're essentially programming the von Neumann machine directly -- and the raw von Neumann model requires you to figure out where to put data, that your types are consistent, and many other things that languages like Perl train you to take for granted.
18:36 jeffreykegler1 So this blog post (?) listed all the gotchas, in a good concise way for a programming expert who has yet to take on the raw von Neumann machine, just to let him know what he's going to be up against.
21:17 ronsavage jeffreykegler: Yes, I've read the new hash randomization could be turned off, but that's not something I wish to do. I do want to know if it fails, and to fix it.
21:18 ronsavage Up until know I've been bamboozled by the (rare) reports of failure, since I only checked the code for dependencies on the order of keys returned.
21:19 ronsavage I did not consider the possibility of remote manipulation (via a method call to a sub which did it) of the underlying hash, which it what was happening in this case.
22:39 jeffreykegler joined #marpa

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary