Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2016-05-15

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

All times shown according to UTC.

Time Nick Message
01:48 ilbot3 joined #marpa
01:48 Topic for #marpa is now Start here: http://savage.net.au/Marpa.html - Pastebin: http://scsys.co.uk:8002/marpa - Jeffrey's Marpa site: http://jeffreykegler.github.io/Marpa-web-site/ - IRC log: http://irclog.perlgeek.de/marpa/today
04:45 idiosyncrat_ My latest build of Marpa::R3 now starts a Lua interpreter -- it does not do anything with it, just starts it :-)
04:46 idiosyncrat_ But this will probably be more than enough to break our newly debugged build logic :-)
04:46 idiosyncrat_ Good night!
05:29 pczarn joined #marpa
05:37 ronsavage joined #marpa
05:47 kaare_ joined #marpa
05:58 ronsavage joined #marpa
08:16 kaare_ joined #marpa
08:38 pczarn joined #marpa
08:56 koo7 joined #marpa
09:52 jdurand joined #marpa
09:53 jdurand Jeffrey, I was you recently commited in libmarpa's cmake directory - is it working as expected ? I'd like to git subtree libmarpa nad using cmake makes my life much easier - thx
10:01 pczarn For example, `elem**1..4` is rewritten to `elem**0..2 elem**1..2`. Input `elem elem elem` can be divided between these subsequences in two ways: `elem**1 elem**2` or `elem**2 elem**1`
10:04 pczarn I discovered this in a test for the parser engine.
10:22 idiosyncrat_ joined #marpa
10:22 idiosyncrat_ jdurand: IIRC cmake worked fine -- it's been a while since I've used it.
10:23 idiosyncrat_ rns: what is your recollection?  Is our Libmarpa cmake build working well?
10:23 idiosyncrat_ pczarn: yes, that is is ambiguous, but it's not the rewrite I intended.
10:24 idiosyncrat_ For elem**1..4 the rewrite is
10:24 idiosyncrat_ elem**1..2 | elem*2 elem**1..2
10:25 idiosyncrat_ Does my original misstate this?
10:25 idiosyncrat_ Back to sleep!
10:25 idiosyncrat_ AFK
10:30 pczarn Yes, the original description says "1, n" is binarized to "1, pow2(n)" and "0, n-pow2(n)"
10:47 kaare_ joined #marpa
12:14 koo7 joined #marpa
12:15 rns joined #marpa
12:19 rns idiosyncrat, jdurand: re http://irclog.perlgeek.de/m​arpa/2016-05-15#i_12480888 -- last time I checked (a month ago, give or take) libmarpa built fine from the top-level Makefile.
12:19 rns left #marpa
13:11 kaare_ joined #marpa
13:15 teatime left #marpa
15:07 pczarn joined #marpa
16:23 idiosyncrat_ joined #marpa
16:24 idiosyncrat_ pczarn: re http://irclog.perlgeek.de/m​arpa/2016-05-15#i_12480912
16:24 idiosyncrat_ Could you give me the link, so I can fix the original description?  Thanks!
16:58 koo7 joined #marpa
18:27 idiosyncrat_ I've decided to move Marpa::R3's Lua up to Lua 5.3.2
18:28 idiosyncrat_ I'd be keeping it at Lua 5.1 because of LuaJIT compatibility, but decided that's suboptimal because
18:29 idiosyncrat_ Lua 5.3 has quite a few features I can use -- integers, GOTO's, etc.
18:30 idiosyncrat_ 2.) The Lua inside Marpa is special, and may further fork, so trying to keep LuaJIT compatibility might be a lost cause.
18:32 idiosyncrat_ 3.) Using LuaJIT means being building with, and handling bugs with, all the various LuaJIT's on the various architectures, whereas if I stick to pure Lua, Marpa::R3 can carry its own along with it.
18:33 idiosyncrat_ Briefly, keeping LuaJIT compatibility had some cost, and the payoff was in the future and dubious.
18:34 idiosyncrat_ LuaJIT is adding features from new Lua's, so this decision may be reversable at a point where we know how both LuaJIT and Lua actually evolved.
19:03 beaugunderson joined #marpa
19:55 ilbot3 joined #marpa
19:55 Topic for #marpa is now Start here: http://savage.net.au/Marpa.html - Pastebin: http://scsys.co.uk:8002/marpa - Jeffrey's Marpa site: http://jeffreykegler.github.io/Marpa-web-site/ - IRC log: http://irclog.perlgeek.de/marpa/today
22:47 idiosyncrat_ joined #marpa
23:55 ronsavage joined #marpa

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