Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2014-12-03

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

All times shown according to UTC.

Time Nick Message
00:00 jeffreykegler There are texts which lay this topic out carefully, and connect the idea of atomic transactions with locking and race conditions.
00:01 jeffreykegler It's the sort of thing that I'd need half an hour and a blackboard for.
00:02 jeffreykegler But you can have race conditions without simultaneity.  That's how they first occurred.
00:02 koo5 one last shot and i will give you a break:) : do any libmarpa functions use static or global variables or ..
00:02 koo5 ok. ill try to catch up on my theory sometime.
00:04 jeffreykegler jdurand: You'll be pleased, I thnk, to know that Kollos builds are likely to be cmake-based
00:05 jeffreykegler I'm starting the code with your SWIG wrapper, and I plan to expand on your build system ...
00:05 jeffreykegler the idea of converted to autoconf is not very attractive.
00:05 jeffreykegler * converted -> converting
00:42 ronsavage koo5: Consider this. A thread can be interrupted by force by the OS at any time (i.e. not just when the thread does something (I/O) which takes a 'long' time). And, if there are other threads /with the same priority/, or higher, another thread which you thought was not concurrent, may just happen to be run while the first thread is dormant. Yep - it happens, and and JK said, did so with single-cpu architectures. Of course, if you can preempt
01:04 flaviu joined #marpa
01:30 flaviu joined #marpa
02:17 jeffreykegler joined #marpa
02:17 jeffreykegler rns: I'm going to put the JSON parser together.
02:18 jeffreykegler In the first version I might "cheat", and skip the LUIF parsing and just use Jean-Damien's SWIG wrapper to create a grammar.
02:19 jeffreykegler That's no test of the LUIF at all, of course, but it will give me a chance to get most of the build put together ...
02:19 jeffreykegler I will need to gather up the wrapper & Libmarpa and include them.
02:20 jeffreykegler In the process, I may wind up creating a cmake build for Libmarpa, which might eliminate a lot of annoyances with Marpa::R2
02:22 jeffreykegler Putting all of this together may take me into the weekend and beyond.
05:05 jdurand joined #marpa
05:06 jdurand Re http://irclog.perlgeek.de/marpa/2014-12-03#i_9750506 - IMHO, good idea - thx!
05:28 ronsavage joined #marpa
05:51 rns joined #marpa
05:54 rns jeffreykegler: re http://irclog.perlgeek.de/marpa/2014-12-03#i_9750740 — good, so I won't be a bottleneck. :)
05:57 rns BTW, for the moment you might use the json parser you've written in C — https://github.com/jeffreykegler/libmarpa/blob/master/test/json.c — as a base.
05:58 rns left #marpa
09:46 koo6 joined #marpa
09:46 jluis joined #marpa
09:48 koo5 joined #marpa
09:53 rns joined #marpa
10:12 rns jeffreykegler: re LUIF-to-lua — there was a breakdown @work, so I've had time to fix the table building logic so here is how the first cut of JSON LUIF grammar looks in lua — https://github.com/rns/MarpaX-Languages-Lua-AST/blob/master/bin/kollos_json.lua — feedback is appreciated.
10:12 rns You might take a look to see if it is suitable for grammar rewriting.
10:13 rns There is also a luif2lua.pl script in bin directory of the repo, which takes a LUIF source from the specified LUIF file or stdin and writes transpiled Lua to the specified lua file or specified LUIF file (replacing its extention to lua) or stdout.
10:13 rns I think I should make writing a cpanfile (so that you can cpanm install from the repo) a priority or you can work off the cloned repo?
10:14 rns left #marpa
11:03 rns joined #marpa
11:05 rns joined #marpa
11:05 rns left #marpa
11:17 rns joined #marpa
11:21 rns left #marpa
11:39 flaviu joined #marpa
11:42 flaviu joined #marpa
13:20 koo5 joined #marpa
15:26 lwa joined #marpa
15:39 jluis joined #marpa
15:46 jeffreykegler joined #marpa
15:48 jeffreykegler rns: re http://irclog.perlgeek.de/marpa/2014-12-03#i_9751064 -- never a bottleneck!  I appreciate your volunteering.
15:50 jeffreykegler Speaking of which, when the time comes, I you'd do benchmarks of the various JSON parsers vs JSON::XS, JSON::PP, perhaps others, it would be helpful.  But I'm not yet ready for that.
15:50 jeffreykegler rns: re http://irclog.perlgeek.de/marpa/2014-12-03#i_9751072 -- good idea
15:52 jeffreykegler rns: re http://irclog.perlgeek.de/marpa/2014-12-03#i_9751735 -- looks OK.  One reason I figure I need to at least do that that of the rewrite code is that that structure needs to be worked out by actually doing it
15:53 jeffreykegler rns: re http://irclog.perlgeek.de/marpa/2014-12-03#i_9751746 -- I will work off the cloned repo, no need for a cpanfile.
16:13 heinz_ joined #marpa
16:13 heinz_ left #marpa
18:46 koo6 joined #marpa
18:48 koo5 joined #marpa
20:39 flaviu joined #marpa
20:56 ronsavage joined #marpa
23:32 jeffreykegler joined #marpa

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