Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2016-03-28

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

All times shown according to UTC.

Time Nick Message
00:08 ronsavage joined #marpa
00:09 ronsavage OK. Let's agree then that handouts are what /they/ call modern!
01:36 idiosyncrat_ It is interesting that it requires a close reading to determine if it was not written 30 years ago
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
03:58 kaare_ joined #marpa
05:09 idiosyncrat_ Good night!
06:55 pczarn joined #marpa
08:12 sivoais joined #marpa
08:48 koo7 joined #marpa
10:20 jdurand joined #marpa
10:21 jdurand FYI I have released https://github.com/jddurand/MarpaX-Java-ClassFile - remains the blog post
14:27 jdurand Hope you will enjoy the blog -; http://blogs.perl.org/users/jean-damien_durand/20​16/03/binary-parsing-with-marpa-class-files.html
14:50 idiosyncrat_ joined #marpa
15:05 idiosyncrat_ jdurand: re http://irclog.perlgeek.de/m​arpa/2016-03-28#i_12248055 -- Cool, cool, cool!
15:10 maybekoo2 joined #marpa
15:17 ceridwen joined #marpa
15:17 ceridwen joined #marpa
15:20 jdurand idiosyncrat_: thx
15:38 ceridwen joined #marpa
18:05 pczarn joined #marpa
18:08 ceridwen joined #marpa
18:08 ceridwen joined #marpa
19:32 ceridwen_ joined #marpa
19:49 pczarn joined #marpa
20:00 ted____ joined #marpa
20:00 jdurand joined #marpa
20:01 dvxd hi everyone.
20:01 dvxd does anyone do marpa consulting?  And what are your rates?
20:28 jdurand dvxd: for an official thingy, I let the others react - it also depend on country of residence as you may already know, in mine I would not be allowed to charge for such thing, I think
20:31 jdurand because I am already at 100%
20:39 dvxd I'm in Canada.
20:39 dvxd My desire is to provide some EBNF, and get back a parser for a simple version of lisp.
20:39 dvxd so, parenthesis, symbols, integers, floats, strings.  I think that should cover it.
20:40 dvxd If I provide the EBNF, what would it take to go from that to blob of C code I can use as my parser for my dialect of LISP.
20:42 dvxd I might need to tweak the EBNF, so a workflow to go from EBNF to blob of C code would be nice.
20:42 dvxd for instance, my lisp dialect (as yet unnamed) will be UTF-8 from beginning to end.
20:51 ceridwen_ joined #marpa
21:03 jdurand dvxd: you want to generate C code from the EBNF am I understanding right ? The easiest for us to understand would be that you post tiny examples of input and output you expect.
21:05 jdurand Though I have to go AFK, late enough here in the west of europe (I'm in France) - will backlog tomorrow - hoping I understood ok what you meant - thx
21:14 dvxd yep, you understand.
21:14 dvxd basically, I want to create some sort of link list I can traverse in my interpreter.
21:15 dvxd so, EBNF (or BNF if that works) to translate into a blob of C that parses my dialect of LISP, and turns it into a linked list of "atoms", so, some sort of "tree" structure.
21:15 dvxd I know I could do it by recursive descent directly in C; just thought it would be cool to do in marpa, if marpa is suitable.
21:16 dvxd also, by doing it in EBNF, the language spec would be easier for people to read, and there are tools to turn EBNF into railroad diagrams, which are AWESOME.
21:25 jdurand (not really absent yet) - I wrote an EBNF to C thingy like that some months ago - for XML parsing - input was W3C EBNF, output was a generated parser in C - really I have to AFK now lol)
21:43 dvxd ok, talk to you tomorrow. thanks.
21:52 idiosyncrat_ joined #marpa
22:32 idiosyncrat_ I just tweeted Jean-Damien's blog post.
22:32 idiosyncrat_ https://twitter.com/jeffreykeg​ler/status/714580637328416768
23:50 teatime joined #marpa
23:54 teatime I kindof fell in love w/ Perl6 regexes and started working on a long-postponed project that involves parsing a simple expression language.  Then, I hit a kindof wall because a certain feature isn't yet implemented in Perl6.  There may be other bugs or TODO features waiting for me in Perl6, if I continue along that route.  Is Marpa perhaps exactly what I didn't know I wanted all along?
23:55 idiosyncrat_ I'm pretty sure it can implement anything in Perl 6 regexes ...
23:55 idiosyncrat_ because the class of grammars it handles is much larger and it allows procedural parsing.
23:55 idiosyncrat_ And it'll be linear if Perl 6 is, but ...
23:56 idiosyncrat_ there may be speed differences.
23:56 teatime Perl6 is not exactly a cheetah.
23:56 idiosyncrat_ Marpa might make it look like one. :-)
23:56 teatime lol, *facepalm*
23:57 idiosyncrat_ Marpa itself is blazing fast, it's a C library, hand-optimized ...
23:57 idiosyncrat_ but to do all of the Perl 6 features, you have to use Perl 5 callbacks, and those are not fast.
23:58 teatime it seems to have been around for quite a while, and to be a mature, reliable tool. would you say that's accurate?
23:58 idiosyncrat_ But maybe your implementation will be a prototype, and convince the Perl 6 team that Marpa was the way to go, and they'll recode all the Perl/XS stuff in C.
23:58 teatime LOL.  that is... unlikely.  for so many reasons.
23:58 idiosyncrat_ teatime: you mean Marpa?
23:58 teatime yes.
23:58 idiosyncrat_ Yes
23:59 idiosyncrat_ In fact, Marpa::R2 is so solid, I've frozen it -- no new features only bug fixes.
23:59 teatime now, the $10,000 question:  (and where Perl6 let me down):  Can I match against a lone/mis-matched Unicode combining character

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