Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2014-12-10

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

All times shown according to UTC.

Time Nick Message
01:16 ronsavage joined #marpa
02:04 jeffreykegler joined #marpa
02:05 jeffreykegler I'm at work on a custom Lua wrapper for Libmarpa.  I'd hoped to make do with one from SWIG, but I realize I need to write one by hand, for any number of reasons.
02:06 jeffreykegler One reason is that various issues are raised in the process, which it is best I resolve ...
02:06 jeffreykegler such as error handling.
02:07 jeffreykegler Libmarpa is very systematic in its error reporting, but much of the benefit of this is lost, because Marpa::R2 follows the more casual Perl standards.
02:09 jeffreykegler [ By systematic, I mean that every Libmarpa error has a numeric error code, and a mnemonic, both of which applications can rely on.  There's also the usual human-readable description, which *is* subject to change and therefore is not reliable for programmatic checking. ]
02:09 flaviu joined #marpa
02:10 jeffreykegler Anyway, the LUIF, this new Marpa-to_Lua wrapper, and Libmarpa, will have an integrated system of error numbers and mnemonics. ]
02:11 jeffreykegler So I'm busy creating this, which I have to have before wrappering the first function.
04:26 ronsavage joined #marpa
05:12 prontotest joined #marpa
05:12 prontotest left #marpa
07:29 jluis joined #marpa
10:48 koo6 joined #marpa
10:50 koo5 joined #marpa
11:34 flaviu joined #marpa
11:46 jluis joined #marpa
13:46 lwa joined #marpa
15:01 koo6 joined #marpa
15:02 koo5 joined #marpa
15:53 jeffreykegler joined #marpa
17:15 pczarn joined #marpa
17:57 jdurand_ joined #marpa
17:57 jdurand_ I do not remember and is lazy to check... do L0 support the notion of SLIF's () ? Thx
18:23 jeffreykegler jdurand: Yes and no
18:23 jeffreykegler L0 rules support exactly the same *syntax* as G1 rules.
18:24 jeffreykegler But L0 has no semantics, so those aspects of syntax relevant to semantics have no effect.
18:24 jeffreykegler parens - () - hide symbols from the semantics, which is easily done when there is none. :-)
18:26 jeffreykegler Btw, my decision to allow the exact same syntax for L0 and G1 is a debatable one.  It may be convenient for quickly moving rules from one to another, but I mainly did it out of laziness -- one syntax is easier to write and test than two syntaxes.
18:26 jeffreykegler Kollos initially won't have a lexer.  When it does, it will probably have its own syntax.
20:35 ronsavage joined #marpa
20:39 flaviu joined #marpa
21:22 pczarn I wrote a poor man's SLIF
22:41 jeffreykegler joined #marpa
22:42 jeffreykegler re http://irclog.perlgeek.de/​marpa/2014-12-10#i_9787342 -- poverty level is something Kollos has yet to achieve :-)

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