Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2016-05-10

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

All times shown according to UTC.

Time Nick Message
03:05 idiosyncrat_ rns: I think I'll rename all the sl_*.t tests, dropping the "sl_" prefix.
03:06 idiosyncrat_ The "sl_" identified SLIF tests, but almost all tests are SLIF tests, and those few which are not have their own prefix to identify them.
03:36 ronsavage joined #marpa
05:10 rns joined #marpa
05:11 rns idiosyncrat: re http://irclog.perlgeek.de/marpa/2016-05-10#i_12457014 -- sure, now that there is only SLIF, makes sense.
05:24 rns idiosyncrat: re -- http://irclog.perlgeek.de/marpa/2016-05-08#i_12450723 -- Carp handling of objects:
05:24 rns The Carp routines don't handle exception objects currently. If called with a first argument that is a reference, they simply call die() or warn(), as appropriate.
05:24 rns http://perldoc.perl.org/Carp.html#BUGS
05:25 rns So looks like Carp's object handling is die() or warn() anyway. The passage is under BUGS though.
06:58 idiosyncrat_ Good night!
07:31 rns left #marpa
09:27 koo7 joined #marpa
12:25 kaare_ joined #marpa
15:30 koo7 joined #marpa
16:45 JPGainsborough joined #marpa
19:00 JPGainsborough joined #marpa
21:06 Cheery joined #marpa
21:54 idiosyncrat_ joined #marpa
21:55 idiosyncrat_ A thought about the new interface for Marpa::R3
21:55 idiosyncrat_ Currently parentheses hide RHS symbols from the semantics, that is, in
21:55 idiosyncrat_ A ::= B (C D) E
21:56 idiosyncrat_ The semantics only see <B> and <E>
21:56 idiosyncrat_ It's already been decided that parentheses are going to revert to their traditional use for grouping, and will lose this special meaning.
21:57 idiosyncrat_ I'm thinking have having the bang ("!") symbol mean visible, so that our previous example is\
21:57 idiosyncrat_ A ::= B! C D E!
21:57 idiosyncrat_ and
21:57 idiosyncrat_ A ::= B C D! E
21:58 idiosyncrat_ means return just the <D>, while
21:59 idiosyncrat_ A ::= B C! D! E means return an array of 2 elements, the values of <C> and <D>, that is
21:59 idiosyncrat_ [value(<C>), value(<D>)]
22:00 idiosyncrat_ A ::= B C! D! E could also be written as\
22:00 idiosyncrat_ A ::= B (! C D !) E
22:46 ronsavage What's the advantage of putting ! after the token and not before?
22:55 idiosyncrat_ Well, I think of "!" as analogous to quantifiers, as in
22:55 idiosyncrat_ <A>+
22:55 idiosyncrat_ <Symbol>*
22:55 idiosyncrat_ or
22:55 idiosyncrat_ <X?
22:55 idiosyncrat_ Oops -- <X>?
22:56 idiosyncrat_ It certainly *could* work as a prefix operator, but to my eye that's seem strange.
22:57 idiosyncrat_ As a better indicator than my own personal quesses, I look at the Perl stuff and I think all the operands for symbols in Perl 6 grammars are postfix.

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