Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2016-06-01

| 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
04:28 idiosyncrat_ Good night!
06:46 ronsavage joined #marpa
09:50 editable-log joined #marpa
09:52 editable-log joined #marpa
14:47 ceridwen joined #marpa
15:16 idiosyncrat_ joined #marpa
15:38 kaare__ joined #marpa
18:41 idiosyncrat_ rns: t/panda1.t uses an undocumented call, rule_closure(), which is very inconvenient to support during this transition.
18:42 idiosyncrat_ It seems largely to duplicate t/panda.t, and at a quick glance is just a second way of doing the same thing.
18:42 idiosyncrat_ Could you look at this, and check whether the panda1.t test has independent value?
18:44 idiosyncrat_ Otherwise, I'll just delete Marpa::R3::Scanless::R::rule_closure
19:04 rns joined #marpa
19:17 rns idiosyncrat: re http://irclog.perlgeek.de/m​arpa/2016-06-01#i_12588353 -- other than rule_closure(), t/panda1.t also tests all_choices() and rh_values() which are documented shortcuts for t/panda.t lines 151-162 and 220-221, respectively.
19:21 rns By and large, all three (and t/panda1.t as their test) can be deleted (that would roll back to the state after introduction of t/panda.t for ASF)
19:24 rns or only rule_closure() can be deleted, tests for all_choices() and rh_values() can be added to t/panda.t and t/panda1.t deleted.
19:27 idiosyncrat_ Perhaps rule_closure() could be replaced by adding names to the grammar, indexing the closures by name, and then going
19:27 idiosyncrat_ rule_id -> name -> closure
19:28 idiosyncrat_ ?
19:35 idiosyncrat_ Come to think of though, my last suggestion really adds nothing to the test suite.
19:40 idiosyncrat_ rns: if we want to keep panda.t as is, and keep the test for all_choices()/rh_values() in panda1.t, but eliminate rule_closure, what is the best way to do that?
19:44 rns idiosyncrat: off the top of my head -- just duplicate panda.t semantics (bless adverb) in panda1.t will render rule_closure() unneeded.
19:45 rns then panda1.t would become 'panda.t with shortcuts'
20:11 idiosyncrat_ Would you do that?
20:16 idiosyncrat_ Off on errands.  AFK
23:01 ronsavage joined #marpa
23:50 idiosyncrat_ joined #marpa

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