Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2014-12-31

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

All times shown according to UTC.

Time Nick Message
00:10 ronsavage jeffreykegler: I've submitted 2 doc patches for Marpa::R2, but did not receive the github emails. I checked and I am subscribed to notifications, so I don't know what's happening there.
00:23 jeffreykegler #228 and #229 -- I've been getting all these.
00:32 ronsavage Good.
01:42 flaviu joined #marpa
02:58 jeffreykegler joined #marpa
04:56 sirdancealot joined #marpa
05:53 ronsavage joined #marpa
06:50 ronsavage jdurand: I've put in all your suggestions, but I can't think how to test exhaustion (due to the design of the grammar). If you can think of a way, please let me know.
06:51 ronsavage I'm still deciding how to handle printing of fatal error messages. Perhaps I'll print nothing, and just let the user call error_number(). That's that V 1.02 is not yet released.
06:55 ronsavage Everyone: Since error and warning messages are saved, and can be retrieved with calls to error_number() and error_message(), would it be best to print nothing? Or should I 'print' when there is a warning and 'die' when there is an error? The more I think about this, the more I'm tempted to print nothing. The parse() method already returns 0 for ok and 1 for error.
06:57 ronsavage The advantage of printing something, is that the caller does not have to do any extra work to get errors. And there is already a print_warnings flag. Hmmm. Undecided.
07:47 rns joined #marpa
07:49 rns jeffreykegler, jdurand: re http://irclog.perlgeek.de/marpa/2014-12-30#i_9870571 -- done, https://github.com/jeffreykegler/kollos/pull/17
07:54 rns re http://irclog.perlgeek.de/marpa/2014-12-30#i_9867903 -- perhaps we should add a README to that effect in kollos/components/ like Jean-Damien did in marpaWrapper. It can save one google search for those not versed enough in cmake.
08:00 rns left #marpa
09:06 lwa joined #marpa
10:13 jdurand joined #marpa
10:15 jdurand ronsavage: Re http://irclog.perlgeek.de/marpa/2014-12-31#i_9871374 - I guess that either setting exhaustion => 'event' in the recognizer constructor (which will give an event named "'exhausted") and explicitely querying $r->exhausted() if you pause correctly in your grammar can do it
10:16 jdurand "or explicitely"
10:17 jdurand the idea is to return true if there is a balanced match even before the end of the input stream, a bit like Regexp::Common::balanced reacts
11:23 jdurand joined #marpa
11:24 jdurand or maybe a completion event on your start symbol - to be confirmed
12:12 jdurand joined #marpa
12:14 jdurand for instance in my latest (experimental, will tell more about this later) grammar, I use exhaustion event to trigger a call to value and instantiate another recce on the rest of the input and so on, and call value() again at the very end
13:13 sirdancealot joined #marpa
14:33 LLamaRider joined #marpa
15:45 jdurand_ joined #marpa
15:47 shadowpaste "jdurand_" at 217.168.150.38 pasted "trace_values program" (24 lines) at http://fpaste.scsys.co.uk/453257
15:47 shadowpaste "jdurand_" at 217.168.150.38 pasted "trace_values output" (5 lines) at http://fpaste.scsys.co.uk/453258
15:48 jdurand_ Jeffrey, I was wondering if this is normal that trace_values show \undef in the test case pasted upper... Thnx
16:01 jeffreykegler joined #marpa
16:14 jeffreykegler jdurand: re http://irclog.perlgeek.de/marpa/2014-12-31#i_9872456 -- this does look strange
16:14 pczarn joined #marpa
16:36 sirdancealot ronsavage, thx for the typo fix:)
16:49 jdurand joined #marpa
17:28 jdurand jeffreykegler: Re http://irclog.perlgeek.de/marpa/2014-12-31#i_9872601 - ok I'll submit an issue tomorrow if you think it deserve it -; AFK & Happy New Year to everybody
17:31 jeffreykegler It does deserve it.  I'm working on the problem now, a Github issue would help me track this.  Thanks.
17:58 LLamaRider joined #marpa
20:35 pczarn joined #marpa
22:16 jeffreykegler joined #marpa
22:16 jeffreykegler On perlmonks, a question about parsing sudoer/config files, mentioning Marpa: http://www.perlmonks.org/?node_id=1111865
22:17 jeffreykegler The poster may go with regexes, but does seem interested in the more systematic, maintainable approach that Marpa offers.
22:29 btyler_ joined #marpa
22:29 sivoais_ joined #marpa
22:48 hobbs joined #marpa
22:48 hobbs joined #marpa
22:49 jeffreykegler joined #marpa
22:51 ronsavage joined #marpa
23:46 sirdancealot joined #marpa
23:51 LLamaRider joined #marpa

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