Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2015-02-13

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

All times shown according to UTC.

Time Nick Message
01:05 koo7 joined #marpa
01:17 flaviu joined #marpa
02:49 ilbot3 joined #marpa
02:49 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
03:04 jeffreykegler joined #marpa
05:31 ronsavage joined #marpa
06:12 sivoais joined #marpa
06:39 pczarn joined #marpa
10:26 rns jeffreykegler: re method census for trivial grammars -- a couple of questions, hope you'll backlog.
10:26 rns sequences -- while proper sequence cannot be added to trivial grammar, I'd add an inaccessible sequence rule with non-nullable seq and item symbols and test methods from 11.6 Not sure: makes sense?
10:26 rns events -- nulled events are tested in trivial.c, should I test other event methods from 11.8?
12:09 lwa joined #marpa
15:13 pczarn joined #marpa
16:36 jeffreykegler joined #marpa
16:40 jeffreykegler rns: re http://irclog.perlgeek.de/m​arpa/2015-02-13#i_10110016 -- re sequences
16:41 jeffreykegler Actually, I think try to add a nulling sequence, and make sure that it fails with an appropriate error code.
16:41 jeffreykegler You can test for specific error code values, by the way -- in the documentation I promise these will be stable.
16:42 jeffreykegler rns: re http://irclog.perlgeek.de/m​arpa/2015-02-13#i_10110018 -- re testing events other than nulled.
16:42 jeffreykegler A test that attempts to create events, other than nulled events, results in a reasonable error, would be good, I think.
16:42 jeffreykegler Thanks!
18:12 koo7 joined #marpa
18:25 koo7 joined #marpa
20:23 jeffreykegler joined #marpa
21:21 koo7 joined #marpa
22:11 ronsavage joined #marpa
22:16 koo7 joined #marpa
23:22 jeffreykegler I finished the 2nd test of "discard events"
23:23 jeffreykegler This one focused on event initialization -- you can initialize an event to on or off in the grammar, and reset that initialization when you create the recognizer.
23:24 jeffreykegler This allows you, for example, to define a lot of discard events in your grammar, initialize them all to off, and then selectively turn them on whenever you do a parse.
23:24 jeffreykegler I have one more test script to do -- a tiny "real life" example, and then the documentation.
23:25 jeffreykegler [ Since I test all the documentation's displays in my test suite, the documentation phase often involves writing new tests as well. ]
23:26 jeffreykegler By the way, AFAIK I'm the only CPAN author who systematically tests displays which, if true, is a little surprising.
23:27 jeffreykegler For a small module, you can get away without it, but for a large complicated one, I can't imagine how you keep the examples in the POD accurate without regression testing.
23:37 jeffreykegler Of course many modules, even large can complex ones, do not need many displays -- for an HTML parser, for example, you don't need your POD to teach HTML, you can assume a module reader will learn it elsewhere.

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