Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2015-02-09

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

All times shown according to UTC.

Time Nick Message
00:06 koo7 joined #marpa
00:09 koo5 joined #marpa
02:05 ronsavage joined #marpa
02:34 jeffreykegler joined #marpa
02:34 * jeffreykegler reports on his doings
02:35 jeffreykegler It's taken some time, mainly due to several detours, but I now have discard events working.
02:35 jeffreykegler A very minimal test script works.
02:36 jeffreykegler Next is to expand the test script, to make it a more thorough test, and to fix any problems spotted.
02:36 jeffreykegler Then document.
02:36 jeffreykegler The new Marpa::R2 should actually be faster, though probably not measurably so --
02:37 jeffreykegler I had been planning on allowing multiple lexers in Marpa::R2, and I backed all of that out, eliminating one layer of indirection.
02:38 ronsavage jk: http://irclog.perlgeek.de/marpa/2015-02-08#i_10082459. No, I've had not time to implement anything on this, and my way of thinking was constrained by the previously available code, which did not allow changing the event's initial status, so I did not ever consider it. OTTOMH (Off the top of my head) I don't think I have any need of it right now, but I don't question that switching events off and on will be very useful in some situations.
02:38 jeffreykegler I backed out multiple lexers for two reasons -- first, finishing it would involve a lot more hacking than is in the cards for Marpa::R2 ...
02:38 jeffreykegler and, second, multiple lexers turns out not to be that good of an idea, compared to a single lexer with lots of options and features.
02:38 ronsavage By multiple lexers, did you mean running 2 at the same time (on the same input)?
02:39 jeffreykegler The problem with multiple lexers is that, in practice, switching back and forth, gets into all sorts of complications.
02:39 jeffreykegler ronsavage: no, I meant 2 lexers with the ability to switch back and forth.
02:39 ronsavage Yes, I assume the complications would be difficult for the you and the end-user.
02:40 ronsavage 'for the you' => 'for both you'
02:41 jeffreykegler Yes, the other features, such as the ability to re-prioritize lexemes on the fly, turn out to allow the much the same flexibility, but to be easier to handle.
02:42 jeffreykegler ronsavage: re the new initializations -- I *did* decide to add them for all events, not just the new discard events, and I imagine they will come in handy soon enough.
02:52 ronsavage Understand. Indeed, I assumed they were being added for *all* events.
02:53 jeffreykegler Actually, I initially said I would only be added for discard events.  But I guess you assumed that I'd come to my senses, and make it orthogonal. :-)
02:55 jeffreykegler * I would only be added -> they would only be added
03:09 ronsavage Exactly. How could /not/ adding them to other events be justified!
05:26 ronsavage joined #marpa
05:33 ronsavage I went for a swim and had a dream about a new module. It would be based on a trivial grammar,
05:33 ronsavage say SQL's 'create table x();', but also have the 3 types of comments I've used in some grammars,
05:33 ronsavage (1) #..., (2) //..., and (3) /*...*/. It could have discard events for various combinations
05:33 ronsavage of these, and non-discard processing of them too. Because, some command-line clients, e.g.
05:33 ronsavage mysql, can analyze structured comments which contain instructions destined for the client
05:33 ronsavage or the server, while ostensibly processing simple SQL. By processing that embedded grammar,
05:33 ronsavage or ignoring it, the new demo could be tour-de-force of events (present, absent/on, off, switched),
05:33 ronsavage comment handling (3 types), and multiple grammars, all rolled into one. So, what am I bid
05:33 ronsavage for this outline, and when do you intend to have it ready for me to test? ROFL.
05:36 jeffreykegler joined #marpa
05:37 jeffreykegler ronsavage: I believe my first attempt at justification would have been laziness. :-)
05:49 jeffreykegler joined #marpa
05:50 jeffreykegler ronsavage: re https://groups.google.com/forum/#!forum/marpa-parser -- maybe you were kidding, but this is exactly what I'm looking for, for the test suite.
05:50 jeffreykegler So the answer is that, if you give me an outline, I'll have it tested and ready probably in a few days.
05:51 jeffreykegler That is, just a test of the discard processing.
05:54 jeffreykegler I was trying to think of an example that justified various kinds of discard events, so I can justify turning them on and off individually in various combinations, like what I do for other events in https://github.com/jeffreykegler/Marpa--R2/blob/master/cpan/t/sl_evsyn2.t
05:55 jeffreykegler A grammar that goes crazy with the various comment types is exactly what I was looking for.  Thanks!
06:23 ronsavage No, I wasn't kidding. I was thinking of writing it and calling it MarpaX :: Demo :: $Something.... Perhaps StructuredComments, or EventHandling. And it would make a great article!
06:24 ronsavage PS Your link in http://irclog.perlgeek.de/marpa/2015-02-09#i_10083505 is to Google Groups, but I assume it was meant to be to http://irclog.perlgeek.de/marpa/2015-02-09#i_10083473.
06:26 jeffreykegler Yes, you are right about the link.
06:27 jeffreykegler It's late CA time -- time to quit!  AFK
06:27 ronsavage Brief outline. Have a set of test input files, which different combinations of comments, i.e. combining the 3 types with different places within the input. So a file with all 3 comments can be processed in multiple ways. All (discard) events on/off/switched at start up on/off to ensure they are detected/ignored at the 1st input char, etc. That's what I was planning.
06:27 ronsavage Good night!
12:36 lwa joined #marpa
16:36 koo5 joined #marpa
17:45 jeffreykegler joined #marpa
21:07 ilbot3 joined #marpa
21:07 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
22:00 ronsavage joined #marpa
22:38 ilbot3 joined #marpa
22:38 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

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