Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2017-07-26

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

All times shown according to UTC.

Time Nick Message
01:30 idiosyncrat joined #marpa
01:30 idiosyncrat Demat!
01:31 idiosyncrat My current task had been converting the test suite for the mitosis of symbols into external and internal versions.
01:32 idiosyncrat I knew in advance I was leaving some tasks undone into the test conversion phase, because they were more convenient to work on in conjunction with their tests.
01:32 idiosyncrat In particular, I left progress reports needing work, which I planned to do (and was doing) in the course of converting the test suite.
01:32 idiosyncrat But ...
01:33 idiosyncrat Work on the progress reports has made me realize I need at this point to create "recognizer traversers" inside Libmarpa.
01:34 idiosyncrat Traversers enable upper layer to conveniently and efficiently access data in the Earley sets and Earley items.
01:35 idiosyncrat I already did this using the "trace methods", and a lot of the logic I can copy from one to the other.
01:35 idiosyncrat The major difference is that traversers are first-class objects, which can be created, copied, stored on their own ...
01:35 idiosyncrat while the "trace" methods were methods of the recognizer.
01:36 idiosyncrat In practical terms, the trace methods were limited to tracing one Earley set and one Earley item at a time.
01:39 idiosyncrat And any change to the current trace Earley set, Earley item, or other trace data affected every other user of that recognizer -- in other words, the recce trace mutators could cause "action at a distance" if two pieces of logic tried to trace the same recce at the same time.
01:39 idiosyncrat The trace mechanism underlies the ASF (Abstract Syntax Forests), and this strained it almost to the breaking point.
01:40 idiosyncrat I'll soon make it convenient to have several ASFs and several valuers for the same recce at once, and this would strain the trace method beyond the breaking point.
01:41 idiosyncrat So what I am now doing is creating a new traverser class inside Libmarpa.
01:50 cheako joined #marpa
01:52 ilbot3 joined #marpa
01:52 Topic for #marpa is now Start here: http://savage.net.au/Marpa.html - Code paste/run: https://f.perlbot.pl/#marpa - Jeffrey's Marpa site: http://jeffreykegler.github.io/Marpa-web-site/ - IRC log: http://irclog.perlgeek.de/marpa/today - Youtube channel: https://www.youtube.com/channel/UCYKVfGBtfTqbs1JdYq-dc5g
02:49 ronsavage Sounds good to me.
07:07 ronsavage joined #marpa
12:49 ernimril joined #marpa
15:10 idiosyncrat joined #marpa
16:22 lucs joined #marpa
17:30 cheako joined #marpa
19:05 sirdancealot joined #marpa
20:06 kaare_ joined #marpa
20:43 cheako joined #marpa
23:16 idiosyncrat joined #marpa
23:26 ronsavage joined #marpa
23:59 sivoais joined #marpa

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