Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2017-11-05

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

All times shown according to UTC.

Time Nick Message
00:36 ronsavage joined #marpa
01:29 idiosyncrat Demat!
01:30 idiosyncrat Marpa-R3-4.001_050 has been uploaded and initial CPANtesters results look good.
01:30 idiosyncrat Testing is appreciated.
01:31 idiosyncrat So is a glance at the documents new with this release:
01:31 idiosyncrat Ext_Scan.pod describing external scanning; and
01:31 idiosyncrat Valuer.pod describing the new valuer objects.
01:32 idiosyncrat Also Recognizer.pod (in addition to be renamed) has been heavily rewritten.
01:32 idiosyncrat This release does have "value added" from the user POV.
01:33 idiosyncrat The valuer objects are new and they make it possible, for example, to do ASF's and regular evaluation for the same grammar at the same time.
01:34 idiosyncrat Also, multi-block input is new with this release: the ability to add new input strings as the parse proceeds.
01:35 idiosyncrat (This was actually present in an earlier release, but it wasn't documented and so was not really fully available to the user.)
01:38 idiosyncrat In more minor stuff, the Changes POD, which tracks changes from R2 to R3 has been reorganized -- the number of changes to individual methods and named arguments reached the point where just adding paragraphs for each one made the document unwieldy.  So there is now a section which lists all R2 grammar/recognizer methods/named-arguments alphabetically, so that you can look up those which are relevant to you.
02:09 ronsavage JK: All tests passed here, for Marpa-R3-4.001_050!
02:09 idiosyncrat ronsavage: Thanks!
02:54 ronsavage JK: Have you seen this: http://nullprogram.com/blog/2015/02/17/
02:56 ilbot3 joined #marpa
02:56 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
03:16 lucs (for the bot:) ronsavage> JK: See also http://nullprogram.com/blog/2017/08/20/
03:44 ronsavage joined #marpa
03:44 ronsavage lucs: Thanx! I was beginning to think the bot had stated its Xmas holidays a bit too early :-).
03:46 lucs :)
03:57 idiosyncrat ronsavage: re https://irclog.perlgeek.de/marpa/2017-11-05#i_15403007 -- interesting -- he's obviously looked into the matter closely.
03:58 idiosyncrat The ref-counter / destructor pair occurs a lot within Marpa, which involves similar code being repeated, unlike in the post you cite, where it'
03:59 idiosyncrat s grouped into a .h file.  It's not clear to me which is better -- not repeating logic is good, but so is having everything in one place --
04:01 idiosyncrat it's often a real question whether an encapsulation is really a simplification -- that is, the new IF may be nearly as complex as the code it replaces, plus having the disadvantage that, unlike the original code which was all basic C, it requires looking up and learning.
04:01 idiosyncrat That is, does the encapsulation really hide complexity, and just convert it into a different form.
04:03 idiosyncrat The Makefile piece was IMHO less interesting, though not bad.  The advice to not use recursive Makefiles is sort of like advice not to break your C (or Perl) program into multiple files.  One big file has a lot of advantages, and a lot of disadvantages.
04:39 ronsavage JK: No problem. I posted because I thought you might like some different perspectives.
05:08 ronsavage joined #marpa
08:13 koom_ joined #marpa
11:46 koom_ joined #marpa
20:00 hobbified joined #marpa
20:00 gabiruh_ joined #marpa
21:01 idiosyncrat joined #marpa
21:50 iarna joined #marpa
23:22 ronsavage joined #marpa

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