Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2014-11-10

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

All times shown according to UTC.

Time Nick Message
00:20 ronsavage joined #marpa
00:36 jeffreykegler joined #marpa
01:29 koo6 ive seen various mentions of scannerless parsing, do i understand it right that it can mean two things, 1 as in marpas SLIF, where i suppose the regexes are collected and a standard separate tokenization is performed, and 2) using a grammar that defines everything down to characters?
01:30 jeffreykegler koo6: yes, that's right.
01:30 koo6 i tried the second approach and i like it and it works...
01:31 jeffreykegler That's why, when I'm being careful I refer to scanless parsing, explaining that it means using a built-in internal scanner, rather than having to provide your own.
01:31 jeffreykegler koo6: glad to hear that it works.
01:32 koo6 i see
01:32 jeffreykegler I experimented with it, but did not go with it for the SLIF, because my feeling was that that scannerless parsing was rarely something that folks really wanted.
01:33 jeffreykegler koo6: Are you aware of the article "That solved problem that isn't"
01:33 jeffreykegler * That -> The
01:33 koo6 well i thought i'd describe the details of how i did it to get some feedback before trying it on a bigger scale, but i think ill just keep going
01:34 koo6 yeah, im not sure if i read it whole, ill review it
01:34 jeffreykegler Actually Marpa goes in a very different direction from Tratt's suggestions.
01:35 jeffreykegler Tratt's article was important for getting me thinking though.
01:35 jeffreykegler Re describing details vs. forging ahead, do whichever you prefer.
01:35 jeffreykegler I'll be very happy to see the results of your research.
01:36 jeffreykegler I'm sure that some people out there really *do* want scannerless parsing in Tratt's sense of the phrase.
01:36 jeffreykegler And Marpa can do it, I believe.
01:39 koo6 ok. i did read that article before. for what my comprehension is worth
01:42 koo6 its just that im getting tired here, but: https://github.com/koo5/new_shi​t/blob/dev/marpa_cffi/test2.py
01:42 koo6 i create a symbol for every digit, for every letter of every word etc (unless one exists already)
01:43 koo6 i think thats the bit im most unsure about
01:43 jeffreykegler This may be a first -- Python's first Marpa driven parser!
01:43 koo6 well, the first was yesterday or so, wasnt it?:p
01:44 jeffreykegler (A caveat: I think Peter Stuifzand did some stuff with Python many months ago, and I'm not sure where he abandoned his efforts.  I'm pretty sure he's done nothing with it for a while.)
01:45 jeffreykegler koo6: You said "seems" about it working, which made me think you hadn't produced an example yet.
01:45 koo6 ive seen a post on reddit from someone about marpa and python, but no code anywhere
01:46 jeffreykegler I know folks have asked for a Python port of Marpa, including on our G+ group.
01:46 koo6 test0.py reproduces the example in the doc page of marpa's Thin
01:46 koo6 in case you didnt explore the rest of the repo, im making a syntax directed editor
01:47 jeffreykegler Why don't you announce this on the Google group -- it gets a larger readership than this channel
01:49 koo6 umm, i will think about it, but i have no idea if anything i will build on top of this would be usable by others
01:50 koo6 but yeah, for what its worth, the lowest layer or two are there, if you want to announce it, feel free
01:50 jeffreykegler Others might want to take your start and make it into something useable
01:51 koo6 i'll add some comments describing what file is what, for a start
01:52 jeffreykegler They might want to help package it for example.
01:53 jeffreykegler I will post something on the Google+ group, to see what kind of response there is.
01:54 koo6 cool
01:54 jeffreykegler Since I won't have tested it myself, what I say will have to be very tentative.
01:54 jeffreykegler But I will encourage Python-oriented folks to look at your repo.
01:54 jeffreykegler What's a good link to give them?
01:55 jeffreykegler (More than one is OK, if there's not *just* one.)
01:57 koo6 hmm, this files's messed up, this one is crappy..i think we should better leave it for tomorrow Y(
01:57 koo6 * :)
01:58 jeffreykegler OK.  Let us know.
01:58 koo6 yeah
02:18 ronsavage A Perl parser for C++ header files: http://starship.python.net/crew/tbryan/PERCEPS/
02:39 koo6 okay, i think https://github.com/koo5/new_shi​t/blob/dev/marpa_cffi/marpa.py is pointable-to
02:42 jeffreykegler koo6: Another thing.  I know it's a nuisance, but could you make the name of the repo more family-friendly.
02:45 jeffreykegler If it gets wide circulation it might fall afoul of some filter.
02:45 jeffreykegler Sorry to be so prissy, but I think it keeps us on the safe side.
02:47 ilbot3 joined #marpa
02:47 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
02:48 koo6 hmm, at least a good argument for once, but no, i dont think i want to take on the hassle of changing the name of my projects repo. Or the hassle of splitting the marpa_cffi part into its own repo, unless somebody actually comes around who will want to take part in it. Would running the url thru a shortener work, for now?
02:50 jeffreykegler OK.  I'll just warn about the name when I post.
02:50 koo6 hehe
04:56 ronsavage joined #marpa
10:03 lwa joined #marpa
11:00 koo6 joined #marpa
13:41 koo6 joined #marpa
16:45 jeffreykegler joined #marpa
18:28 jdurand joined #marpa
18:29 jdurand Re http://irclog.perlgeek.de/​marpa/2014-11-09#i_9632984 - nice writeup - although why did you wanted to point me to it lol?
18:46 jdurand joined #marpa
20:42 flaviu joined #marpa
20:57 ronsavage jdurand: I don't expect you to actually do anything, but I think you should be aware of it :-)
22:41 jdurand_ joined #marpa
22:53 flaviu joined #marpa
23:00 jdurand_ Jeffrey, in a custom action, is it possible to know is the rule being fired is a sequence ? I would like to assign automatically a name to every item on the stack based on the rhs name.
23:01 jdurand_ "to know if the rule"
23:01 jdurand_ In the meantime I will assume that if I receive more than needed I retake latest name and that is a sequence -;
23:43 jeffreykegler joined #marpa
23:45 jeffreykegler jdurand: re http://irclog.perlgeek.de/​marpa/2014-11-10#i_9639063 -- only by using the LHS -- which should be enough, I hope
23:45 jeffreykegler IIRC the LHS of a sequence rule is required to be unique -- you can always make it so, if it is not.

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