Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2016-02-08

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

All times shown according to UTC.

Time Nick Message
00:11 koo7 joined #marpa
00:14 ceridwen joined #marpa
00:14 ceridwen joined #marpa
00:16 koo7 joined #marpa
00:31 koo8 joined #marpa
01:09 Idiosyncrat joined #marpa
01:32 lucs Re. the backlog: yes, if someone notices that 'ilbot3' is AWOL, please inform moritz, who will gladly bring it back up for us.
01:41 Idiosyncrat Perhaps ilbot3 is PEG-based, and is tired of us throwing shade. :-)
01:42 Idiosyncrat ronsavage: re http://irclog.perlgeek.de/m​arpa/2016-02-07#i_12003027
01:42 Idiosyncrat Thanks.
01:42 Idiosyncrat A not-necessarily-all-that-pedantic point.
01:44 Idiosyncrat PEG *does* have a theoretical representation, and so in that sense, its current implementations are reference implementations, and Marpa's would be "wrong".
01:44 Idiosyncrat The algorithm is discussed in the 2-volume Aho&Ullmann.
01:46 Idiosyncrat PEG implement TDPL's (top down parsing languages).
01:49 Idiosyncrat A&U make a comment (Vol. I, p. 466) later overlooked:
01:49 Idiosyncrat that it can be quite difficult to determine what language a TDPL program is actually parsing.
01:51 Idiosyncrat But by definition the TDPL does implement that language, and Marpa+PEG if it was different (including different and better) would not be a reference implementation of the TDPL program.
01:51 Idiosyncrat This is why if someone does a Marpa+PEG, I suggest they rename it.
01:52 Idiosyncrat When folks use PEG, they expect the language specified by the BNF, which is *not* what happens, unless your BNF happens to be LL(1).
01:53 Idiosyncrat Marpa+PEG would deliver the language the BNF implies, even if it is not LL(1).
01:54 Idiosyncrat But that is not the TDPL language (whatever it might be).
01:56 Idiosyncrat This means, for example, if you *have* managed to get your PEG spec to run, dropping in a Marpa+PEG replacement will change the behavior.
01:56 Idiosyncrat Which if you are just starting development is almost certainly a good thing.
01:56 Idiosyncrat But if you have a production implementation that you just want to keep running as it has been in the past, may not be a good thing.
02:07 ronsavage JK: Fair enough!
02:07 ronsavage lucs: And just how does one contact Moritz?
02:09 ronsavage JK: But to clarify, my comments were somewhat tongue-in-cheek, since it would be, errr, cheeky, to out-PEG PEG and still call it PEG.
02:42 lucs ronsavage: Oh, he hangs out on freenode, so just  /msg moritz ...
02:48 ilbot3 joined #marpa
02:48 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
03:01 ronsavage lucs: Thanx.
04:39 Idiosyncrat ronsavage: and I suppose when you talked about putting nails in PEG's coffin, no pun was intended? :-)
04:59 ronsavage joined #marpa
04:59 ronsavage Weeelllll. Nails and coffins go together naturally, just like Marpa and parsing......
05:00 ronsavage Off-topic: Anyone know why my new module gets this error: "bsd_glob" is not defined in %File::Glob::EXPORT_TAGS. I used: use File::Glob ':bsd_glob';.
06:00 Idiosyncrat Good night!
06:04 kaare_ joined #marpa
06:30 jdurand ron: maybe this can help; https://github.com/clintong​ormley/ConfigMerge/issues/3
06:30 jdurand AFK
06:39 ronsavage jdurand: Yes, thanx. bsd_glob is only available from Perl V 5.16 on. See http://search.cpan.org/~shay/pe​rl-5.22.1/pod/perl5160delta.pod. I switched to just glob, and the tests all work, so released a new version.
07:17 ronsavage joined #marpa
09:16 kaare_ joined #marpa
09:47 ronsavage It turns out that bsd_glob was available V 5.6 but the tag :bsd_blog was only added for V 5.16.
10:29 koo8 joined #marpa
14:31 koo8 joined #marpa
20:45 aredridel joined #marpa
21:53 ronsavage joined #marpa
23:07 koo7 joined #marpa

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