Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2014-05-06

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

All times shown according to UTC.

Time Nick Message
00:57 jeffreykegler joined #marpa
01:08 vivtek_ joined #marpa
03:46 ronsavage joined #marpa
05:07 jeffreykegler1 joined #marpa
06:42 vivtek joined #marpa
07:50 LLamaRider joined #marpa
08:26 LLamaRider joined #marpa
09:20 LLamaRider joined #marpa
16:03 jeffreykegler joined #marpa
16:48 jdurand joined #marpa
16:53 jdurand Jeffrey, suppose I would like to support terminals_expectedd stuff: is calling once marpa_r_terminals_expected() really more efficient than calling marpa_r_terminal_is_expected() at every MARPA_STEP_TOKEN
17:08 jeffreykegler Are we talking about a case with multiple (ambiguous) tokens at an earleme location?  Otherwise the two approaches involve the same # of calls, I think.
17:13 jdurand Yes - ok, thx
17:22 jdurand Successive calls to marpa_g_start_symbol_set() will overwrite start symbol, am I right ?
17:24 jeffreykegler Yes, that is correct.
18:18 jeffreykegler1 joined #marpa
18:29 jeffreykegler joined #marpa
18:32 jeffreykegler1 joined #marpa
18:45 jeffreykegler joined #marpa
19:05 jdurand Jeffrey, my wrapper on Marpa is up and running well -;
19:10 jeffreykegler Great!
19:10 jeffreykegler What do you plan as a next step?
19:11 jdurand Use it in my XML stuff (I wrote it for that primilarly), revisit SLIF with it, integration within Marpa if you dare
19:12 jdurand Documentation at the prologue, although an experience C user would understand from the .h but well
19:13 jdurand Did you something else in mind
19:13 jeffreykegler So there will be a Marpa-based XML parser?
19:15 jdurand yes
19:15 jeffreykegler I look forward to it.
19:16 jeffreykegler As for "something else", I always have lots of "somethings else" in mind, but I think it is best if people pursue what excites them.
19:19 jdurand that is good, although I believe at some time a collaborative effort on a big project could be necessary to put Marpa in the top places - it deserve it IMHO - I am thinking to the very short but explicit citation in a link you gave very recently
19:20 jdurand Anyway, an XML parser will certainly have big audience - nowadays every phrase containing XML is looked at, whatever the phrasing -;
19:20 jeffreykegler Perhaps.  The Perl world has unfortunately seen a lot of bit colloborative efforts that have done nothing but waste valuable time and effort.
19:20 jeffreykegler ... and more which have been disappointing.
19:22 jdurand My wrapper have interesting things: a built-in trace mechanism, built-in stack manipulation, and is writen in the "extreme programming way" from symbol names point of view
19:22 jeffreykegler Btw, my Pienoir outline was a sort of project suggestion, but it is more on the line of an "framework" or outline within which I see progress happening -- it's intentionally without any immediate objectives, timelines, etc.
19:23 jdurand Yep, I remember it.
19:23 jeffreykegler Symbol names, "extreme programming"?
19:23 jeffreykegler Btw, another reason I am careful about not pushing project ideas too hard ...
19:24 jdurand all pointers end with "p", all void with "v", all integers with "i" and so on... so a pointer to a pointer to an pointer to a void will end with "pppv"
19:24 jeffreykegler Marpa is as original as it is, because I am about as out of touch with orthodoxy and fashion as anyone out there, which is fine but ...
19:25 jeffreykegler it makes me very poor at predicting what will excite people.
19:25 shadowpaste "jdurand" at 88.160.190.154 pasted "t/thin_eq.t test using my Marpa wrapper in TRACE mode" (508 lines) at http://scsys.co.uk:8002/366765
19:26 jeffreykegler Nice, does your C wrapper produce that trace?
19:27 jdurand Yes. Of course the logging mechanism can be overwriten using a user's function pointer
19:28 jdurand After my 20 years in programming, I tend to believe that what is exciting people is that what they are given to eat, crude but well...
19:28 jeffreykegler Btw, I do notice your use what we call camel-case.  That is, LongVarName instead of long_var_name.  I take it that's a personal preference?
19:28 jdurand I mean: big companies have promoted the non-sense JavaScript language: this is what is exciting people the most
19:29 jdurand Naming ? That's the only thing that I adopted almost immediately from the Java language, almost a 100% camel case yes
19:29 jeffreykegler Yes, I have been missing current trends ever since it was COBOL and OS/360.
19:31 jeffreykegler I've gotten used to reading non-camel-case, so it's hard on my eye.
19:31 jdurand 10 years ago, it was XML everywhere, and that persisted because of Java popularity with beans abstraction, that Flex, one the best thing that happened in the past years; adopted as well to do methods injection
19:31 jeffreykegler "what they are given to eat" -- sounds like a literal translation -- would "have rammed down their throat" capture the sense in English?
19:31 jdurand Nowdays, as I said, JavaScript unfortunately
19:32 jdurand Let me google translate -;
19:32 jeffreykegler Actually, if you'd give the French I might puzzle it out.
19:33 jdurand Ok, I meant: "ce qui est populaire est ce qu'on donne a manger aux gens", in other words, popularity is not a side-efffect of some willing, it is a pure consequence of gigantic marketing pressure
19:35 jdurand Btw, in the trace, there are indications that I manage entirely the stack on the behalf of the user, lines like "tack [0..2] -> Stack 0" and so on... Doing that wrapper I relaized the big effort you did with writing SLIF and XS
19:36 jdurand "Stack [0..2] -> Stack 0"
19:36 jeffreykegler My attempt at translation would be "Popularity is just another word for prolefeed"
19:37 jeffreykegler or "popularity is nothing but the result of prolefeed"
19:37 jdurand Interesting, that makes me reminding that 1984 is definitely true more than ever -;
19:39 jdurand Anyway, this is not the scope of this IRC -; sorry - yes, I was thinking that Marpa deserve a probable collaborative project if we want to adress a popular suject
19:39 jdurand subject
19:42 jdurand Ah... coming back to terminals expected, I implemented both. These are the only methods that are a one-to-one with libmarpa, probably. But they had to be present IMHO
19:42 jeffreykegler Perhaps certain large collorative projects will realize they need Marpa. :-)
19:48 jeffreykegler1 joined #marpa
19:49 jdurand I am thinking to a generic rules engine for example, that could derive to code generation
19:49 jdurand well, I have plenty of ideas coming around, have to digest them
19:50 jdurand first I have to concentrate on this xml stuff, it is because of that that I paused my turing machine for ecmascript -;
19:53 jeffreykegler joined #marpa
19:59 jeffreykegler My internet connection is getting flaky, so I'm going to AFK
19:59 jdurand No pb - bye
21:21 ilbot3 joined #marpa
21:21 Topic for #marpa is now Logs: http://irclog.perlgeek.de/marpa/today Nopaste: http://scsys.co.uk:8002/marpa Stable release: https://metacpan.org/pod/Marpa::R2 Developer release, 2014-02-14: https://metacpan.org/release/JKEGL/Marpa-R2-2.079_015 Source: https://github.com/jeffreykegler/Marpa--R2
22:38 ronsavage jdurand: Can I use your code to parse Javascript and then pretty-print it? That would be marvellous in cases where the original JS is obfuscated.

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