Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2017-08-25

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:00 samcv err no  9.24x faster
00:00 samcv yay
00:00 raschipi yay!
00:00 raschipi taht's awesome, samcv
00:02 samcv to get a really nice boost when the haystack is made up of strands, i could have the KMP algorithm use a grapheme iterator but that'd be a lot of work since instead of getting a grapheme by index in the string it gets each one, one after another
00:07 hunter_ joined #perl6
00:08 samcv raschipi, i wonder if that makes building rakudo any faster
00:08 samcv since that may happen a lot
00:09 Kyo91_ joined #perl6
00:09 raschipi It's like the linux kernel hackers, the usecase they care most when developing the scheduler is someone working on code while GCC runs on the back
00:09 ccntrq joined #perl6
00:12 samcv what i think i'm gonna make is a caching grapheme iterator
00:13 samcv it stores a normal grapheme iterator plus also stores its current location and it caches the value it last returned, so you can request the same grapheme a second time and it'll pass it back to you
00:13 samcv since the knuth morris pratt algorithm never goes backwards but very often can check the same grapheme twice
00:14 skids Heck I'm enjoying the 60-second 'Stage parse' these days.  Used to be like 79.
00:14 mcmillhj joined #perl6
00:14 raschipi You know linux has special support for that, right? Building in the background while coding... Launch builds in a separate session and the kernel will ensure the active one is completely responsive.
00:15 samcv raschipi, ah has it already sped up a lot?
00:15 skids Well, the 79 seconds was probably like 4 months ago or so.
00:15 raschipi <ctrl>-<alt>-<f1-12>  will start a new session, and you can go back with <alt>-<f1-12>.
00:15 samcv yeah i know that
00:16 raschipi samcv: No, it will be s little bit slower, but won't make a text editor lag.
00:16 skids raschipi: us ancients just call that "switching VTs"
00:16 Cabanossi joined #perl6
00:18 raschipi Yeah, but the important part here is that it also changes the session, not just th vt. vt-> console subsystem concept. session->scheduler concept
00:26 raschipi This is a sysadmin nitpick, of course, but it's possible to build linux with CONFIG_VT=n so one doesn't have virtual terminals anymore (except for the pseudo ones) but there will still be sessions.
00:29 cpage_ joined #perl6
00:30 ryu0 joined #perl6
00:32 raschipi The init system will launch each service in a new session but won't connect them to a vt. When connecting the kernel log ring to a vt, there will be a vt but no session.
00:35 skids just use setsid(1)?
00:40 samcv wow. ok it works. my caching grapheme iterator. neato
00:45 cdg joined #perl6
00:46 raschipi skids, then youll have to be the session leader and there's plenty of setup, it's not straighforward.  Anywy I was just explaining why I used precise terminology instead of the traditional one.
00:46 raschipi samcv: how much faster?
00:46 raschipi samcv is a ricer.
00:46 llfourn joined #perl6
00:51 Geth ¦ doc: 8bef23dbba | (Will "Coke" Coleda)++ | 2 files
00:51 Geth ¦ doc: remove trailing whitespace
00:51 Geth ¦ doc: review: https://github.com/perl6/doc/commit/8bef23dbba
00:51 Geth ¦ doc: 8cc9276751 | (Will "Coke" Coleda)++ | doc/Language/faq.pod6
00:51 Geth ¦ doc: fix some nbsps"
00:51 Geth ¦ doc: review: https://github.com/perl6/doc/commit/8cc9276751
00:51 Geth ¦ doc: 43d883053c | (Will "Coke" Coleda)++ | doc/Language/faq.pod6
00:51 Geth ¦ doc: fix typo
00:51 Geth ¦ doc: review: https://github.com/perl6/doc/commit/43d883053c
00:58 Zoffix Wonder if GitHub tries to be too smart and converts nbsps to spaces when you're using the online editor... I'm fairly certain I did insert nbsps in those FAQ sentences when I edited them
01:00 samcv nice i'm getting 2.5x faster nqp::index when the Haystack is made up of strands with caching and saving the graphemeiterator
01:00 samcv \o/
01:00 mst I wouldn't trust the online editor for anything personally; ingy's git-hub plus a command line actually does what it's told. but I know that's me being a curmudgeon, so please don't take that as a criticism or whatever
01:00 samcv did some statistics doing nqp::index("aaaag" x 100000000 ~ 'h', 'aaagh')
01:01 samcv cache_hit: 200000000 jump: 0 next: 500000000 # so 200000000 times it requested the same grapheme again. and 500000000 times requested the grapheme after the previous
01:01 mcmillhj joined #perl6
01:01 Zoffix samcv++ awesome
01:10 BenGoldberg joined #perl6
01:16 Cabanossi joined #perl6
01:17 mcmillhj joined #perl6
01:35 pilne joined #perl6
01:40 Dunearhp_ joined #perl6
01:51 ilbot3 joined #perl6
01:51 Topic for #perl6 is now »ö« Welcome to Perl 6! | https://perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:,  or /msg camelia p6: ... | irclog: http://irc.perl6.org or http://colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
01:56 BenGoldberg joined #perl6
02:01 Cabanossi joined #perl6
02:17 mcmillhj joined #perl6
02:24 kent\n joined #perl6
02:25 riatre joined #perl6
02:26 Kyo91_ joined #perl6
02:32 samcv these are some of the stats i just got: https://gist.github.com/samcv/f885c0590515603c518c76644198055d
02:32 samcv that's pretty awesome
02:33 samcv 2.5x faster, even matching 'ba' which doesn't have the 1st grapheme matching the haystack anywhere
02:34 noganex_ joined #perl6
02:35 samcv gonna push `index` 9x speed boost when Haystack is strand, needle is 1 long. will hopefully merge the grapheme caching with the next week after input from jnthn
02:40 aindilis joined #perl6
02:42 japhb samcv: Is there a fast path failure for the needle being longer than the haystack?  What about same length (where index is essentially a check for string equality)?
02:46 Cabanossi joined #perl6
02:47 samcv yeah there is
02:47 samcv if needle is longer than the haystack nqp::index will just return -1
03:04 Sgeo joined #perl6
03:08 protium Is running with the bignum pragma conceptually similar to perl6 rats? Any big differences performance wise?
03:11 u-ou- joined #perl6
03:11 u-ou- left #perl6
03:11 u-ou- joined #perl6
03:11 Xliff How do I insure rakudobrew uses a parallelized build?
03:15 samcv protium, no. that creates big numbers but it still does floating point logic i think from looking at the documentation page. perl 6 has rational numbers whic hcontain the numerator and denominator so it doesn't loose accuracy
03:16 protium samcv: Ah, ok, so even more precision
03:16 samcv yep
03:29 kshannon joined #perl6
03:37 * Xliff *cries* -- His rakudobrew still takes almost 90 seconds for stage parse.
03:37 Xliff Of course, this is on a 32-bit VM.
03:38 Xliff I quickly hacked rakudobrew to add parallel builds. Didn't notice much improvement.
03:45 raschipi joined #perl6
03:46 Cabanossi joined #perl6
03:52 samcv Xliff, parallel how.
03:53 BenGoldberg protium, There is a perl5 pragma 'bigrat' which you might compare perl6's rats to.
03:53 protium BenGoldberg: Maybe I should use that. I am reconciling thousands of currency transactions at the office
03:54 protium BenGoldberg: I like the idea of perl6 rats, but there's no perl6 at the office, so perl5 it is
03:57 charsbar joined #perl6
04:00 Cabanossi joined #perl6
04:31 Xliff samcv
04:31 Xliff samcv: Adding the proper argument to GNU's make
04:31 Xliff -J I think
04:31 samcv -j
04:31 Xliff so -J <num threads>
04:31 Xliff Oh. Well.
04:31 samcv that won't do too much
04:32 Xliff :(
04:32 Xliff No, it won't help stage parse, but it should get there faster. :/
04:32 samcv i have: alias make="make -j $(($(nproc) - 1))"
04:33 samcv but most of the perl 6 stuff is linear since it's bootstrapping compilier
04:41 japhb protium: Currency math is often fixed-point (neither floating point nor abitrary-denominator rational arithmetic).
04:42 protium japhb: That makes sense — is there a better way to do it than I was thinking? I am certainly no expert
04:42 japhb Just something to watch out for ... the difference may show up as e.g. a few pennies off in a pile of calculations.
04:43 japhb As for a better way ... well, when I did that kind of work, I did all calculations in whole millicents.
04:43 japhb (So every amount specified in fractional dollars was multiplied by 100,000 and carried out in exact integer math)
04:43 AnotherNick joined #perl6
04:44 japhb I carried millicents in all internal data structures, and converted at first input and final output.
04:44 protium japhb: I thought about that too, and then thought it would be tedious converting back and forth. Nevertheless, that's probably what I should do, too
04:45 Cabanossi joined #perl6
04:46 protium And you're right about the differences, I am pennies out here and there
04:46 protium But that gets annoying when reconciling, so I finally thought I should clean it up
04:46 japhb However you do it, make sure your test suite is strong enough to catch common mistakes (mixing floating point, rational, and fixed-point match in different calculations -- easy to do when you e.g. calculation taxes or interest -- forgetting to convert fixed point back and forth)
04:47 japhb (Trivia: One of the first major cases of international computer espionage was discovered because of a penny-scale failure to reconcile two different computer billing systems.)
04:48 protium Very interesting.
04:48 protium Is there somewhere I can read about that?
04:48 japhb Clifford Stoll: The Cuckoo's Egg.
04:48 japhb I can't believe I can still remember that off the top of my head.
04:48 protium And yes, that's a good point, I should write some more tests when I refactor
04:51 raschipi If you're up to a deep refactoring, you could write types that did all the boring stuf and them teach the operators how to deal with them. This way you'd get rid of the cruft and have clean code at the end.
04:52 teatime japhb: neat.
04:52 raschipi And have them write out the correct format when you convert to string.
04:52 japhb Oh, one final bit: When you do need to do a calculation where you'll have to temporary work in higher precision -- calculating tax on your value in millicents, for example -- remember to round-to-even when going back to your fixed-point format.  This keeps half values from always rounding the same way and slowly drifting the calculation in a single direction, rather than jittering around zero error.  That drift was why round-to-even was invented; it would become v
04:53 teatime cut off, "it would become vi" ...
04:53 protium Thanks for that! Yes, that's a really good idea
04:53 japhb it would become visible in stock exchange transaction levels.
04:54 protium Both, though I am not sure that I want to do as deep a refactoring as to start making types, though that would probably yield the cleanest code in the end...
04:54 japhb At some point you have to stop yak shaving.  :-)
04:57 raschipi protium: do it and them publish it as a module so that other people can ensure your math works
04:59 protium Well, could roll my own there, or probably I could just use Math::Currency
05:00 raschipi You know how it works, the easiest way to get the correct answer is not to ask for it, but to publish the wrong one.
05:00 japhb raschipi: That only works if the right person happens to notice said wrong answer.  :-)
05:01 raschipi Well, he could come here and tell you he figured a better way of doing it.
05:02 raschipi protium: Yeah, write perl6 bindings to it and you're set.
05:03 protium Decisions, decisions...
05:03 protium Well, thanks for all the info, this will be good. Hopefully doing this will eliminate the small current amount of glitchiness that I've got
05:05 raschipi G'night people.
05:12 japhb Good night all, and good luck protium
05:12 protium Thanks, night
05:15 Cabanossi joined #perl6
05:22 cpage_ joined #perl6
05:30 Dunearhp joined #perl6
05:38 |oLa| joined #perl6
05:42 kannan joined #perl6
05:48 lowbro joined #perl6
05:48 lowbro joined #perl6
05:52 wamba joined #perl6
05:52 cdg joined #perl6
06:02 nadim joined #perl6
06:04 ufobat_ joined #perl6
06:26 Kyo91_ joined #perl6
06:41 timotimo 8o/ from the venue
06:41 yoleaux 24 Aug 2017 17:47Z <brrt> timotimo: that's way overthinking it
06:41 timotimo oops, the 8 came out of nowhere
06:43 lancew joined #perl6
06:43 mcmillhj joined #perl6
06:45 andrzejku joined #perl6
06:49 timotimo looks like all my friends who already saw dconway's "three little words" talk decided to stay at the hotel for a bit longer :P
06:49 timotimo making slides and such
06:51 ryu0 timotimo: were those three little words "three little words"?
06:52 ufobat joined #perl6
06:53 domidumont joined #perl6
06:53 lancew_ joined #perl6
06:55 nadim good morning. I've had some problems with promises. I shared them with jnth yesterday but I thought it would be of in terest for more people https://gist.github.com/nkh/f482d398b8db001a7778f89975723fbc
06:55 nadim timotimo: ^^ may interest you too
06:57 timotimo ryu0: that'd be cool, too, but it was actually different words
06:57 nadim jnth proposes I report at least some of those, would that format be enough?
07:02 timotimo i'm not exactly sure which refers to which, i.e. "the first" vs "the second" vs "the third"
07:05 nadim right, I did change the text a bit to make it  clearer... I failed :). Let me change that again :)
07:06 domidumont joined #perl6
07:08 nadim with the nuber and text where it hould be https://gist.github.com/nkh/9942e800c8e0f8a65e0ed87f79cb4c25
07:10 leont joined #perl6
07:16 timotimo the random errors aren't supposed to happen, that much i know
07:17 timotimo the memory map under the stack trace is probably completely worthless, i'd throw it out to reduce the necessary amount of scrolling
07:17 timotimo it'd be nice if you could reproduce these issues with --ll-exception so we get at least a bit of a clue what's happening
07:18 nadim The other errors, too many threads and the slowing down, may not be fatal errors but to a lambda dev, ike me, they are total show stoppers.
07:19 nadim I'll write a little blog entry about this, it be interesting to see if other have had similar problems.
07:24 itaipu joined #perl6
07:25 abraxxa joined #perl6
07:26 parv joined #perl6
07:39 alimon joined #perl6
07:40 parv joined #perl6
07:40 darutoko joined #perl6
07:59 Cabanossi joined #perl6
08:02 nhywyll joined #perl6
08:23 AlexDaniel joined #perl6
08:24 zakharyas joined #perl6
08:25 alimon joined #perl6
08:27 robertle joined #perl6
08:31 aindilis joined #perl6
08:32 ufobat joined #perl6
08:37 mniip joined #perl6
08:54 zakharyas joined #perl6
08:57 timotimo i don't know what a "lambda dev" is :)
08:57 timotimo jnthn talk now \o/
08:59 imcsk8 joined #perl6
09:00 Cabanossi joined #perl6
09:09 timotimo oh crap i think the junction folding optimization is broken again …
09:25 sena_kun joined #perl6
09:26 nadim timotimo: you don't want to know ;)
09:31 lancew joined #perl6
09:32 lancew_ joined #perl6
09:35 ufobat joined #perl6
09:39 parv joined #perl6
09:45 Cabanossi joined #perl6
09:47 lizmat joined #perl6
09:47 mr-fooba_ joined #perl6
09:52 cdg joined #perl6
09:53 rindolf joined #perl6
09:59 cdg joined #perl6
10:00 cdg_ joined #perl6
10:03 llfourn joined #perl6
10:15 Cabanossi joined #perl6
10:19 cdg joined #perl6
10:26 piojo joined #perl6
10:26 piojo tadzik: I seem to have made rakudobrew (windows) play nice with msys and cygwin
10:27 tadzik oh, cool!
10:27 piojo But it in required a less "libified" approach--for example, FindBind just doesn't work with cygwin
10:27 piojo and of course, chomp() is not smart enough to detect which line endings it's dealing with
10:27 tadzik heh
10:27 piojo If that's cool with you, I'll clean it up, upload it, and submit a pull request
10:28 piojo actually, I ought to do a bit more testing. I haven't used any capabilities besides "zef" yet
10:28 piojo I'm sure I'm not done yet, but I haven't found the rest of the incompatibilities
10:30 Zoffix \o/
10:30 Zoffix Awesome.
10:30 piojo tadzik: though I'm not talking about building it--just running. (rakudo star can run from msys, and it can PARTIALLY run from cygwin, but rakudobrew can't do either)
10:31 tadzik nodnod
10:46 Actualeyes left #perl6
11:03 AlexDaniel joined #perl6
11:07 margeas joined #perl6
11:12 epony joined #perl6
11:13 vendethiel joined #perl6
11:19 wamba joined #perl6
11:31 kannan joined #perl6
11:31 lizmat joined #perl6
11:35 timotimo yo tadzik o/
11:36 tadzik o/
11:44 wamba joined #perl6
11:48 wamba joined #perl6
11:54 rindolf joined #perl6
11:59 MasterDuke joined #perl6
11:59 giraffe joined #perl6
12:01 lancew joined #perl6
12:10 piojo tadzik: I pushed to github and made the pull request
12:10 piojo That was way too much debugging to just change a few lines of code!
12:11 tadzik :)
12:11 piojo (Err, I mean it only needed a few lines changed in the end)
12:11 tadzik I remember spending about 4 hours to fix 4 characters of code at $work long time ago :)
12:14 piojo You got me beat for sure :)
12:15 piojo though I bet someone has spent a month removing one byte from binary data...
12:16 eroux joined #perl6
12:17 tadzik hehe, probably
12:18 lichtkind joined #perl6
12:20 epony joined #perl6
12:22 gfldexwork joined #perl6
12:29 moritz or several weeks chasing a bug that turned out to be not a bug
12:30 lizmat .tell nadim re http://blogs.perl.org/users/nadim_khemir/2017/08/promises-promises.html , will probably take until after the weekend before it will see more core eyes due to SPW
12:30 yoleaux lizmat: I'll pass your message to nadim.
12:32 masak joined #perl6
12:33 xiaomiao joined #perl6
12:33 xiaomiao joined #perl6
12:33 pecastro joined #perl6
12:33 APic joined #perl6
12:33 leedo_ joined #perl6
12:33 ryu0 joined #perl6
12:33 Khisanth joined #perl6
12:33 kannan joined #perl6
12:33 Voldenet joined #perl6
12:33 Voldenet joined #perl6
12:33 R0b0t1_ joined #perl6
12:33 Levex joined #perl6
12:33 spider-mario joined #perl6
12:34 itaylor57 joined #perl6
12:38 ccntrq joined #perl6
12:40 eroux joined #perl6
12:42 pilne joined #perl6
12:46 zakharyas joined #perl6
12:56 mcmillhj joined #perl6
12:56 ufobat max last bug was a configuration issue :)
13:00 cdg joined #perl6
13:10 cdg_ joined #perl6
13:19 committable6 joined #perl6
13:25 mcmillhj joined #perl6
13:28 geekosaur joined #perl6
13:32 Geth ¦ doc: 3556160fd8 | (Zoffix Znet)++ (committed using GitHub Web editor) | doc/Language/glossary.pod6
13:32 Geth ¦ doc: Add "Huffmanize" to glossary
13:32 Geth ¦ doc: review: https://github.com/perl6/doc/commit/3556160fd8
13:37 mcmillhj joined #perl6
13:46 nattefrost joined #perl6
13:49 lizmat joined #perl6
13:50 bdmatatu joined #perl6
13:54 skids joined #perl6
13:55 benchable6 joined #perl6
13:59 raschipi joined #perl6
14:00 wamba joined #perl6
14:04 Actualeyes joined #perl6
14:10 piojo joined #perl6
14:13 geekosaur joined #perl6
14:14 epony joined #perl6
14:16 timotimo i wonder if slang::tuxic distorts line number information?
14:16 timotimo because the line numbers i get in my profile are weird
14:16 timotimo i really really need to finish the profiler tui explorer
14:19 lizmat yes, you do!
14:21 raschipi profiler tui explorer sounds sweet
14:21 timotimo it can already display a tiny bit of stuff
14:22 timotimo then i got distracted wanting to write a unicode-and-ansi-color segmented-bar-graph thingie
14:22 timotimo sadly it's not trivial :P
14:22 timotimo but it is pretty
14:22 timotimo http://i.imgur.com/tnhR6Zp.png
14:24 epony joined #perl6
14:33 Kyo91_ joined #perl6
14:45 lizmat joined #perl6
14:51 nadim lizmat: as long as it get's some attention I am good :) My worry is not for my ownm insignificant little toy, it's the fear that people who come to P6 to get a feel  for what it can get discouraged. I've been waiting for Christmas long enough to be able to wait for the next Christmas.
14:51 yoleaux 12:30Z <lizmat> nadim: re http://blogs.perl.org/users/nadim_khemir/2017/08/promises-promises.html , will probably take until after the weekend before it will see more core eyes due to SPW
14:53 raschipi joined #perl6
14:53 nadim and Jonathan dig give a few answers yersterday. It's just that if the basic 'start' crashes, burns, and core dumps, there is little incentive to try the rest.
14:55 raschipi_ joined #perl6
14:55 cdg joined #perl6
14:56 raschipi_ Grrr, Linux 4.12 doesn't like my computer
14:56 geekosaur joined #perl6
14:57 cdg__ joined #perl6
14:57 Xliff timotimo: OooOOOoooOOoo colors!
14:57 MilkmanDan joined #perl6
14:58 Xliff raschipi_: Kernel panics? Or just doesn't install?
14:58 AlexDaniel joined #perl6
15:00 cdg_ joined #perl6
15:01 raschipi_ Reboots without showing a backtrace, nothing in the logs.
15:07 dct joined #perl6
15:08 Xliff https://gist.github.com/Xliff/c66c6b6871ec2b321c5a82374fd6ee76
15:08 Xliff Trying to resolve a segfault when using zmq 4.2 and Net-ZMQ
15:10 lancew joined #perl6
15:15 Cabanossi joined #perl6
15:17 levex joined #perl6
15:23 raschipi joined #perl6
15:23 lancew Hello all, I am experiencing issues installing DBIish, with NativeHelpers::Blob failing, is this something others have encountered?
15:25 lancew Here is example of the errors I see: https://gist.github.com/lancew/cd281098cf75973ebcb494f1a4a9098e
15:26 raschipi_ joined #perl6
15:28 ugexe hey lancew - it ended up being an error in DBIish that never bubbled up until the last rakudo release
15:29 lancew ugexe ok, I thought someone else would have to have encountered this
15:29 ugexe give me a minute and i'll see if i cant get you a workaround
15:30 lancew Thanks!
15:30 st_elmo joined #perl6
15:35 ugexe darn, looks like there is another error though (which I remember nine found a week or so back)
15:35 ugexe zef install  https://github.com/ugexe/NativeHelpers-Blob.git@3108ac4 (this has your error patched, but then hits the second problem :( )
15:36 setty1 joined #perl6
15:36 ugexe what perl6 -v btw?
15:39 cdg joined #perl6
15:40 ugexe a good project for someone might be to remove NativeHelpers::Blob from DBIish
15:41 lancew I'm using Rakudobrew and have this: Rakudo version 2017.07-148-g5fcce6732 built on MoarVM version 2017.07-378-g5e94da03
15:41 ugexe i saw this having no idea what NativHelpers::Blob does, however
15:41 ugexe say
15:41 mcmillhj joined #perl6
15:41 cdg joined #perl6
15:42 lancew LOL... me neither
15:42 ugexe lancew: what about `rakudobrew switch moar-2017.07`?
15:43 lancew I shall try
15:43 cdg joined #perl6
15:43 MasterDuke or 2017.08
15:43 ugexe it fails on 08
15:43 ugexe it works on 07
15:44 ugexe (the original fork, not the modified one I posted)
15:44 MasterDuke ah
15:45 Cabanossi joined #perl6
15:47 lancew Hmm... that switch is not working for me, says:$ rakudobrew switch moar-2017.07
15:47 lancew Sorry, I have no idea what 'moar-2017.07' is
15:48 ugexe you'd have to `rakudobrew build moar 2017.07` first then
15:48 ryu0 But waits, there's moar!
15:48 lancew ahhh ok
15:48 ryu0 :)
15:48 raschipi lancew neds moar moar
15:48 lancew lol
15:49 * lancew sits and watches some git cloning...
15:49 ryu0 just what I always wanted to watch. the git reproductive process. :P
15:52 sumdoc joined #perl6
15:53 sumdoc Doesn't perl6 support argument -e ?
15:53 Sgeo joined #perl6
15:53 [Coke] sumdoc: sure it does.
15:53 sumdoc Why perl6 -e 'say "Hello" ' throwing error?
15:54 [Coke] $ perl6 -e 'say 3.^name'
15:54 [Coke] Int
15:54 [Coke] It's just you. What's the rror? (if it's big, nopaste it)
15:55 sumdoc Coke This is the error I am getting ===SORRY!=== Error while compiling -e Unable to parse expression in single quotes; couldn't find final "'" at -e:1 ------> 'say<HERE><EOL>     expecting any of:         single quotes         term
15:55 robertle joined #perl6
15:55 [Coke] sumdoc: perl6 --version ?
15:55 [Coke] what OS?
15:56 sumdoc Coke This is Rakudo version 2017.07 built on MoarVM version 2017.07 implementing Perl 6.c.
15:56 [Coke] did you cut and paste the command you're running exactly?
15:56 sumdoc Windows 10 64 bit
15:56 sumdoc Coke Yes
15:57 sumdoc Coke Can you see the last comment in this discussion by Yihui
15:57 sumdoc https://stackoverflow.com/questions/45857934/executing-perl-6-code-in-rmarkdown/45869682?noredirect=1#comment78728989_45869682
15:57 lancew sumdoc, does Coke's example work for you (  perl6 -e 'say 3.^name'  )?
15:58 sumdoc lancew No. Its throwing this error "===SORRY!=== Error while compiling -e Unable to parse expression in single quotes; couldn't find final "'" at -e:1 ------> 'say<HERE><EOL>     expecting any of:         single quotes         term"
15:59 jnthn iirc, the Windows command prompt doesn't deal with single quotes, only double
15:59 jnthn perl6 -e "say 'hello'" likely works
15:59 [Coke] checking.
16:00 sumdoc Coke lancew Looks like bug in the windows build. It though works in this docker image https://hub.docker.com/r/sumdoc/exper2/
16:01 [Coke] jnthn++ has it.
16:01 lancew That would make sense sumdoc that docker image is Debian. :-)
16:01 mcmillhj joined #perl6
16:01 [Coke] works fine with double quotes as the command quotes.
16:02 * [Coke] hurls https://stackoverflow.com/questions/24173825/what-does-single-quote-do-in-windows-batch-files
16:02 * lancew now has a happy looking DBIish install, cheers!
16:02 [Coke] the docker image is no doubt running a linux variant, where the shell can handle either quoting mechanism.
16:02 TimToady on windows you have to use "" for shell quoting, don't you?
16:03 * lancew goes back to trying his hand at Bailador
16:03 TimToady oh, jnthn++ got it already
16:03 TimToady and [Coke]++
16:03 * TimToady crawls back to his coffeemaker
16:04 lancew lol... now watching "zef install bailador" install. Should have figured out that I'd need to install that again too. :-)
16:05 [Coke] TimToady: I was kidding about nbsp'ing the version, probably don't do that. :)
16:05 TimToady figgered
16:06 ugexe strange, after installing DBIish in 2017.07 I switched back to 2017.08 and its no longer failing
16:06 lancew ugexe, odd. I'll try in a sec also
16:08 ufobat lancew, i am going to make a new bailador release today or tomorrow, fyi
16:08 sumdoc Coke TimToady jnthn With double quote, yeah it works
16:09 ugexe or just use powershell
16:11 lancew ufobat, great stuff. I'm slowly getting there with it. :-)
16:11 lancew p.s. you were totally right about the PR I sent about the prefix-enter. #newbie
16:13 lancew ufobat, are you likely to be attending the London Perl Workshop in November?
16:15 Cabanossi joined #perl6
16:19 * Zoffix waves to sumdoc
16:19 Zoffix Glad to know you're planning to work on some tutorials :)
16:20 * lancew is Entering the dance floor: http://127.0.0.1:3000 :-)
16:20 sumdoc Zoffix Aye thanks :)
16:21 mcmillhj joined #perl6
16:34 lancew ugexe, I just built 2017.08 and DBIish installed OK for me also
16:34 lancew Back in a while folsk, thanks for the help
16:35 lancew S/folsk/folks/
16:35 Zoffix Any time.
16:36 mcmillhj joined #perl6
16:37 mst do we have any UK people who'd like to talk at http://freenode.live/ ?
16:43 cdg_ joined #perl6
16:45 Cabanossi joined #perl6
16:55 mcmillhj joined #perl6
16:56 nowan joined #perl6
17:08 mcmillhj joined #perl6
17:16 moritz .oO( is UK a new UX trend? :D )
17:19 domidumont joined #perl6
17:24 El_Che _sfiguser: but if you're just trying, that's something for lat
17:26 devmikey joined #perl6
17:26 nadim joined #perl6
17:38 |\n joined #perl6
17:38 khw joined #perl6
17:45 Cabanossi joined #perl6
17:45 mcmillhj joined #perl6
17:46 * Xliff goans at moritz, then giggles because that was so bad, it was actually pretty good.
17:47 huf moritz: yes, it stands for User Killing, the final solution to all UI problems.
17:47 Xliff huf++ # mic drop
17:47 huf unfortunately it has other unrelated flaws that make it not terribly attractive
17:48 Xliff huf: What? Like not working well with other, similar initiatives for no good reason? ;>
17:48 nhywyll joined #perl6
17:49 huf nah, things like weeping widows/mothers/orphans, being jailed, losing paying customers
17:49 Xliff s/not working well/not wanting to work/
17:50 huf the third one is the most serious problem, the first two can be solved with socks and money, respectively.
17:50 Xliff Any kind of UI trend that wants to apply anything like "*exit" should be looked at very closely.
17:52 raschipi joined #perl6
17:55 raschipi_ joined #perl6
17:58 MilkmanDan joined #perl6
17:59 leont joined #perl6
18:09 raschipi http://funroll-loops.info/ is out...
18:09 AlexDani` joined #perl6
18:21 dct joined #perl6
18:21 mr-foobar joined #perl6
18:30 espadrine joined #perl6
18:32 cdg joined #perl6
18:34 pmurias joined #perl6
18:34 pmurias hi
18:38 Zoffix \o
18:42 armin joined #perl6
18:43 pmurias Zoffix: modules.perl6.org always displays the builtin modules when doing a tagged search, should I make a PR to fix that?
18:46 Zoffix pmurias: sure
18:46 Zoffix Probably can add tags to core modules here: https://github.com/perl6/modules.perl6.org/blob/master/lib/ModulesPerl6/Model/CoreModules.pm
18:47 Zoffix And then do a similar grep on tag, but on $core_dists here: https://github.com/perl6/modules.perl6.org/blob/master/lib/ModulesPerl6/Controller/Root.pm#L75
18:53 Zoffix (P.S.: the test suite is bitrotten and many tests fail, so don't use it to gauge your cahnges)
18:54 mr-foobar joined #perl6
19:02 wamba joined #perl6
19:28 raschipi joined #perl6
19:45 Cabanossi joined #perl6
19:55 Kyo91_ joined #perl6
19:56 cdg joined #perl6
20:20 cdg joined #perl6
20:24 mcmillhj joined #perl6
20:40 lancew joined #perl6
20:40 mcmillhj joined #perl6
20:45 mr-foobar joined #perl6
20:47 lizmat joined #perl6
21:13 pyrimidine joined #perl6
21:14 TEttinger joined #perl6
21:18 mcmillhj joined #perl6
21:21 pyrimidi_ joined #perl6
21:35 AlexDani` joined #perl6
21:38 Ulti https://freenode.live/ <--- this is a ten minute walk from my flat... \o/
21:38 mcmillhj joined #perl6
21:43 mcmillhj joined #perl6
21:44 pyrimidine joined #perl6
21:45 Cabanossi joined #perl6
21:45 pyrimidine .oO ( Funny, just when I join back I get dropped by a bad wireless connection ) :P
21:55 lizmat joined #perl6
22:00 Zoffix :)
22:01 mcmillhj joined #perl6
22:04 cdg_ joined #perl6
22:21 mcmillhj joined #perl6
22:22 mniip joined #perl6
22:31 mcmillhj joined #perl6
22:44 mcmillhj joined #perl6
23:02 raschipi joined #perl6
23:09 skids joined #perl6
23:22 grondilu_ joined #perl6
23:25 Kyo91_ joined #perl6
23:44 kannan joined #perl6
23:56 BenGoldberg joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo