| --- Log opened Sat Aug 24 00:00:18 2013 |
| 00:00 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 00:00 | | mode/#code [+o himi] by ChanServ |
| 00:00 | < simon_> | RichyB, how does an LL(1) grammar verify that tags have corresponding closing tags in the right order to arbitrary depth? |
| 00:01 | < simon_> | RichyB, you need more than an LL(1) grammar, I think. HTML tag soup is LL(1). |
| 00:02 | < simon_> | RichyB, LL(1) and a stack, or LR(1) (which has a stack, so...) |
| 00:02 | < RichyB> | Oh yes. Hrmn. The fact that the tags have to match up looks problematic at first, like you might even need backreferences⦠|
| 00:02 | < simon_> | I wouldn't say that's necessary. but you can actually parse XML using regex+backrefs, I think. |
| 00:02 | < simon_> | it's crazy, though. |
| 00:02 | < RichyB> | â¦but actually, they only match up in pairs, which is pretty easy. |
| 00:03 | < simon_> | no, wait. you need PCRE's recursion stuff too |
| 00:03 | < simon_> | yes |
| 00:03 | <&McMartin> | Yeah, you'd just do "type checking" when parsing the close tags |
| 00:03 | < RichyB> | You can very easily augment LL(1) with a stack because it never backtracks. |
| 00:03 | < simon_> | so you just need a stack and check that whenever a tag ends, it's the one on top of the stack. |
| 00:03 | < simon_> | RichyB, yes. |
| 00:04 | < RichyB> | Haskell's Parsec is LL(unbounded) for pretty much precisely this reason |
| 00:04 | | * simon_ is taking a course called Semantics and Types this fall... |
| 00:04 | < RichyB> | it's a Monad instance that implements LL parsing and also implements the State monad too on the side because that's useful to have lying around for things like symbol tables. |
| 00:05 | | * simon_ has recently started using Reader monads for keeping lexically scoped symbol tables. |
| 00:08 | < RichyB> | You can think of State as a Reader that you can update, and the updated state will be visible to all subsequent actions in the same computation. |
| 00:32 | | Syka [the@Nightstar-e0f5fd20.iinet.net.au] has quit [Connection closed] |
| 00:40 | < simon_> | RichyB, still around? |
| 00:42 | < RichyB> | Indeed. |
| 00:43 | < RichyB> | I just accidentally made the internet worse for myself by watching http://vooza.com/videos/radimparency/ and, halfway through, wondering, "Wait, what the heck is 'Grindr'?", googling it, and now I'm probably going to get different adverts on the internet than I used to get. :| |
| 00:45 | < simon_> | I've seen that. |
| 00:45 | < simon_> | I'm trying to figure out how to use GeneralizedNewtypeDeriving on my (very small) stack of monad transformers: https://gist.github.com/sshine/652cbf8df87a9759f3ee |
| 00:46 | < simon_> | unfortunately I get an error related to not using my deconstructor, runCodeGen |
| 00:46 | < simon_> | (error present at bottom) |
| 00:47 | < RichyB> | I have no idea how to use GeneralizedNewtypeDeriving (or even a clear idea of what it does), sorry. |
| 00:48 | < simon_> | oh, ok |
| 00:49 | < RichyB> | It's one of a few GHC extensions that I avoid on purpose because it opens holes in the type system. http://joyoftypes.blogspot.co.uk/2012/08/generalizednewtypederiving-is.html |
| 00:50 | < simon_> | when you make a monad transformer stack, you end up with the problem that all those neat get/put/modify/local/etc. monadic actions that applied to individual components of the stack don't work for your new type, and you have to write a bunch of boilerplate code so WhateverNewType can be viewed as a state-monad by lifting the 'get' action to the proper level in the stack. |
| 00:51 | < simon_> | s/(?<=the 'get' action)/e.g. / |
| 00:52 | | You're now known as TheWatcher[T-2] |
| 00:52 | < simon_> | then I can do 'get' as if this were just a state-monad, and 'local' as if it were just a reader-monad, when in fact I make this happen by declaring my newtype an instance of MonadState and having its get/etc. replaced by some wrapper function that handles the complexity I add to the type. |
| 00:52 | < simon_> | e.g. 'get' becomes 'lift get', etc. |
| 00:53 | < simon_> | instead of writing all these instances myself, GeneralizedNewtypeDeriving will let me specify which sub-parts of my type are to be treated as the state of the state-monad, and so on for all the other monad transformers. |
| 00:54 | < simon_> | I made a little exercise for myself to add all these instances manually just to get a grasp of what magic GHC is doing for me. |
| 00:54 | < RichyB> | Oh. |
| 00:54 | < RichyB> | Your problem here is that you wrote "evalStateT (translateProg ... ...)" when you meant to write "evalState (runCodeGen (translateProg ... ...))" |
| 00:55 | < simon_> | and yeah, I did read that blog post you linked to... I'm not sure I like it yet. but the people who've taught me Haskell use it all the time for similar purposes, so... |
| 00:55 | < RichyB> | Yeah, just don't use it in the presence of type-level witnesses or anything. |
| 00:55 | < simon_> | dangit! you're right. I kept trying add runCodeGen to the outside of those calls. |
| 00:55 | < simon_> | hehe |
| 00:56 | < simon_> | I don't know what a type-level witness is yet, but our ICFP team name was "Type Witness Protection Programme" |
| 00:56 | < RichyB> | This is a perfectly boring case of - newtypes needing to be wrapped and unwrapped - even though the wrap and unwrap operations are (AIUI) always no-ops and get compiled to nothing. |
| 00:56 | < simon_> | yup |
| 00:56 | < simon_> | somehow I can't distinguish top from bottom in these monad transformer stacks. |
| 00:58 | < simon_> | apparently you lift from the inside out, the top of the stack is the outermost(?) transformer, |
| 00:58 | < simon_> | thanks, RichyB. |
| 00:59 | | * simon_ is translating to a reversible architecture on which register allocation isn't solved yet. |
| 00:59 | < RichyB> | No trouble. Just the kind of thing that takes two pairs of eyes to stop. |
| 00:59 | < RichyB> | Oh Hell, good luck with that. :) |
| 01:03 | | You're now known as TheWatcher[zZzZ] |
| 01:04 | < RichyB> | simon_, eh, I'd go on continuing with -XGeneralizedNewtypeDeriving. If you're not using type familier or GADTs then I don't think it's likely that you'll accidentally implement unsafeCoerce with it. ;) |
| 01:04 | < simon_> | right :) |
| 01:04 | < simon_> | I haven't really started using type families. |
| 01:07 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has quit [[NS] Quit: Gone.] |
| 01:10 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has joined #code |
| 01:10 | < RichyB> | Oh interesting |
| 01:10 | < RichyB> | They fixed the segfault bugs that GND used to enable by adding one more type of types to the type-checker. :) |
| 01:11 | < RichyB> | http://ghc.haskell.org/trac/ghc/ticket/7148 |
| 01:11 | < RichyB> | Paper linked from here -> http://ghc.haskell.org/trac/ghc/wiki/Roles |
| 01:11 | < RichyB> | So the problem "I can accidentally violate another module's invariants" is still there but the "I can accidentally write unsafeCoerce" isn't. |
| 01:12 | < simon_> | oh |
| 01:13 | < simon_> | when I was writing my bachelor's project, a friend was doing his at the same time on x86 code obfuscation. he wrote an obfuscator which was essentially a bijective function that did a lot of weird jumping around, encrypting things. at his defense, he said that the best way to obfuscate your code is still writing it in Haskell. |
| 01:14 | < RichyB> | ahahahaha no |
| 01:14 | < RichyB> | Write it in Haskell's type-checker instead. ;) |
| 01:14 | < simon_> | right :) |
| 01:14 | < simon_> | my TA did extend it with Peano arithmetic once :) |
| 01:15 | < RichyB> | Haskell is also much simpler (to me, at least) than, say, Coq. |
| 01:17 | < simon_> | I haven't looked into dependent types yet. |
| 01:18 | | * simon_ reads http://research.microsoft.com/en-us/um/people/simonpj/papers/assoc-types/fun-wit h-type-funs/typefun.pdf |
| 01:18 | | Derakon is now known as Derakon[AFK] |
| 01:19 | < RichyB> | Yyyyes |
| 01:20 | < RichyB> | fundeps let you encode functions in the type system |
| 01:20 | < RichyB> | argh for fuck's sake the internet is so fucking useless |
| 01:20 | | * simon_ has in the last month gotten addicted to Hacker News and has to say yes. |
| 01:21 | < RichyB> | *the web |
| 01:27 | < RichyB> | I may have hallucinated it, but I think I remember reading at some point about a Lisp implementation that compiled to Haskell typechecker code. |
| 01:27 | < RichyB> | Bah. |
| 01:36 | | Turaiel[Offline] [Brandon@Nightstar-949d7402.resnet.mtu.edu] has joined #code |
| 01:36 | | Turaiel[Offline] is now known as Turaiel |
| 02:14 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has joined #code |
| 02:31 | | Vorntastic [Vorn@Nightstar-70ca46e4.sub-70-211-9.myvzw.com] has joined #code |
| 03:25 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has quit [Client closed the connection] |
| 03:26 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has joined #code |
| 03:31 | | ktemkin is now known as ktemkin[awol] |
| 03:33 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has quit [Client closed the connection] |
| 03:33 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has joined #code |
| 03:37 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has quit [Client closed the connection] |
| 03:51 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 03:51 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 03:51 | | Vorntastic [Vorn@Nightstar-70ca46e4.sub-70-211-9.myvzw.com] has quit [[NS] Quit: Bye] |
| 04:10 | | Kindamoody[zZz] is now known as Kindamoody |
| 04:16 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has joined #code |
| 04:33 | | iospace [Alexandria@Nightstar-e67f9d08.com] has quit [Connection closed] |
| 04:36 | | iospace [Alexandria@Nightstar-e67f9d08.com] has joined #code |
| 04:36 | | mode/#code [+o iospace] by ChanServ |
| 04:55 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has quit [Ping timeout: 121 seconds] |
| 05:00 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down] |
| 05:28 | | Turaiel is now known as Turaiel[Offline] |
| 05:37 | | iospace [Alexandria@Nightstar-e67f9d08.com] has quit [Connection closed] |
| 05:40 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has joined #code |
| 05:41 | | iospace [Alexandria@Nightstar-e67f9d08.com] has joined #code |
| 05:41 | | mode/#code [+o iospace] by ChanServ |
| 06:02 | | Karono [Karono@Nightstar-13c26ed9.optusnet.com.au] has quit [Client closed the connection] |
| 06:02 | | Karono_ [Karono@Nightstar-13c26ed9.optusnet.com.au] has joined #code |
| 06:04 | | ErikMesoy|sleep is now known as ErikMesoy |
| 06:20 | | Karono_ [Karono@Nightstar-13c26ed9.optusnet.com.au] has quit [Ping timeout: 121 seconds] |
| 06:32 | | Karono [Karono@Nightstar-a97724cd.optusnet.com.au] has joined #code |
| 06:53 | | Vornicus is now known as Vash |
| 06:55 | | Vash is now known as Vornicus |
| 07:07 | | Alek [omegaboot@Nightstar-56dbba0f.in.comcast.net] has quit [Ping timeout: 121 seconds] |
| 07:11 | | Alek [omegaboot@Nightstar-56dbba0f.in.comcast.net] has joined #code |
| 07:11 | | mode/#code [+o Alek] by ChanServ |
| 07:11 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 07:47 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 07:47 | | mode/#code [+o himi] by ChanServ |
| 07:49 | | Kindamoody is now known as Kindamoody|out |
| 08:06 | | Netsplit *.net <-> *.split quits: @iospace, RichyB, @Pandemic, Chutzpah, Xires, @Zemyla, @Syloq, simon_, AnnoDomini, @froztbyte, (+14 more, use /NETSPLIT to show all of them) |
| 08:07 | | Netsplit over, joins: RichyB, Chutzpah, @PinkFreud, @Tamber, @Reiv, &jerith, @Pandemic, @froztbyte, @iospace, &McMartin (+14 more) |
| 08:26 | | VirusHome [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 08:26 | | Netsplit *.net <-> *.split quits: @Alek, @Namegduf, simon_, @Tamber, Xires, @Zemyla, @Pandemic, @Azash, Chutzpah |
| 08:26 | | Netsplit over, joins: Chutzpah, @Tamber, @Pandemic, @Namegduf, Xires, simon_, @Azash, @Zemyla |
| 08:28 | | Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Ping timeout: 121 seconds] |
| 08:31 | | Alek [omegaboot@Nightstar-56dbba0f.in.comcast.net] has joined #code |
| 08:31 | | mode/#code [+o Alek] by ChanServ |
| 09:03 | | Syka [the@Nightstar-250464d0.iinet.net.au] has joined #code |
| 10:00 | | You're now known as TheWatcher |
| 10:09 | | AverageJoe [evil1@Nightstar-4b668a07.ph.cox.net] has joined #code |
| 10:23 | | Syka_ [the@Nightstar-e02c231f.iinet.net.au] has joined #code |
| 10:27 | | Syka [the@Nightstar-250464d0.iinet.net.au] has quit [Ping timeout: 121 seconds] |
| 11:20 | | thalass [thalass@Nightstar-de48278f.bigpond.net.au] has joined #code |
| 11:40 | | AverageJoe [evil1@Nightstar-4b668a07.ph.cox.net] has quit [[NS] Quit: Leaving] |
| 12:13 | | Typherix [Typherix@Nightstar-ea8fda76.lnngmi.sbcglobal.net] has quit [Operation timed out] |
| 12:19 | | Typherix [Typherix@Nightstar-ea8fda76.lnngmi.sbcglobal.net] has joined #code |
| 12:54 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 12:54 | | PinkFreud [WhyNot@NetworkAdministrator.Nightstar.Net] has quit [[NS] Quit: Moving!] |
| 12:57 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [Operation timed out] |
| 12:58 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 12:58 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 13:30 | | Kindamoody|out is now known as Kindamoody |
| 13:36 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving] |
| 14:26 | | VirusJTG_ [VirusJTG@BAD19E.09A45B.582A63.5AE998] has joined #code |
| 14:29 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Ping timeout: 121 seconds] |
| 15:05 | | thalass [thalass@Nightstar-de48278f.bigpond.net.au] has quit [[NS] Quit: *floop*] |
| 15:33 | | Kindamoody is now known as Kindamoody|afk |
| 15:37 | | Karono [Karono@Nightstar-a97724cd.optusnet.com.au] has quit [[NS] Quit: nini] |
| 16:22 | | Syka_ [the@Nightstar-e02c231f.iinet.net.au] has quit [Ping timeout: 121 seconds] |
| 16:23 | | Syka [the@A6D346.0419D1.CCD1D2.871D30] has joined #code |
| 16:46 | | Syloq [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds] |
| 16:48 | | Syloq [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code |
| 16:48 | | mode/#code [+o Syloq] by ChanServ |
| 17:27 | | VirusJTG__ [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 17:30 | | VirusJTG_ [VirusJTG@BAD19E.09A45B.582A63.5AE998] has quit [Ping timeout: 121 seconds] |
| 18:56 | | gnolam_ [lenin@Nightstar-f7705974.cust.bredbandsbolaget.se] has joined #code |
| 18:56 | | gnolam is now known as NSGuest20067 |
| 18:56 | | gnolam_ is now known as gnolam |
| 18:57 | | mode/#code [+o gnolam] by ChanServ |
| 18:58 | | NSGuest20067 [lenin@Nightstar-f7705974.cust.bredbandsbolaget.se] has quit [Ping timeout: 121 seconds] |
| 19:14 | | celticminstrel [celticminst@Nightstar-ae361035.dsl.bell.ca] has quit [[NS] Quit: KABOOM! It seems that I have exploded. Please wait while I reinstall the universe.] |
| 19:14 | | celticminstrel [celticminst@Nightstar-ae361035.dsl.bell.ca] has joined #code |
| 19:14 | | mode/#code [+o celticminstrel] by ChanServ |
| 19:45 | | ErikMesoy [Erik@Nightstar-0fb48670.80-203-17.nextgentel.com] has quit [Ping timeout: 121 seconds] |
| 20:23 | | Kindamoody|afk is now known as Kindamoody |
| 20:56 | | Derakon[AFK] is now known as Derakon |
| 21:09 | | VirusJTG__ [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down] |
| 21:09 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 21:26 | | Kindamoody is now known as Kindamoody[zZz] |
| --- Log closed Sat Aug 24 22:58:56 2013 |
| --- Log opened Sat Aug 24 22:59:04 2013 |
| 22:59 | | TheWatcher [chris@Nightstar-3762b576.co.uk] has joined #code |
| 22:59 | | Irssi: #code: Total of 36 nicks [16 ops, 0 halfops, 0 voices, 20 normal] |
| 22:59 | | mode/#code [+o TheWatcher] by ChanServ |
| 22:59 | | Irssi: Join to #code was synced in 38 secs |
| 23:53 | <@Azash> | http://i.imgur.com/vz0fKG9.png?1 |
| --- Log closed Sun Aug 25 00:00:34 2013 |