Re: [Haskell-cafe] GHCI Feature Request: Last Successful Compilation State Saved
On 30 August 2011 06:46, Albert Y. C. Lai tre...@vex.net wrote: In #haskell, we came up with the idea of running two instances of ghci. Try loading in one instance first; if good, commit to the other instance too; if bad, you still have the other instance in a functional state. Presumably, this should be automated by some middleware. See my post from two days ago. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] GHC API question
Hi, This is what I've was referring to in my previous mail. Even though you're compiling to machine code, you are using the in-memory linker (i.e., the GHCi linker). It seems like that this is a fundamental limitation of the internal linker. You may be using it in a way that doesn't trigger the sanity check and end up causing a panic. I had similar problems when trying to profile an executable that was using certain parts of the GHC API [1]. If you are using parts of the GHC API that are related to the GHCi profiling API bug [2], then you will get errors/segfaults. AFAIK, a better error message was never implemented for GHC API users, only for building GHCi in profiling mode. -- Christiaan Baaij [1] http://hackage.haskell.org/trac/ghc/ticket/3285 [2] http://hackage.haskell.org/trac/ghc/ticket/2197 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] bitSize
That's reasonably believable – streaming units on current CPUs can execute
multiple floating point operations per cycle.
if (*ra4 != 0xffc78948) { return false; }
On 30 Aug 2011, at 02:30, Richard O'Keefe wrote:
On 29/08/2011, at 10:32 PM, Maciej Marcin Piechotka wrote:
According to random side (http://gruntthepeon.free.fr/ssemath/) not so
new computers can compute 15.5 milions of serial logarithms per second
(62 millions in total). I'd say that overhead of Integer might be much
bigger then cost of logarithm.
That's floating-point logarithms, not Integer logarithms.
Single-precision floats, at that.
The code in question does not link at optimisation level 4.
At least some of the benchmark results are impossible to believe:
benching cephes_sinf ..
- 12762.3 millions of vector evaluations/second
- 0 cycles/value on a 2000MHz computer
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] GHCI Feature Request: Last Successful Compilation State Saved
Oh, I didn't reply all. Whoops. On 28 August 2011 13:40, Christopher Done chrisd...@googlemail.com wrote: On 27 August 2011 22:59, aditya siram aditya.si...@gmail.com wrote: I would like for the GHCI interpreter to save its environment before reloading a file and allowed the user to revert back to that state if the compilation was unsuccessful. Many times I've changed files, loaded them, hit a compilation error and needed, for example, the inferred type signature of some function. Even though that function hasn't been changed I have to either fix my code, undo a bunch of changes or comment out the last change I made and reload in order to do a :type on the function. This really breaks the flow of development. This has been bugging me for a long time and it's been on my TODO list for haskell-emacs: https://github.com/chrisdone/haskell-emacs I just implemented it. http://i.imgur.com/A71T5.png I just run two ghci processes, one for compiling, one for merely loading the code in when the former succeeds. It's not a very pleasant hack but in conjunction with :set -fobject-code it is fast and does not consume that much more memory. Considering tibbe's report is 4 years old, I think this solution is better than nothing. Set hs-config-preliminary-load-file to t in hs-config.el and it will do a preliminary load in a separate GHCi process. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Decompressing and http-enumerator
On Mon, 2011-08-29 at 16:50 +0100, Colin Adams wrote: On 29 August 2011 16:45, Michael Snoyman mich...@snoyman.com wrote: chunked is the only valid transfer-encoding[1], while gzip must be specified on the content-encoding header[2]. For a simple example of these two, look at the response headers from Haskellers[3] in something like Chrome developer tools. [1] http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.6 [2] http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.5 [3] http://www.haskellers.com/ Well [1] contradicts your claim: The Internet Assigned Numbers Authority (IANA) acts as a registry for transfer-coding value tokens. Initially, the registry contains the following tokens: chunked (section 3.6.1), identity (section 3.6.2), gzip (section 3.5), compress (section 3.5), and deflate (section 3.5). This added paragraph was actually one of the major changes between RFC2068 and RFC2616, in order to enhance the Transfer-Encoding field to become full-fledged as the Content-Encoding field was, as the previous definitions were deemed to have significant problems.[1] So, using 'Content-Encoding: gzip' was required before RFC2616... but w/ RFC2616 'Transfer-Encoding: gzip, chunked' becomes possible too... It's also been a long standing bug/deficiency in Firefox[2]. So, RFC2616 differentiates between the 'entity' (= the actual information content, mostly end-to-end) and the 'message' (= transfer wire format, mostly hop-by-hop). The content-* headers pertain to the 'entity', whereas the transfer-* headers describe the 'message' format. But I have to admit, that we have an unfortunate situation here, where the real-world implementations of RFC2616 don't follow the intent of the RFC (or alternatively: they implement the obsolete RFC2068). And I'm well aware that a 100% compliant RFC2616 implementation most likely will cause a lot of trouble in terms of interoperability... maybe the HTTPbis[3] effort will improve on the situation... Long story short: HTTP client libraries should be flexible enough to let the client-code decide whether transparent decoding of the specified Content-Encoding is desired or not. [1]: http://www.w3.org/Protocols/rfc2616/rfc2616-sec19.html#sec19.6.3 [2]: https://bugzilla.mozilla.org/show_bug.cgi?id=68517 [3]: http://tools.ietf.org/wg/httpbis/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Pointed, but not Applicative
Tony Morris wrote: Pointed f = Pointed (StateT s f) but not Applicative f = Applicative (StateT s f) I see. So StateT cannot be what you could call an applicative transformer. (Unlike for example ReaderT.) Thanks. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] [CUFP 2011] Birds of a Feather sessions
= Birds of a Feather sessions (BOFs) Commercial Users of Functional Programming Workshop (CUFP 2011) http://cufp.org/bofs-2011 Tokyo, Japan, September 22 - 23 = Following last year's success, CUFP 2011 will again include Birds of a Feather sessions. BoFs provide a place for our community to gather informally at CUFP and reach consensus on matters of importance. Any CUFP attendee can propose a BoF session, and grab one of the spare rooms to hold a discussion. Attendance in the evening is open to all. BoF sessions facilitate ad-hoc discussions and provide a place to gather and start off the chat, before moving on during the evening to a local restaurant or pub. There is no deadline for proposing a session, but we encourage you to do so at the earliest possible time to help us with planning and to publicize your proposed discussion. Attendees are also free to show up even if you have not confirmed, but again earlier confirmation will help us with planning. Please see http://cufp.org/bofs-2011 for details on how to propose a BoF session or confirm attendance at one of them. Scheduled BoFs will be listed here as they are confirmed. General information on CUFP 2011 can be found at http://cufp.org/conference/2011. The conference is co-located with ICFP. You can register for CUFP at the main ICFP website http://www.icfpconference.org/icfp2011. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] installing/reinstalling haskell platform with profiling enabled on windows
How can I install/reinstall haskell platform with profiling enabled on windows? Or the other way to put it, how can I install/reinstall haskell platform on windows with my own cabal options? I googled on it and didn't find any solutions for windows OS. Thanks, Malik ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] GHCI Feature Request: Last Successful Compilation State Saved
Chris, this is amazing! One question - do you have support for tramping in and running ghci remotely (or would that be easy to add)? I primarily develop inside virtual-machines, so while the source code lives where emacs lives, all the libraries are installed inside the virtual-machine (and the source also lives there, of course). On Aug 30, 2011, at 4:46 AM, Christopher Done wrote: Oh, I didn't reply all. Whoops. On 28 August 2011 13:40, Christopher Done chrisd...@googlemail.com wrote: On 27 August 2011 22:59, aditya siram aditya.si...@gmail.com wrote: I would like for the GHCI interpreter to save its environment before reloading a file and allowed the user to revert back to that state if the compilation was unsuccessful. Many times I've changed files, loaded them, hit a compilation error and needed, for example, the inferred type signature of some function. Even though that function hasn't been changed I have to either fix my code, undo a bunch of changes or comment out the last change I made and reload in order to do a :type on the function. This really breaks the flow of development. This has been bugging me for a long time and it's been on my TODO list for haskell-emacs: https://github.com/chrisdone/haskell-emacs I just implemented it. http://i.imgur.com/A71T5.png I just run two ghci processes, one for compiling, one for merely loading the code in when the former succeeds. It's not a very pleasant hack but in conjunction with :set -fobject-code it is fast and does not consume that much more memory. Considering tibbe's report is 4 years old, I think this solution is better than nothing. Set hs-config-preliminary-load-file to t in hs-config.el and it will do a preliminary load in a separate GHCi process. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Pointed, but not Applicative
I suspect this definition is what Sebastian meant by converting back and forth to ordinary lists. 2011/8/29 Ryan Ingram ryani.s...@gmail.com On Sun, Aug 28, 2011 at 8:24 PM, Maciej Marcin Piechotka uzytkown...@gmail.com wrote: f `fmap` FList g = _|_ f `fmap` FList g = map id f `fmap` FList g = map _|_ (+ variation of _|_*) f `fmap` FList g = \bs - map f (g []) ++ bs ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] GHCI Feature Request: Last Successful Compilation State Saved
On 30 August 2011 17:48, Daniel Patterson lists.hask...@dbp.mm.st wrote: Chris, this is amazing! One question - do you have support for tramping in and running ghci remotely (or would that be easy to add)? I primarily develop inside virtual-machines, so while the source code lives where emacs lives, all the libraries are installed inside the virtual-machine (and the source also lives there, of course). So either 1) You put all your source files on the server, too and then run the GHCi process under tramp and access the files via tramp. 2) Sounds like you'd prefer to keep your files local, which, I suppose means you want to send code to the server when you trigger a compile? Could be done, could be done. I'll ponder it. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Project Euler: request for comments
You might like this zipping folding version. Explicit recursion has the disadvantage that one has to read the entire function in order to figure out what's going on; whereas using the higher order functions makes things much easier to grasp. listof4tuples xs = (zip4 xs (tail xs) (tail (tail xs)) (tail (tail (tail xs prods xs = map prods4 (listof4tuples xs) prods4 (t,u,v,w) = t*u*v*w maxprods4 xs = maximum $ prods xs On Mon, Aug 29, 2011 at 9:40 AM, Oscar Picasso oscarpica...@gmail.com wrote: Got it. f :: [Int] - Int f (t:u:v:xs) = helper t u v xs helper :: Int - Int - Int - [Int] - Int helper t u v (w:ws) | ws == [] = t*u*v*w | otherwise = max (t*u*v*w) (f (u:v:w:ws)) I tend to overlook mutual recursion in my toolbox. Thanks for the nnlightenment. On Sun, Aug 28, 2011 at 4:54 PM, KC kc1...@gmail.com wrote: Try something like the following: -- Project Euler 11 -- In the 20×20 grid below, four numbers along a diagonal line have been marked in red. -- snip -- The product of these numbers is 26 × 63 × 78 × 14 = 1788696. -- What is the greatest product of four adjacent numbers in any direction (up, down, left, right, or diagonally) in the 20×20 grid? import Data.List -- Doing the one dimensional case. f011 :: [Int] - Int f011 (t:u:v:xs) = f011helper t u v xs f011helper :: Int - Int - Int - [Int] - Int f011helper t u v (w:ws) | ws == [] = t*u*v*w | otherwise = yada nada mada -- What are yada nada mada? -- The 20x20 grid case will become: f0112D :: [[Int]] - Int -- where [[Int]] is a list of lists of rows, columns, major diagonals, minor diagonals. On Sun, Aug 28, 2011 at 5:10 AM, Oscar Picasso oscarpica...@gmail.com wrote: No. The answer I posted is not good. It worked, by chance, on a couple of small examples I tried but it could end up comparing sequence of 4 numbers that where not initially adjacent. On Sun, Aug 28, 2011 at 12:32 AM, Oscar Picasso oscarpica...@gmail.com wrote: Maybe this? f x@(a:b:c:d:[]) = x f (a:b:c:d:e:ys) = if e = a then f (b:c:d:e:ys) else f (a:b:c:d:ys) On Sat, Aug 27, 2011 at 8:26 PM, KC kc1...@gmail.com wrote: Think of the simplest version of the problem that isn't totally trivial. e.g. A one dimensional list of numbers. What would you do? Note: you only want to touch each element once. The 2 dimensional case could be handled by putting into lists: rows, columns, major diagonals, and minor diagonals. This isn't the fastest way of doing the problem but it has the advantage of avoiding indexitis. On Fri, Aug 26, 2011 at 6:15 PM, Oscar Picasso oscarpica...@gmail.com wrote: Like: 20*19*21*18 is bigger than 100*100*3*2 ? If so I need to think about how to formalize it. Thanks for the hint. On Fri, Aug 26, 2011 at 8:55 PM, KC kc1...@gmail.com wrote: Is Problem 11 the 4 consecutive #'s problem? If so what must be true for 4 #'s to have a large product? Hint: x * y * z * 2 is that going to be larger? -- -- Regards, KC -- -- Regards, KC -- -- Regards, KC -- -- Regards, KC ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Ambiguous module name `System.Directory'
Note that once you upgrade it (to =0.4), you'll still need to remove the older version to fix the error. I wish cabal-install defaulted to hiding every package it installs. The current behavior of exposing every installed module is unreasonable and confusing. Packages should be namespaces, not just installation aliases. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Off-topic: Mathematics
On 29/08/2011 01:13 PM, Christopher Done wrote: There's also #math on freenode, but it's a scary wilderness. On 29 August 2011 13:34, Benedict Eastaughionf...@gmail.com wrote: On 29 August 2011 09:34, Andrew Coppinandrewcop...@btinternet.com wrote: This is fairly wildly off-topic but... does anybody know of a good forum where I can ask questions about mathematics and get authoritative answers? Apart from math.stackexchange.com and mathoverflow.net, which people have already mentioned, people often discuss mathematics on #haskell-blah on Freenode. I know of several places where I can ask maths questions and half a dozen people will take guesses at what the correct solution might be. I haven't yet found anywhere where I can say when would a chi-squared test be more appropriate than a KS test? and get an informed, knowledgeable answer. (Answers from people who /know/ what they're talking about rather than just /think/ they know.) Anyway, from this thread I've got the names of a few places to start looking. And that's really what I was hoping for. Thanks for the tips. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Off-topic: Mathematics and modesty
Andrew Coppin : I haven't yet found anywhere where I can say when would a chi-squared test be more appropriate than a KS test? and get an informed, knowledgeable answer. (Answers from people who /know/ what they're talking about rather than just /think/ they know.) With all my respect: I think I know several mathematicians who learning that a person asking for help begins with trying to distinguish between knowledgeable, and those who just think they are, will simply - to say it politely - refuse to engage. (There might be some impolite answers as well, but you don't need the details). Jerzy Karczmarczuk ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Off-topic: Mathematics and modesty
On Tue, 2011-08-30 at 20:58 +0200, Jerzy Karczmarczuk wrote: With all my respect: I think I know several mathematicians who learning that a person asking for help begins with trying to distinguish between knowledgeable, and those who just think they are, will simply - to say it politely - refuse to engage. I don't agree with this. It's the most natural thing in the world to listen to an answer and then try to figure out whether the speaker knows what they are talking about or not. Those who expect us to forego that step aren't really engaged in mathematics any more. -- Chris Smith ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Text.Parsec.Prim
On 11-08-29 12:35 PM, Roly Perera wrote: Cool. I assumed that since I already had Parsec installed as part of the Haskell Platform it wouldn't have made any difference, but actually it sorted it. Thanks a lot! The combination of ghc 6.12.1 and haskell platform suggests it is a pretty old haskell platform, whose parsec is version 2. It does not have ParsecT. Your recent cabal install parsec gets you version 3. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Off-topic: Mathematics and modesty
He said Trying to distinguish first between knowledgeable, wise, guru, wizard, etc. before asking the actual mathematics related question. Beware: the expert X is an unknown quantity spert is a drip under pressure :D On Tue, Aug 30, 2011 at 12:16 PM, Chris Smith cdsm...@gmail.com wrote: On Tue, 2011-08-30 at 20:58 +0200, Jerzy Karczmarczuk wrote: With all my respect: I think I know several mathematicians who learning that a person asking for help begins with trying to distinguish between knowledgeable, and those who just think they are, will simply - to say it politely - refuse to engage. I don't agree with this. It's the most natural thing in the world to listen to an answer and then try to figure out whether the speaker knows what they are talking about or not. Those who expect us to forego that step aren't really engaged in mathematics any more. -- Chris Smith ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe -- -- Regards, KC ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] installing/reinstalling haskell platform with profiling enabled on windows
On 11-08-30 10:52 AM, MH wrote: How can I install/reinstall haskell platform with profiling enabled on windows? It already is. Program Files (x86)\Haskell Platform\2011.2.0.1\lib\extralibs\text-0.11.0.6\ghc-7.0.3\libHStext-0.11.0.6_p.a Program Files (x86)\Haskell Platform\2011.2.0.1\lib\extralibs\text-0.11.0.6\ghc-7.0.3\Data\Text.p_hi are just a few examples. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Off-topic: Mathematics
Hi Andrew, I know of several places where I can ask maths questions and half a dozen people will take guesses at what the correct solution might be. I haven't yet found anywhere where I can say when would a chi-squared test be more appropriate than a KS test? and get an informed, knowledgeable answer. (Answers from people who /know/ what they're talking about rather than just /think/ they know.) I believe this phenomenon is quite natural and easily explained. When you're asking a non-trivial question, hardly anyone just knows the correct answer -- especially when it comes to math. In order to answer your question, people have to dedicate time and effort to study the problem you're asking about. (Furthermore, formulating a coherent response is usually be a bit of an effort, too.) Now, a person who has profound knowledge of the subject you're asking about is not very likely to do this, because he is probably not going to learn anything in the process. Dedicating time and effort to studying your particular problem is not an appealing prospect. A person who has superficial understanding of the subject, however, is more likely to be fascinated by the problem, and consequently he is more likely to dedicate time and effort into formulating a response. In other words, even if Donald Knuth himself is reading the forum you're posting to, it doesn't mean that he is actually going to respond. On the other hand, if you're asking the right question, Donald Knuth just might respond to it, but not necessarily in the forum that you were originally asking in. Take care, Peter ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Off-topic: Mathematics
Peter Simons : Now, a person who has profound knowledge of the subject you're asking about is not very likely to do this, because he is probably not going to learn anything in the process. Dedicating time and effort to studying your particular problem is not an appealing prospect. You might be right. You may be dead wrong... Some people like to teach, find some pleasure in explaining things they know, and find a pleasure in trying to be useful. But they might refuse, if they hear: ...if you are a real specialist and you know what you are talking about, tell me what is better, the chi-square test or (...). Knuth admitted that he had learnt a lot while teaching things he already knew. So did Feynman. And Landau. All the best. Jerzy K. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Pointed, but not Applicative
On Tue, Aug 30, 2011 at 9:42 AM, Conal Elliott co...@conal.net wrote: I suspect this definition is what Sebastian meant by converting back and forth to ordinary lists. Yep, I know; and technically it violates 'fmap id' == 'id' for example, fmap id (FList $ \xs - xs ++ xs) = FList $ \xs - xs If you add this FList law, though, you're OK: runFList fl as = runFList fl [] ++ as But, yes, this definition of fmap converts back to an ordinary list representation. -- ryan ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Project Euler: request for comments
You also don't need mutual recursion for this explicit recursion since I imagine it would use up more stack space. -- Doing the one dimensional case. f011 :: [Int] - Int f011 (t:u:v:xs) = f011helper t u v xs f011helper :: Int - Int - Int - [Int] - Int f011helper t u v (w:ws) | ws == [] = t*u*v*w | otherwise = max (t*u*v*w) (f011helper u v w ws) -- Note: f011helper does not call f011. On Mon, Aug 29, 2011 at 9:40 AM, Oscar Picasso oscarpica...@gmail.com wrote: Got it. f :: [Int] - Int f (t:u:v:xs) = helper t u v xs helper :: Int - Int - Int - [Int] - Int helper t u v (w:ws) | ws == [] = t*u*v*w | otherwise = max (t*u*v*w) (f (u:v:w:ws)) I tend to overlook mutual recursion in my toolbox. Good going! :) Thanks for the enlightenment. On Sun, Aug 28, 2011 at 4:54 PM, KC kc1...@gmail.com wrote: Try something like the following: -- Project Euler 11 -- In the 20×20 grid below, four numbers along a diagonal line have been marked in red. -- snip -- The product of these numbers is 26 × 63 × 78 × 14 = 1788696. -- What is the greatest product of four adjacent numbers in any direction (up, down, left, right, or diagonally) in the 20×20 grid? import Data.List -- Doing the one dimensional case. f011 :: [Int] - Int f011 (t:u:v:xs) = f011helper t u v xs f011helper :: Int - Int - Int - [Int] - Int f011helper t u v (w:ws) | ws == [] = t*u*v*w | otherwise = yada nada mada -- What are yada nada mada? -- The 20x20 grid case will become: f0112D :: [[Int]] - Int -- where [[Int]] is a list of lists of rows, columns, major diagonals, minor diagonals. -- -- Regards, KC ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Off-topic: Mathematics
Use Usenet. sci.math, sci.math.research, and sci.logic are some of the best mathematics and logic resources on the internet. On Mon, Aug 29, 2011 at 1:34 AM, Andrew Coppin andrewcop...@btinternet.comwrote: This is fairly wildly off-topic but... does anybody know of a good forum where I can ask questions about mathematics and get authoritative answers? (Apart from go visit the nearest university, that is.) __**_ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/**mailman/listinfo/haskell-cafehttp://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Pointed, but not Applicative
On Wed, Aug 31, 2011 at 6:13 AM, Ryan Ingram ryani.s...@gmail.com wrote: technically it violates 'fmap id' == 'id' [...] If you add this FList law, though, you're OK: runFList fl as = runFList fl [] ++ as I think the idea of functional lists is that the monoids of 'lists' and 'functions on lists' are isomorphic with isomorphisms toFList and toList: toFList [] = id toFList (xs++ys) = toFList xs . toFList ys toList id = [] toList (f . g) = toList f ++ toList g These can be defined as: toFList = (++) toList = ($[]) Eliding newtypes, runFList is just the identity function so we need to check f l = toList f ++ l to verify your law. This is the same as f = toFList (toList f) which (once we know that toList and toFList are isomorphisms) does indeed hold because: toFList . toList = id toList . toFList = id Sebastian ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Pointed, but not Applicative
toFList [] = id toFList (xs++ys) = toFList xs . toFList ys toList id = [] toList (f . g) = toList f ++ toList g These laws do not *define* the isomorphisms because their behavior on singletons is not fixed. Combining them with laws using a 'point' function for functional lists point = (:) the isomorphisms are characterized uniquely: toFList [x] = point x toList (point x) = [x] This might be an argument in favor of a Pointed class without Functor constraint as it shows that other pointed structures have reasonable laws involving 'point' as well. Sebastian ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] GHCI Feature Request: Last Successful Compilation State Saved
On Aug 30, 2011, at 1:15 PM, Christopher Done wrote: 1) You put all your source files on the server, too and then run the GHCi process under tramp and access the files via tramp. 2) Sounds like you'd prefer to keep your files local, which, I suppose means you want to send code to the server when you trigger a compile? Could be done, could be done. I'll ponder it. Actually, 1. is what happens, which seems like the easier thing. The folder does exist locally (ie, on the host), but it is shared with the virtualmachine, so it can be edited in either place. The sharing is either done through a a special mechanism provided by the virtualmachine (ie, in virtualbox, shared folders), or with plain old NFS (the latter is faster, for unknown reasons). My normal workflow is to have an editor open that edits the files on the host. Then with a ssh connection to the client (the virtualmachine), I build and/or run ghci through that. I'm not sure if this is how other people do work with virtualmachines, but I would imagine it was relatively common. I'd be pretty excited about that working. As an added benefit, the same strategy could be used to jump into a project on an actually remote host. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] bitSize
On 30/08/2011, at 7:45 PM, Thomas Davie wrote:
That's reasonably believable – streaming units on current CPUs can execute
multiple floating point operations per cycle.
The figures for cephes_{sinf,cosf} are difficult to believe
because they are so extremely at variance with the figures that
come with the software.
First off, the program as delivered, when compiled, reported that the
computer was a 2000 MHz one. It is in fact a 2.66 GHz one. That
figure turns out to be a #define in the code. Fix that, and the
report includes
benching cephes_sinf .. - 12779.4 millions of vector
evaluations/second
- 0 cycles/value on a 2660MHz computer
benching cephes_cosf .. - 12756.8 millions of vector
evaluations/second
- 0 cycles/value on a 2660MHz computer
benching cephes_expf .. -7.7 millions of vector evaluations/second
- 86 cycles/value on a 2660MHz computer
The internal documentation in the program claims the following results
on a 2.4 GHz machine:
benching cephes_sinf .. - 11.6 millions of vector evaluations/second
- 56 cycles/value on a 2600MHz computer
benching cephes_cosf .. - 8.7 millions of vector evaluations/second
- 74 cycles/value on a 2600MHz computer
benching cephes_expf .. - 3.7 millions of vector evaluations/second
- 172 cycles/value on a 2600MHz computer
It seems surpassing strange that code compiled by gcc 4.2 on a 2.66 GHz
machine should run more than a thousand times faster than code compiled
by gcc 4.2 on a 2.60 GHz machine with essentially the same architecture.
Especially as those particular functions are *NOT* vectorised.
They are foils for comparison with the functions that *ARE* vectorised,
for which entirely credible 26.5 million vector evaluations per second
(or 25 cycles per value) are reported. Considering that sin and cos
are not single floating point operations, 25 cycles per value is well
done.
28 cycles per single-precision logarithm is not too shabby either,
IF one can trust a benchmark that has blown its credibility as badly
as this one has. But it's still not an Integer logarithm, just a
single precision floating point one.
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] iOS and Haskell
Hello, I know nothing about the architecture of iOS running on an iPad. However, I have a friend who just bought the newer iPad. It seems that either C++ or Objective C is available on iPad for development(actually the icon says C++). What are the prospects of getting Haskell ported to iOS? Vasili ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] iOS and Haskell
On Wed, Aug 31, 2011 at 00:32, Vasili I. Galchin vigalc...@gmail.comwrote: I know nothing about the architecture of iOS running on an iPad. However, I have a friend who just bought the newer iPad. It seems that either C++ or Objective C is available on iPad for development(actually the icon says C++). What are the prospects of getting Haskell ported to iOS? http://www.haskell.org/haskellwiki/IPhone is a compendium of discussion and work on the subject. I don't know the current state of it. -- brandon s allbery allber...@gmail.com wandering unix systems administrator (available) (412) 475-9364 vm/sms ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
