[Haskell-cafe] Parsec in Haskell platform
Since Parsec 2 and Parsec 3 are usually considered to be different packages, it's ambiguous to say that 'parsec' is included in Haskell Platform [1]. Could someone please: 1. Clarify which version(s) are included? 2. Fix the page [1] It would be convenient to have a page which would list all the HP packages with their versions. The release page [2] only has a list of packages whose versions has changed since the last release, as I understood. [1] http://hackage.haskell.org/platform/contents.html [2] http://hackage.haskell.org/platform/changelog.html -- Roman I. Cheplyaka :: http://ro-che.info/ Don't let school get in the way of your education. - Mark Twain ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Re: Splay tree in a pattern matching way
Hi, I just tried a smoke test case, seems the tree is balanced enough: import System.Random lookup :: (Ord a) = STree a - a - STree a lookup E _ = E lookup t@(Node l x r) y | x == y= t | x y = splay (Node (lookup l y) x r) y | otherwise = splay (Node l x (lookup r y)) y testSplay = do xs - sequence (replicate 100 (randomRIO(1, 10))) putStrLn $ show (foldl lookup t xs) where t = foldl insert (E::STree Int) [1..10] Where STree a and insert are defined as in my previous email, except I changed `xy' to `xy' (sorry for the mistake). By running the testSplay I got a BST like this: *SplayHeap testSplay Node (Node E 1 (Node (Node E 2 E) 3 (Node (Node E 4 (Node E 5 E)) 6 (Node E 7 E 8 (Node (Node E 9 E) 10 E) So I got a bit more confident about the correctness of my code. Best regards -- Larry, LIU Xinyu http://sites.google.com/site/algoxy On Oct 24, 11:06 am, Xinyu LIU liuxiny...@gmail.com wrote: Hi, I checked the Hackage Splay Tree implementation from.http://hackage.haskell.org/packages/archive/TreeStructures/0.0.1/doc/... Basically it's based on the strategy mentioned in Okasaki's ``Purely Functional Programming'', Chapter 5.4. That in case we traverse twice in left (or right), we rotate the tree, so in long term of view, the tree turns more and more balanced. There are 3 cases for splay: zig-zig case, zig-zag case and zig case according tohttp://en.wikipedia.org/wiki/Splay_tree So I wonder if Splay tree can also be implemented by using pattern matching in a similar way as red black tree. Here is a draft source code: data STree a = E -- Empty | Node (STree a) a (STree a) -- left, element, right deriving (Eq, Show) -- splay by pattern matching splay :: (Eq a) = STree a - a -STree a -- zig-zig splay t@(Node (Node (Node a x b) p c) g d) y = if x == y then Node a x (Node b p (Node c g d)) else t splay t@(Node a g (Node b p (Node c x d))) y = if x == y then Node (Node (Node a g b) p c) x d else t -- zig-zag splay t@(Node (Node a p (Node b x c)) g d) y = if x == y then Node (Node a p b) x (Node c g d) else t splay t@(Node a g (Node (Node b x c) p d)) y = if x == y then Node (Node a g b) x (Node c p d) else t -- zig splay t@(Node (Node a x b) p c) y = if x == y then Node a x (Node b p c) else t splay t@(Node a p (Node b x c)) y = if x == y then Node (Node a p b) x c else t -- otherwise splay t _ = t -- insert by using pattern matching insert :: (Ord a) = STree a - a - STree a insert E y = Node E y E insert (Node l x r) y | x y = splay (Node (insert l y) x r) y | otherwise = splay (Node l x (insert r y)) y I am not quite sure if the solution is correct and what about the efficiency of this code. Regards. -- Larry. LIU Xinyuhttp://sites.google.com/site/algoxyhttp://sites.google.com/site/algoxy/home ___ Haskell-Cafe mailing list haskell-c...@haskell.orghttp://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] concurrency vs. I/O in GHC
On 23/10/10 23:17, Donn Cave wrote:
Quoth Claude Heiland-Allenclaudiusmaxi...@goto10.org,
...
The conclusion I drew was that unsafe foreign functions block the
current capability (OS thread) and any threads (Haskell forkIO etc)
currently scheduled on that capability, but other capabilities and
threads continue executing as normal.
... until GC time when all capabilities must be ready. (?)
If a trivial test program would help, here I call the sleep() function,
which I believe on a POSIX platform suspends the thread until receipt
of a SIGALRM.
I wrote a program which shows some interesting behaviour:
8
{-# LANGUAGE ForeignFunctionInterface #-}
module Main (main) where
import GHC.Conc (forkOnIO, numCapabilities)
import Control.Concurrent (threadDelay)
import Foreign.C (CInt)
import System.Environment (getArgs)
foreign import ccall unsafe sleep sleep :: CInt - IO CInt
delayer :: Int - IO ()
delayer n = do
print (delayer, n)
threadDelay 10 -- 10Hz
delayer n
sleeper :: Int - IO ()
sleeper n = do
print (sleeper, n)
_ - sleep 1 -- 1Hz
sleeper n
main :: IO ()
main = do
m - (read . head) `fmap` getArgs
mapM_ (\n - forkOnIO n $ delayer n) [1 .. numCapabilities]
mapM_ (\n - forkOnIO n $ sleeper n) [1 .. numCapabilities - m]
threadDelay 1 -- 100s
8
$ ghc --version
The Glorious Glasgow Haskell Compilation System, version 6.12.3
$ uname -a
Linux zebimus 2.6.32-25-generic #44-Ubuntu SMP Fri Sep 17 20:05:27 UTC
2010 x86_64 GNU/Linux
$ ghc -O2 -Wall -threaded --make DelayedSleep.hs
$ ./DelayedSleep +RTS -N4 -S -RTS 3
[snip]
8
By interesting I mean there is lots of output from the delayer threads
on capabilities without sleeper threads (as you would expect), with the
delayer threads on capabilities also having sleeper threads being much
less frequent (as you might also expect). But then there are some long
pauses where there is no output from any thread: my hypothesis is that
the whole runtime is blocked waiting for all threads to be ready for GC
(because +RTS -S shows some GC stats after the end of those pauses).
Claude
--
http://claudiusmaximus.goto10.org
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Parsec in Haskell platform
On 24/10/2010 09:10 AM, Roman Cheplyaka wrote: It would be convenient to have a page which would list all the HP packages with their versions. The release page [2] only has a list of packages whose versions has changed since the last release, as I understood. It would be nice to have a page that lists everything included in every HP release, together with their version numbers. (So that, e.g., I can see at a glance what version of GHC, Haddock or cabal-install is in HP-2009.1.0.0.) All this information must exist somewhere, it's just not easily viewable on the web. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Parsec in Haskell platform
On Sun, Oct 24, 2010 at 11:57 AM, Andrew Coppin andrewcop...@btinternet.com wrote: It would be nice to have a page that lists everything included in every HP release, together with their version numbers. (So that, e.g., I can see at a glance what version of GHC, Haddock or cabal-install is in HP-2009.1.0.0.) All this information must exist somewhere, it's just not easily viewable on the web. http://hackage.haskell.org/platform/changelog.html There's a Cabal file somewhere specifying all the package in the platform (including binaries like Happy.) Johan ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Converting Values Between Lua And Haskell
Hello aditya, Sunday, October 24, 2010, 8:05:55 AM, you wrote: HsLua page is nothing more but my fantasy about future HsLua development :) you may find even XXX type where i don't found good name :) Hi all, The HsLua page [1] says that Int,Double,String,Bool,[a] and [(a,b)] types can be converted to and from Lua values. However the on hslua API page I don't see a StackValue instance [2] for [a] or [(a,b)]. Am I missing something? -deech [1] http://www.haskell.org/haskellwiki/HsLua#Exchanging_data_between_Haskell_and_Lua_worlds [2] http://hackage.haskell.org/packages/archive/hslua/latest/doc/html/Scripting-Lua.html#t:StackValue -- Best regards, Bulatmailto:bulat.zigans...@gmail.com ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Parsec in Haskell platform
On 24/10/2010 11:10 AM, Johan Tibell wrote: On Sun, Oct 24, 2010 at 11:57 AM, Andrew Coppin andrewcop...@btinternet.com wrote: It would be nice to have a page that lists everything included in every HP release, together with their version numbers. (So that, e.g., I can see at a glance what version of GHC, Haddock or cabal-install is in HP-2009.1.0.0.) All this information must exist somewhere, it's just not easily viewable on the web. http://hackage.haskell.org/platform/changelog.html There's a Cabal file somewhere specifying all the package in the platform (including binaries like Happy.) I'm pretty sure there used to be a link to the .cabal file somewhere, but now I can't seem to find it. Instead, I had to navigate to the Unix download page, download the source tarball, untar it (non-trivial under Windows), and hunt around for the .cabal file. And *then* I can find out what's in [the current release of] HP. It would be nice if there was a table on the website somewhere that I could just glance at to get the same information. [Also, regarding the page linked above... Who the hell uses strong for section heading?! Why wou-- oh, wait. It's Pandoc? I guess that also explains why we have ASCII-art arrows [-] rather than real arrows [rarr;]. And presumably this stuff is written by hand, becuase GHC is missing an arrow...] ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] ANNOUNCE: iteratee-compress 0.1.1
Iteratee-compress provides compressing and decompressing enumerators including flushing. Currently only gzip is provided but at least bzip is planned. Changes from previous version: - Independent from zlib library (Haskell one, not C) - Allow hand-flushing the contents (from outside). - Fix potential memory-leak Next goals: - BZip support - Generic interface for flushing To think about: - Should inner iteratee be able to request flushing? Regards PS. It did change API by removing dependency on zlib but I home such breakage in 0.1.x will be allowed signature.asc Description: This is a digitally signed message part ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Re: Haddock: patch to generate single-index page in addition to perl-letter indexes indices.
2010/10/24 Ryan Newton new...@mit.edu: When I encounter a split-index (A-Z) page it can be quite frustrating if I don't know the first letter of what I'm searching for. I want to use my browser find! For example, tonight I wanted to look at all the functions that END in Window in the Chart package -- no luck: http://hackage.haskell.org/packages/archive/Chart/0.13.1/doc/html/doc-index.html Therefore I propose that even when generating the A-Z individual pages that there also be an All option for the single-page version. Attached is a patch against haddock's HEAD (darcs get http://code.haskell.org/haddock/ right?) that implements this behavior. As an example, here is FGL's documentation built with the patched haddock: http://people.csail.mit.edu/newton/fgl_example_doc/doc-index.html The great thing about hackage being centralized is that if people are happy with this fix it can be widely deployed where it counts, and quickly! Patch applied. Thanks! David ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haddock: patch to generate single-index page in addition to perl-letter indexes indices.
For packages with many items in the index, these pages can get a bit huge. How about a permuted index like http://www.lispworks.com/documentation/HyperSpec/Front/X_Symbol.htm? E.g., for your use case, you would go to E and then the row with all the End entries, which would contain all the names with End anywhere in their name. I don't know if page size can be a problem, but at least for mobile or otherwise low-bandwidth devices this can be a nice alternative. On 24 October 2010 04:41, Ryan Newton new...@mit.edu wrote: When I encounter a split-index (A-Z) page it can be quite frustrating if I don't know the first letter of what I'm searching for. I want to use my browser find! For example, tonight I wanted to look at all the functions that END in Window in the Chart package -- no luck: http://hackage.haskell.org/packages/archive/Chart/0.13.1/doc/html/doc-index.html Therefore I propose that even when generating the A-Z individual pages that there also be an All option for the single-page version. Attached is a patch against haddock's HEAD (darcs get http://code.haskell.org/haddock/ right?) that implements this behavior. As an example, here is FGL's documentation built with the patched haddock: http://people.csail.mit.edu/newton/fgl_example_doc/doc-index.html The great thing about hackage being centralized is that if people are happy with this fix it can be widely deployed where it counts, and quickly! Cheers, -Ryan P.S. At the other end of the spectrum, when considering a central index for all of hackage (as in the below ticket) maybe it would be necessary to have more than 26 pages, I.e. Aa-Am | An-Az or whatever. http://hackage.haskell.org/trac/hackage/ticket/516#comment:6 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe -- Push the envelope. Watch it bend. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] running and understanding a lifting program
hi, I am trying to run and understand a lifting program from [1]. The program lifts points to moving points that vary their position over time. I made some effort to run the progrm but I do not know how to overide the +,-,*,sqr, and sqrt from the Num class. Below is my current attempt. I do not wish to change or imporve the code, rather I wish to understand it as it stands and find out what needs to be added to get the outputs shown below i.e. distance between points p1 and p2 -- 1.55 and the distance between moving points mp1 and mp2 for time 2 5.83. data Point = Point Float Float data Line = Line Point Point data Polygon = Polygon [Point] type Time = Float -- Functor Changing, which adds time parameter t to its input value. -- For example, Changing Float indicates a changing floating number (i.e. a function of time). type Changing v = Time - v -- The abstract lifting functions class Lifts a where lift0 :: a - f a lift1 :: (a - b) - f a - f b lift2 :: (a - b - c) - f a - f b - f c -- Not too sure about this, instance Lifts Changing Float where lift0 a = \t - a lift1 op a = \t - op (a t) lift2 op a b = \t - op (a t) op (b t) class Number a where (+), (-), (*) :: a - a - a sqr, sqrt :: a - a sqr a = a * a -- The class point which support vector plus and minus as well as -- distance operation is defined as follow class Number s = Points p s where x, y :: p s - s x (Point x1 y1) = x1 y (Point x1 y1) = y1 (+), (-) :: p s - p s - p s (+) a b = Point (x a + x b) (y a + y b) (-) a b = Point (x a - x b) (y a - y b) dist :: p s - p s - s dist a b = sqrt(sqr((x a)-(x b))+sqr((y a)-(y b))) -- The lifting the operations for numbers shoukd provide a distance -- function which can be used for both static and moving points: instance Number v = Number (Changing v) where (+) = lift2 (+) (-) = lift2 (-) (*) = lift2 (*) sqrt = lift1 (sqrt) -- Running the code -- If p1 and p2 are two 2D static points, -- their distance d is calculated as follows: p1, p2 :: Point Float p1 = Point 3.4 5.5 p2 = Point 4.5 4.5 -- distance between p1 and p2 -- 1.55 d = dist p1 p2 -- For 2D moving points mp1 and mp2, their distance md, -- which is a function of time, is calculated as follows: mp1, mp2 :: Point (Changing Float) mp1 = Point (\t - 4.0 + 0.5 * t)(\t - 4.0 - 0.5 * t) mp2 = Point (\t - 0.0 + 1.0 * t)(\t - 0.0 - 1.0 * t) -- distance between mp1 and mp2 md = dist mp1 mp2 -- distance md for time 2 5.83 md 2 [1] A Mathematical Tool to Extend 2D Spatial Operations to Higher Dimensions: by Farid Karimipour1,2, Mahmoud R. Delavar1, and Andrew U. Frank2 http://books.google.ie/books?id=JUGpGN_jwf0Cpg=PA153lpg=PA153dq=Karimipour+%22A+Mathematical+Tool+to+Extend+2D+Spatial+Operations+to+Higher+Dimensions%22source=blots=fu-lSkPMr3sig=ztkcRV3Cv6hn9T6iwQCJ9sB75IMhl=enei=QS7ETJHPGoiA5Ab0zZW6Awsa=Xoi=book_resultct=resultresnum=4ved=0CCMQ6AEwAw#v=onepageq=Karimipour%20%22A%20Mathematical%20Tool%20to%20Extend%202D%20Spatial%20Operations%20to%20Higher%20Dimensions%22f=false This message has been scanned for content and viruses by the DIT Information Services E-Mail Scanning Service, and is believed to be clean. http://www.dit.ie ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Converting Values Between Lua And Haskell
I was fooled :). Some indication of that on the page would be very helpful. -deech On Sun, Oct 24, 2010 at 5:29 AM, Bulat Ziganshin bulat.zigans...@gmail.comwrote: Hello aditya, Sunday, October 24, 2010, 8:05:55 AM, you wrote: HsLua page is nothing more but my fantasy about future HsLua development :) you may find even XXX type where i don't found good name :) Hi all, The HsLua page [1] says that Int,Double,String,Bool,[a] and [(a,b)] types can be converted to and from Lua values. However the on hslua API page I don't see a StackValue instance [2] for [a] or [(a,b)]. Am I missing something? -deech [1] http://www.haskell.org/haskellwiki/HsLua#Exchanging_data_between_Haskell_and_Lua_worlds [2] http://hackage.haskell.org/packages/archive/hslua/latest/doc/html/Scripting-Lua.html#t:StackValue -- Best regards, Bulatmailto:bulat.zigans...@gmail.com ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Parsec in Haskell platform
It would be convenient to have a page which would list all the HP packages with their versions. The release page [2] only has a list of packages whose versions has changed since the last release, as I understood. It would be nice to have a page that lists everything included in every HP release, together with their version numbers. (So that, e.g., I can see at a glance what version of GHC, Haddock or cabal-install is in HP-2009.1.0.0.) All this information must exist somewhere, it's just not easily viewable on the web. Yes, I needed the same information recently. I ended up reading the Cabal file at [1]. A convenient way to see what packages are included in what version of the platform would still be very useful, though. Cheers, Simon [1] http://code.haskell.org/haskell-platform/haskell-platform.cabal ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] back doors into the IO monad
On Sun, 24 Oct 2010 00:28:37 +0200, Manlio Perillo manlio_peri...@libero.it wrote: Hi. What are the available methods to execute IO actions from pure code? I know only unsafePerformIO and foreign import (to call a non pure foreign function). unsafeCoerce is a back door for almost everything. -- Nicolas Pouillard http://nicolaspouillard.fr ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haddock: patch to generate single-index page in addition to perl-letter indexes indices.
(Btw, I blame gmail for the mangled title ;-) -- it's been doing some weird stuff recently on Safari for me; but only in the subject line. I think there were backspace characters in an edit that weren't applied.) The permuted indices are interesting. It means really committing to the style/naming conventions, doesn't it? In common lisp, hyphen separated names and I guess camel-case for Haskell. Though you could split on underscores *or* camel case... I think multiple ways of indexing the data never hurts (except by confusing the user a bit). On that common lisp page I especially like how they've indented the words. Frankly I'd like a search box in any interface that displays more than two thingamajigs. That should be a UI commandment. I was expecting the objection of wasted server bandwidth for very large indices. I wasn't so worried about the client (even mobile) case. People can always press escape if a load takes too long. And it only happens if they manually drill down into All. Perhaps a good idea would be to follow the convention used elsewhere http://svnbook.red-bean.com/ and have a link with a size warning -- All Entries (1.6 MB HTML). That should keep people from clicking on it with their smartphone :-). I can tweak it again to do that if people like. Cheers, -Ryan On Sun, Oct 24, 2010 at 8:15 AM, Thomas Schilling nomin...@googlemail.comwrote: For packages with many items in the index, these pages can get a bit huge. How about a permuted index like http://www.lispworks.com/documentation/HyperSpec/Front/X_Symbol.htm? E.g., for your use case, you would go to E and then the row with all the End entries, which would contain all the names with End anywhere in their name. I don't know if page size can be a problem, but at least for mobile or otherwise low-bandwidth devices this can be a nice alternative. On 24 October 2010 04:41, Ryan Newton new...@mit.edu wrote: When I encounter a split-index (A-Z) page it can be quite frustrating if I don't know the first letter of what I'm searching for. I want to use my browser find! For example, tonight I wanted to look at all the functions that END in Window in the Chart package -- no luck: http://hackage.haskell.org/packages/archive/Chart/0.13.1/doc/html/doc-index.html Therefore I propose that even when generating the A-Z individual pages that there also be an All option for the single-page version. Attached is a patch against haddock's HEAD (darcs get http://code.haskell.org/haddock/ right?) that implements this behavior. As an example, here is FGL's documentation built with the patched haddock: http://people.csail.mit.edu/newton/fgl_example_doc/doc-index.html The great thing about hackage being centralized is that if people are happy with this fix it can be widely deployed where it counts, and quickly! Cheers, -Ryan P.S. At the other end of the spectrum, when considering a central index for all of hackage (as in the below ticket) maybe it would be necessary to have more than 26 pages, I.e. Aa-Am | An-Az or whatever. http://hackage.haskell.org/trac/hackage/ticket/516#comment:6 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe -- Push the envelope. Watch it bend. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] back doors into the IO monad
On Sun, Oct 24, 2010 at 10:22 AM, Nicolas Pouillard nicolas.pouill...@gmail.com wrote: On Sun, 24 Oct 2010 00:28:37 +0200, Manlio Perillo manlio_peri...@libero.it wrote: Hi. What are the available methods to execute IO actions from pure code? I know only unsafePerformIO and foreign import (to call a non pure foreign function). unsafeCoerce is a back door for almost everything. The mueval tests also include this fun example: 'runST (unsafeIOToST (readFile /etc/passwd))' -- gwern http://www.gwern.net/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Re: Haskell meeting in Berlin
On Saturday, October 23, 2010 07:50:41 pm Johannes Waldmann wrote: There will be an informal Haskell meeting [...] The bi-monthly lisp meeting [...] [1] http://www.c-base.org/calender/phpicalendar/month.php Pray tell - which of the three above-mentioned languages is from the dark side? Point taken. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Parsec in Haskell platform
andrewcoppin: On 24/10/2010 09:10 AM, Roman Cheplyaka wrote: It would be convenient to have a page which would list all the HP packages with their versions. The release page [2] only has a list of packages whose versions has changed since the last release, as I understood. It would be nice to have a page that lists everything included in every HP release, together with their version numbers. (So that, e.g., I can see at a glance what version of GHC, Haddock or cabal-install is in HP-2009.1.0.0.) All this information must exist somewhere, it's just not easily viewable on the web. http://code.haskell.org/haskell-platform/haskell-platform.cabal ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Eta-expansion and existentials (or: types destroy my laziness)
Max Bolingbroke schrieb: Let's start with a simple example of an existential data type: data Stream a = forall s. Stream s (s - Maybe (a, s)) I use quite the same data type for my signal processing applications: http://code.haskell.org/synthesizer/core/src/Synthesizer/State/Signal.hs You may be interested in my overview: http://hackage.haskell.org/packages/archive/synthesizer/0.2.0.1/doc/html/Synthesizer-Storage.html Now we may wish to define infinite streams using value recursion: ones :: Stream Int ones = cons 1 ones For me a Stream is a List without storage, thus in your example you cannot share the content of 'ones'. The internal state type becomes larger and larger for every new element (every element adds a new layer of Maybe, if I'm not mistaken). In cases, where I want this kind of recursion I store the content generated by a Stream in a list or another more efficient stream type or I write special functions for this purpose (e.g., 'repeat' in the case of 'ones'). ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: iteratee-compress 0.1.1
On 10/24/10 7:09 AM, Maciej Piechotka wrote: Iteratee-compress provides compressing and decompressing enumerators including flushing. Currently only gzip is provided but at least bzip is planned. Changes from previous version: - Independent from zlib library (Haskell one, not C) - Allow hand-flushing the contents (from outside). - Fix potential memory-leak Next goals: - BZip support - Generic interface for flushing Have you thought about adding LZO[1] support? There'd be the usual licensing issues for GPL, but it offers a realtime alternative to gzip (i.e., decompression time is hidden by I/O latency) with comparable compression performance. [1] http://www.oberhumer.com/opensource/lzo/ -- Live well, ~wren ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Bug in HTTP (bad internal error handling)
On Sun, 24 Oct 2010, Bit Connor wrote: On Sat, Oct 23, 2010 at 8:49 PM, Ganesh Sittampalam gan...@earth.li wrote: I'm just looking at fixing this so I can make an upload as discussed with Sigbjorn. I guess the best thing to do is to make all the calls to fail into something more explicit. Yep. Ideally, simpleHTTP and its friends should never throw an IO exception (and should certainly never call prelude's error, as happens now). Network errors and other errors should be reported by returning an appropriate ConnError Result: I'm not going to try to audit for that exhaustively right now, but I've changed all the calls to fail that were broken by the Either change, so hopefully we should be back where we started. I've put a repo up with the changes and a base dependency bump at http://urchin.earth.li/git/HTTP - could you (and anyone else interested) give it a try? I haven't done anything but the most minimal testing and in particular I haven't made any effort to reproduce your specific problem. I plan to upload this as HTTP-4000.0.10 in the next few days. Cheers, Ganesh ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Re: ANNOUNCE: iteratee-compress 0.1.1
On Sun, 2010-10-24 at 15:03 -0400, wren ng thornton wrote: On 10/24/10 7:09 AM, Maciej Piechotka wrote: Iteratee-compress provides compressing and decompressing enumerators including flushing. Currently only gzip is provided but at least bzip is planned. Changes from previous version: - Independent from zlib library (Haskell one, not C) - Allow hand-flushing the contents (from outside). - Fix potential memory-leak Next goals: - BZip support - Generic interface for flushing Have you thought about adding LZO[1] support? There'd be the usual licensing issues for GPL, but it offers a realtime alternative to gzip (i.e., decompression time is hidden by I/O latency) with comparable compression performance. [1] http://www.oberhumer.com/opensource/lzo/ Currently I thought only about bzip2/gzip. Probably .xz support would follow if any. LZO, as you said, is on GPL-2. While I have no problems with GPL-2 some potential users may (Haskell tend to be BSD3 community). Does anyone knows if conditional compilation solves problem? In my interpretation of GPL-2 yes but I'm not sure. Regards signature.asc Description: This is a digitally signed message part ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Re: ANNOUNCE: iteratee-compress 0.1.1
On Sun, Oct 24, 2010 at 9:59 PM, Maciej Piechotka uzytkown...@gmail.com wrote: Currently I thought only about bzip2/gzip. Probably .xz support would follow if any. LZO, as you said, is on GPL-2. While I have no problems with GPL-2 some potential users may (Haskell tend to be BSD3 community). Does anyone knows if conditional compilation solves problem? In my interpretation of GPL-2 yes but I'm not sure. Conditional compilation is tricky for licensing purposes, I think. But a 'iteratee-compress-lzo' or 'iteratee-compress-gpl' package would definitely avoid such issues. Cheers! -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANNOUNCE: iteratee-compress 0.1.1
On 24 October 2010 20:09, Maciej Piechotka uzytkown...@gmail.com wrote: Iteratee-compress provides compressing and decompressing enumerators including flushing. Currently only gzip is provided but at least bzip is planned. Changes from previous version: - Independent from zlib library (Haskell one, not C) - Allow hand-flushing the contents (from outside). - Fix potential memory-leak Next goals: - BZip support - Generic interface for flushing To think about: - Should inner iteratee be able to request flushing? Regards PS. It did change API by removing dependency on zlib but I home such breakage in 0.1.x will be allowed version numbers are cheap :) Conrad. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Re: ANNOUNCE: iteratee-compress 0.1.1
On 10/24/10 8:14 PM, Felipe Lessa wrote: On Sun, Oct 24, 2010 at 9:59 PM, Maciej Piechotkauzytkown...@gmail.com wrote: Currently I thought only about bzip2/gzip. Probably .xz support would follow if any. LZO, as you said, is on GPL-2. While I have no problems with GPL-2 some potential users may (Haskell tend to be BSD3 community). Does anyone knows if conditional compilation solves problem? In my interpretation of GPL-2 yes but I'm not sure. Conditional compilation is tricky for licensing purposes, I think. But a 'iteratee-compress-lzo' or 'iteratee-compress-gpl' package would definitely avoid such issues. I was going to suggest iteratee-compress-gpl... -- Live well, ~wren ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
