- Haskell-Cafe - Haskell.org

Creating a dependency graph (in the hope it may show some structure)
by Bram Neijt 02 May '12

02 May '12

Dear Haskelllers, I've reacently started with Haskell and during the Dutch Hackaton[3], I decided to generate graphs of dependencies. I was thinking about doing this for Java earlier, and that weekend I decided to write the tooling in Haskell. I've written two programs: hs2dot[1] and dotimate[2]. The first generates a dot file (Graphviz) from all .hs files below the current working directory. The second will take one or more dot files and call neato (from Graphviz) multiple times to generate frames of neato output from one to the next. Trying hs2dot and dotimate on two sequencial commits in the cabal/cabal-install/frames directory already gives me a way to complex and full graph. See attached. ==> My question is: What dependencies, if any, would be nice to graph, and/or which dependencies could I filter out to keep this from becoming a mess? Greetings, Bram Neijt [1] https://gist.github.com/bneijt/hs2dot [2] https://gist.github.com/bneijt/dotimate [3] http://www.haskell.org/haskellwiki/DHD_UHac#Utrecht_Hackathon

2 2

ANN: bitwise - fast multi-dimensional unboxed bit packed Bool arrays
by Claude Heiland-Allen 01 May '12

01 May '12

Hi all, I'm pleased to announce bitwise-0.1: http://hackage.haskell.org/package/bitwise --8<-- excerpt from the hackage page Unboxed multidimensional bit packed Bool arrays with fast aggregate operations based on lifting Bool operations to bitwise operations. There are many other bit packed structures out there, but none met all of these requirements: 1. unboxed bit packed Bool array, 2. multi-dimensional indexing, 3. fast (de)serialization, or interoperable with foreign code, 4. fast aggregate operations (fold, map, zip). --8<-- end excerpt bitwise has been tested in hugs -98, ghc-7.0.4, ghc-7.4.1. Bool data structures comparison (features numbered as excerpt above): package features additional notes 1 2 3 4 array Y Y n n bitarray Y n n ? based on array bitstream n n ? ? based on vector bitstring Y n ? ? based on bytestring bit-vector n n Y ? based on vector bitwise Y Y Y Y repa n Y Y ? based on vector uvector Y n ? ? deprecated vector n n Y ? uses a Word8 for each Bool Known shortcomings of this first release of bitwise: * missing common instances (Eq, Ord, Read, Show, ...); * missing additional functions (counting, searching, ...); * missing documentation on ByteString format: the bits are taken from the array and packed into Word8s, with the first bit (in array order) becoming the least significant bit of the first Word8; the last Word8 is padded with 0 at the most-significant-bit end; * PBM reading is not compliant to the official specification; * misleading benchmark (bitwise zipWith is fast, but there may be a faster way to zipWith UArray ix Bool than the one I used). What I'm using bitwise for at the moment: A building block for fractal imaging: an image is made up of cells, each cell can be in one of 4 states (represented with 2 bits): * cell is 100% interior to a set of points in the plane * cell is 100% exterior * partially interior and partially exterior (ie, on the boundary) * unknown with recursive subdivision of cells into images taking place only when the cell is on the boundary of the set being imaged. Thanks, Claude

1 0

Re: [Haskell-cafe] Darcs home page updated
by Paul R 01 May '12

01 May '12

On Mon, 30 Apr 2012 12:27:57 -0700, Simon Michael <simon(a)joyful.com> said: Simon> With 2.8 released, I felt Darcs deserves better presentation. After Simon> surveying other VCS sites I worked on an update to our home page Simon> layout and content over the last few days, with review and input from Simon> #darcs, and it went live last night. It's far from perfect but I hope Simon> it's a good step forward. Thanks for the input, and have at it: Simon> http://darcs.net Hi, that looks good, thanks. Here are some ideas : - The purple color of titles is not very attractive, changing it to something more saturated would make it look fresher. - the purple color rounded box is not very attractive either :) - there is a video right on the first page, good ! Unfortunatly, it is not showing darcs but its cousin. Also, the title is "why do we continue to develop ... [camp]", which isn't the most positive approach one can expect on a start page - the compact tutorial is great, I love it. It deserves more poish. For inspiration, one can have a look at this website for example : http://gembundler.com/ - underlined links looks a bit 90's. You can add a simple css rule to choose an other color and remove the underline (though it is good practice to set it on mouseover) cheers, -- Paul

2 2

ANN: HandsomeSoup-0.3: CSS selectors for HXT
by aditya bhargava 01 May '12

01 May '12

*Homepage:* http://egonschiele.github.com/HandsomeSoup *On Hackage:* http://hackage.haskell.org/package/HandsomeSoup *Blurb:* HandsomeSoup is the library I wish I had when I started parsing HTML in Haskell. It is built on top of HXT and adds a few functions that make is easier to work with HTML. Most importantly, it adds CSS selectors to HXT. The goal of HandsomeSoup is to be a complete CSS2 parser for HXT (it is very close to this right now). -- adit.io

2 2

Darcs home page updated
by Simon Michael 30 Apr '12

30 Apr '12

With 2.8 released, I felt Darcs deserves better presentation. After surveying other VCS sites I worked on an update to our home page layout and content over the last few days, with review and input from #darcs, and it went live last night. It's far from perfect but I hope it's a good step forward. Thanks for the input, and have at it: http://darcs.net -Simon

2 1

Re: [Haskell-cafe] [Haskell] ANN: Portackage - a hackage portal
by Andrew Seniuk 30 Apr '12

30 Apr '12

> On Apr 27, 2012 6:18 PM, "Andrew Seniuk" <rasfar(a)gmail.com> wrote: > Oh, the idea wasn't that you would make the app, but that by providing a web > service interface you would make it easier for me or anyone else to make an > app or other alternative interface. It would also reduce bandwidth usage > over screen scraping. Yes, well as I mentioned in the part of my response you didn't quote, I agree it would be desireable, and am working on it, but my motivation was to provide people with something immediately useful. Once I get something reasonably efficient worked out, I'll look at uploading the Haskell code to Hackage, providing JSON data, etc.. Meanwhile if someone wanted to work on their own app or interface, it's not too hard to get CSV from the table. I made a mistake and the link was down for about 12 hours. It's back now, with a number of improvements and up-to-date hackage info. It is an expensive page to run, you may get "Script not responding" pop-ups depending on your hardware, but I think it's worth it. http://fremissant.net/portackage Kind Reg'ds, Andrew

2 4

ANN: hflags-0.1, a command line flags library similar to Google's gflags
by Gergely Risko 30 Apr '12

30 Apr '12

Hello, I would like to announce the release of HFlags, a command line flags library that makes it very easy to define and use flags. The API is similar to Google's gflags, here is a very simple example program: -=- #!/usr/bin/env runhaskell {-# LANGUAGE TemplateHaskell #-} import HFlags defineFlag "name" "Indiana Jones" "Who to greet." defineFlag "r:repeat" (3 + 4 :: Int) "Number of times to repeat the message." main = do remainingArgs <- $(initHFlags "Simple program v0.1") sequence_ $ replicate flags_repeat greet where greet = putStrLn $ "Hello " ++ flags_name ++ ", very nice to meet you!" -=- As you can see, we have TemplateHaskell functions to help with the definition (and initialization) of the flags, and the values themselves are pure, no need for being in the IO monad to use the value of the flags. Also, the initHFlags function automagically gathers all the flags defined anywhere in the whole program, so if a library defines flags, you don't have to mention it in the main, but the user will still be able to set those flags too. More details behind this design and more example can be found in the following post: http://blog.risko.hu/2012/04/ann-hflags-0.html Comments and criticism is welcome. Code: http://github.com/errge/hflags Examples: https://github.com/errge/hflags/tree/master/examples Hackage: http://hackage.haskell.org/package/hflags Best regards, Gergely Risko

1 0

Hong Kong Haskell hackathon - call for participation
by Matthias Fischmann 30 Apr '12

30 Apr '12

Hi everybody, if you are in or around Hong Kong in May and want to play, please check out this: http://www.haskell.org/haskellwiki/HkHac2012 Otherwise, any wider circulation or suggestion of channels / university people / companies that I should invite directly is welcome. cheers, Matthias

1 0

Printing call site for partial functions
by Michael Snoyman 30 Apr '12

30 Apr '12

I had a bug in a site of mine[1] for a few weeks, where it would just print: Prelude.head: empty list It took a long time to track down the problem, as it came from some other library I was depending on. Eventually I tracked it down, reported it, and the problem was fixed the next day. The weak link in this chain was identifying what package was to blame. I noticed the other day that the docs for undefined[2] state: > It is expected that compilers will recognize this and insert error messages which are more appropriate to the context in which undefined appears. I'm not sure if the current output of `Prelude.undefined` is meant to fit the criterion of "appropriate to the context", but in my opinion it does not. I'd like to propose a bit of a strawman to improve the situation for not just `undefined`, but any arbitrary partial function (including `head`). What we really want is to define some function like: headContext :: String -> [a] -> a headContext context [] = error $ "Prelude.head: empty list (" ++ context ++ ")" headContext _ (x:_) = x And then have the compiler automatically include (optional) package name, module name, and line number where `headContext` was called. How about we borrow a bit from rewrite rules, and have a pragma such as: {-# WITH_CONTEXT head headContext #-} If the compiler supports the feature, and is able to provide context in the given location, it will call `headContext` with an appropriate `String`. Otherwise, normal `head` would be called. I'm sure there are many better ways to approach the problem, and I can't speak to the complexity of implementation within GHC. I *can* say, however, that this would have saved me a lot of time in the example I gave above, and I'd bet many Haskellers have similar stories. This could be a huge debugging win across the board. Michael [1] It was actually yesodweb.com, and resulted in empty Javascript files being generated, thus disabling some features of the site. [2] http://hackage.haskell.org/packages/archive/base/4.5.0.0/doc/html/Prelude.h…

8 10

ANN: unfoldable-0.4.0
by Sjoerd Visscher 30 Apr '12

30 Apr '12

I am pleased to announce the 5th version of the unfoldable package. (This is the first announcement, you didn't miss anything.) http://hackage.haskell.org/package/unfoldable-0.4.0 Just as there's a Foldable class, there should also be an Unfoldable class. This package provides one: class Unfoldable t where unfold :: Unfolder f => f a -> f (t a) Writing instances of Unfoldable is similar to writing Traversable instances. For example, given a data type data Tree a = Empty | Leaf a | Node (Tree a) a (Tree a) a suitable instance would be instance Unfoldable Tree where unfold fa = choose [ pure Empty , Leaf <$> fa , Node <$> unfold fa <*> fa <*> unfold fa ] The choose function comes from the Unfolder class: class Applicative f => Unfolder f where choose :: [f x] -> f x (If f is an Alternative instance, choose is simply Data.Foldable.asum.) Different unfolders provide different ways of generating values, for example: - Random values - Enumeration of all values (depth-first or breadth-first) - Convert from a list - An implementation of QuickCheck's arbitrary should also be possible (still working on that) Some examples can be found in the examples directory in the github repo: https://github.com/sjoerdvisscher/unfoldable Ideas and comments are welcome! greetings, Sjoerd

4 7