May 2011 - Libraries - Haskell.org

Re: Libraries Digest, Vol 93, Issue 20
by Louis Wasserman 23 May '11

23 May '11

I remember that proposal, and it would've definitely supported strictified operations. On Sat, 21 May 2011, Edward Z. Yang wrote: > > > As it stands, we have strict functions for certain operations, such as > > insertWith. But this current selection is somewhat arbitrary. For > > illustrative purposes, here are other functions that ought to be > strictified in > > IntMap, because there is no easy way for a user to implement them. > > Is it possible to solve this with the proposed Location interface? > > "Library proposal: add a Location interface for element-wise operations on > Data.Map" > http://www.haskell.org/pipermail/libraries/2011-January/015593.html > >

1 0

Bugfix for parallel package
by Daniel Fischer 17 May '11

17 May '11

Hi, since libraries@ is the maintainer of parallel, I send the patch for http://hackage.haskell.org/trac/ghc/ticket/5100 here. nonempty-chunks fixes the bug, whitespace removes trailing whitespace in Control.Parallel.Strategies, version-bump bumps the version to 3.1.0.2. Please review.

1 0

ANN: protocol-buffers, protocol-buffers-descriptor, hprotoc version 2.0.5
by Chris Kuklewicz 16 May '11

16 May '11

There is a new version of the Haskell implementation of Google's protocol buffer message serialization format. Performance for serialization has been improved in this release. The new code is up on hackage at http://hackage.haskell.org/package/protocol-buffers http://hackage.haskell.org/package/protocol-buffers-descriptor http://hackage.haskell.org/package/hprotoc The darcs repository is at http://code.haskell.org/protocol-buffers/ The changes in 2.0.5 compared to 2.0.2 are: * Specialize some encoding during serialization to improve performance * Add some options that protoc handles and that were missing in hprotoc * Alter the namespace resolution in hprotoc to better match protoc * Improve error handling during name resolution in hprotoc * Lex and parse the new aggregate options in Google version 2.4.0 Note that the new aggregate syntax in Google's version 2.4.0 is not supported. The change allows the new notation to be parsed and a clear error message to be reported. -- Chris Kuklewicz

1 0

Floating point and RealFrac
by Daniel Fischer 16 May '11

16 May '11

Looking at the old bug list again, there are a couple of tickets I might be able to do something about. Since library submissions are somewhat painful, though, and I'm not bitten by the issues, I'd like to know if there's general interest to do something about them. 1) http://hackage.haskell.org/trac/ghc/ticket/2269 Feature request: add primops word2Double#, double2Word#, word2Float#, float2Word# Well, I don't know how to create primops, but I could twiddle a few bits in C to produce a hopefully faster conversion than we have now. 2) http://hackage.haskell.org/trac/ghc/ticket/2281 a fast 'fraction' function for Double Although fraction x = snd (properFraction x :: (Integer, TYPE_OF_x)) , when compiled with optimisations, is not as slow as it used to be, it's still not very fast. A binding to modf/modff or some bit-twiddling in C should be faster. If there's demand for this, should a `fraction' function be added to RealFrac, should standalone functions fractionFloat and fractionDouble be added to GHC.Float with rewrite rules snd (properFraction x) -> fractionTYPE x ? 3) http://hackage.haskell.org/trac/ghc/ticket/3676 realToFrac misbehaves on NaNs, -0.0 and infinities (for Float -> Double) when the rewrite rule doesn't fire. As long as the realToFrac conversion [without rewrite rules] goes through Rational, this is basically unfixable. So a real fix would necessitate a language change. Some ideas have been shortly discussed in http://www.haskell.org/pipermail/haskell-prime/2010-February/003139.html and replies. The most reasonable candidates IMO are - a dedicated floating point conversion, with * a type alias FloatMax for the largest (real + infinities + NaNs) Floating point type, currently Double * methods toFloatMax and fromFloatMax added to RealFloat * the conversion function floatConvert :: (RealFloat a, RealFloat b) => a -> b - adding a type specifically for conversions between different number types that caters for special values, something along the lines of data NumRep = NegativeZero | Infinity Bool -- to indicate sign | NaN -- possibly with a parameter to encode different NaNs | BinExp Integer Int -- m*2^e | Rat Rational | Entire Integer -- ? and let realToFrac convert via this type. The BinExp constructor would allow a more efficient conversion between base-2 floating point types than fromRational . toRational can be. If such a type would also be used for the Read instances, instead of Rational, adding a constructor for base-10 representations would be useful (cf. http://hackage.haskell.org/trac/ghc/ticket/3897 ) [and Rational could be freed from the invalid values 0:%0, ±1:%0]. Both candidates would be major undertakings and have their downsides. I can live with the current state of affairs, what bugs me most about it is the open ticket.

1 0

Proposal: Tidy up and export PSQ from base
by Edward Z. Yang 15 May '11

15 May '11

The new IO event manager implements a priority search queue internally. This is a kind of handy data structure; in particular, some algorithmic improvements in Hoopl would require a decent priority queue implementation. I propose that we move GHC.Event.PSQ to Data.PSQ, cleaning up the dependency on GHC.Event.Unique, and export it unconditionally (right now it is only available on non-Windows.) Discussion period: 1 week. Cheers, Edward

8 20

test-suite type: detailed-1.0
by Guy 15 May '11

15 May '11

The user guide for Cabal 1.10.1.0 specifies "detailed-1.0" as an acceptable test-suite type, but cabal reports Warning: 'detailed-1.0' is not a supported test suite version. The known test suite types are: exitcode-stdio-1.0

2 1

cabal test = permission denied?
by Guy 14 May '11

14 May '11

I can successfully run cabal test, and my test suite passes. Cabal's output it: Running 1 test suites... Test suite tests: RUNNING... Test suite tests: PASS Test suite logged to: dist\test\foo-0.1-tests.log cabal: permission denied foo-0.1-tests.log is present and correct. There is also an empty file cabal-test-4513.log. I am on Windows 7 x64, cabal 1.10.1.0. Running as Admin makes no difference.

2 1

cabal-latest
by Guy 13 May '11

13 May '11

http://www.haskell.org/cabal/release/cabal-latest/ points to version 1.8.0.4. Is this correct?

1 0

ANN: version 2.0.0 of protocol-buffers, protocol-buffers-descriptor, hprotoc
by haskell＠list.mightyreason.com 12 May '11

12 May '11

A new major version of the Haskell implementation of Google's protocol buffers is done. The new code is in the usual three packages at http://hackage.haskell.org/package/protocol-buffers http://hackage.haskell.org/package/protocol-buffers-descriptor http://hackage.haskell.org/package/hprotoc The darcs repository is still at http://code.haskell.org/protocol-buffers/ This is the first release where performance was tested and improved: specifically parsing messages from the wire protocol. The new library now defaults to using strict fields for generated messages. This vastly improves the memory usage when getting messages from the binary format. The old lazy fields can still be generated by hprotoc with a new command line option. On "OS X" with 64-bit ghc the C++ loading is 0.5 seconds and Haskell is 1.6 seconds (128 MB memory usage). This is down from 8.3 seconds (1597 MB memory usage). A related change in version 2.0.0 is that 'mergeEmpty' no longer exists. With strict messages and required fields this would have been a dangerous value to keep. If you used this, then please use 'defaultValue' instead. The Haskell code has been compiled against Google's version 2.4.0a proto files. But the Haskell code is only compatible up to Google's version 2.3.0, and in particular the new default message syntax is not yet supported in the 'hprotoc' parser. -- Chris

2 1

Messed up with Cabal on Mac
by Harad 09 May '11

09 May '11

Hello everybody, I was trying to install `snap` but I was getting this error: <command line>: cannot satisfy -package-id haskell98-1.1.0.1-6c0d823d8a0fe4966441285c9ec0b534: I still don't fully understand why was this happening but, reading here and there, I ended up running `ghc-pkg unregister --user haskell98`. Some of you may be facepalming at this point, because then if I ran `ghc-pkg check` I got a really, really long list of broken packages. So I decided to throw it all away, removed the whole Haskell Platform from my computer as told here <http://bit.ly/jHBzEk> (I know, pretty Mac-ish solution), installed it back... but I continue getting the same error when installing `snap`, and `ghc-pkg check` still throws me this infinite list: http://pastebin.com/duTC4rqX . I searched but I didn't find any posible solution. Do you have any idea of how can I clean up all that mess? Thanks in advance! (If my English is awful is because I'm Spanish and Spanish people are the worse language learners in the world, so I'm sorry for that.)

2 2