June 2010 - Haskell-prime

Second draft of the Haskell 2010 report available
by Simon Marlow 08 Jul '10

08 Jul '10

The second draft of the Haskell 2010 report is now available in PDF and HTML formats (the PDF looks a lot nicer): http://www.haskell.org/~simonmar/haskell-2010-draft-report-2.pdf http://www.haskell.org/~simonmar/haskell-2010-draft-report-2/haskell.html relative to the first draft, which was only publicised on the haskell-prime mailing list, I have now updated the libraries too. Rather than update all the library documentation manually, I (perhaps rashly) decided to make a LaTeX backend for Haddock instead, and generate the report automatically from the library source code. Getting this to work turned out to be a lot more effort than I anticipated, but I think the results are quite attractive. Once the new Haddock backend is incorporated upstream, we'll finally have the ability to generate decent typeset API documentation. Furthermore, this should make it much easier to incorporate more libraries in future versions of the Haskell standard, should we decide to do so. Right now, the HTML version of the report is generated from the LaTeX sources, including the libraries. We could use the Haddock HTML output instead, but that would entail some difficulties with cross-references from the language part of the report to the libraries, which is why I've left it this way for now. This is why the libraries part of the report is bereft of hyperlinks in HTML; but at least it is well indexed in the PDF version. Summary of the library changes in Haskell 2010 relative to Haskell 98 and the FFI specification: * All libraries have been updated to their hierarchical names * The following library modules were dropped from the standard, due to being obsolete or superseded. Replacements are not part of the standard yet, but it is expected that they will be replaced in the future: Directory, System, Time, Locale, CPUTime, Random In the case of System, some functions have moved to the new modules System.Environment and System.Exit. * Foreign.Marshal.Error: functions on IOError moved to System.Error (this is where they've been in base for ever). * Data.List: added intercalate, subsequences, permutations, foldl', foldl1', stripPrefix (H2010 Data.List matches the current base version). * Data.Char: various additions of Unicode predicates (e.g. isLetter, isMark, isNumber), the GeneralCategory type and generalCategory. (matches the base version) * Control.Monad: added forM, forM_, (>=>), (<=<), forever, foldM_, replicateM, replicateM_ (matches the base version). * System.IO: added fixIO, hSetFileSize, hTell, hIsTerminalDevice, hSetEcho, hGetEcho, hShow. The base version has various additions: hGetBuf/hPutBuf, binary Handles, encodings, and newline support, but I erred on the side of being conservative here: these APIs need discussion, and in some cases are probably not suitable for the standard in their current state at all. * System.IO.Error: new module, providing functionality that was in Foreign.Marshal.Error in the FFI spec. * System.IO.Exit: new module, functionality moved from H98 System * System.Environment: new module, functionality moved from H98 System We expect to provide exactly these libraries in GHC 6.14, although the exact mechanism has yet to be decided; for discussion see http://www.haskell.org/pipermail/haskell-prime/2010-April/003158.html Comments on the draft report are welcome, before I finalise this and sign off on Haskell 2010. Cheers, Simon

7 16

Re: [Haskell] Second draft of the Haskell 2010 report available
by Simon Marlow 30 Jun '10

30 Jun '10

On 29/06/2010 23:31, Henk-Jan van Tuyl wrote: > On Tue, 29 Jun 2010 17:01:54 +0200, Simon Marlow <marlowsd(a)gmail.com> > wrote: > >> Comments on the draft report are welcome, before I finalise this and >> sign off on Haskell 2010. > > Subsection 12.3, "Language extensions", mentions the FFI as a language > extension, but FFI is now part of the standerd; the same goes for the > "extensions" mentioned at the end of the subsection. Thanks; changed to refer to them as "language features" rather than "extensions". Cheers, Simon

1 0

Re: [Haskell-cafe] Proposal to solve Haskell's MPTC dilemma
by Brandon S. Allbery KF8NH 09 Jun '10

09 Jun '10

On Jun 8, 2010, at 15:32 , Job Vranish wrote: > It seems like this would make working with MPTCs much easier. > When programming, I generally want to only specify the minimum > amount of information to make my code logically unambiguous. > If the code contains enough information to infer the proper > instantiation without the use of an FD, then I shouldn't need to add > a FD. > It seems like this would have much more of a "it just works" feel > than the currently alternatives. I can't help but think that the "it just works" mentality leads to duck typing. -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery(a)kf8nh.com system administrator [openafs,heimdal,too many hats] allbery(a)ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH

1 0