- Libraries - Haskell.org

Readline read_history and write_history addition
by Alexander Dunlap 21 Feb '08

21 Feb '08

Hello all, I'd like to propose the addition of read_history and write_history bindings to the readline library. I believe I've followed the instructions on the Haskell wiki appropriately and I've set up a ticket (with a patch) at http://hackage.haskell.org/trac/ghc/ticket/2053. Essentially, the patch adds simple bindings so that Haskell programs can use Readline's read_history and write_history functions. This will be useful for the implementation of http://hackage.haskell.org/trac/ghc/ticket/2050 (add persistent history to GHCi) and potentially other applications. Thanks for your time and consideration. Alex

11 39

Re: [Haskell-cafe] The Proliferation of List-Like Types
by Henning Thielemann 20 Feb '08

20 Feb '08

> Now, to help solve this problem, I wrote ListLike[2], providing a > set of typeclasses that make list operations generic. I also provided > default instances of ListLike for: > > ListLike Data.ByteString.ByteString Word8 > ListLike Data.ByteString.Lazy.ByteString Word8 > ListLike [a] a > (Integral i, Ix i) => ListLike (Array i e) e > (Ord key, Eq val) => ListLike (Map key val) (key, val) It's a multi-parameter type class, right? So it's difficult to push it to the core. > Now, the questions: > > 1) Does everyone agree with me that we have a problem here? I agree. > 2) Would it make sense to make ListLike, or something like it, > part of the Haskell core? Somehow yes. However since the 'base' package is constantly split into smaller parts, there is maybe no need to merge it somewhere, but introduce simply new package dependencies. > 3) Would it make sense to base as much code as possible in the Haskell > core areound ListLike definitions? Here I think of functions such > as lines and words, which make sense both on [Char] as well as > ByteStrings. > > 4) We are missing one final useful type: a Word32-based ByteString. > When working in the Unicode character set, a 32-bit character > can indeed be useful, and I could see situations in which the > performance benefit of a ByteString-like implementation could > be useful combared to [Char]. StorableVector should fill this gap. http://code.haskell.org/~sjanssen/storablevector/

1 0

select and selectSplit
by Cale Gibbard 18 Feb '08

18 Feb '08

I know Bulat will be terribly disappointed by my suggestion to make an addition to Data.List ;) but I mentioned the following couple of functions on #haskell and got some positive response that they were things other people ended up writing all the time as well, so I decided I'd propose them here as additions to Data.List and see what kind of reaction I got: -- | The 'select' function takes a list and produces a list of pairs -- consisting of an element of the list together with a list of the -- remaining elements. select :: [a] -> [(a,[a])] select [] = [] select (x:xs) = (x,xs) : [(y,x:ys) | (y,ys) <- select xs] -- | The 'selectSplit' function takes a list and produces a list of -- triples consisting of a prefix of the list, the element after it, -- and the remainder of the list. selectSplit :: [a] -> [([a],a,[a])] selectSplit [] = [] selectSplit (x:xs) = ([],x,xs) : [(x:lys,y,rys) | (lys,y,rys) <- selectSplit xs] These tend to be very handy in the list monad. The names are off the top of my head. Logan Capaldo also suggested 'pick' for the first, which is a name that I've used as well. Faxathisia mentioned that it's called 'select' in Prolog as well. As a side note, the state transformer makes it relatively easy to pick n elements using this: pick n = runStateT . replicateM n . StateT $ select (Showing that select is secretly an operation in a state transformed list monad, where the state is a list of elements.) If the order of the pairs in the MTL is fixed in the future in order to better reflect the available instances of Functor/etc., at that point we may want to consider flipping the pairs in the result of select to match. - Cale

8 13

HSQL defunct?
by Frederik Eaton 18 Feb '08

18 Feb '08

Hello all, Is there going to be a new maintainer for HSQL? I have submitted a number of patches to fix some painful bugs. It's been almost a year since my previous enquiry and the last release was December 15, 2005, over two years ago: https://sourceforge.net/project/showfiles.php?group_id=65248 The software no longer compiles with the latest version of GHC, due to some (seemingly gratuitous) changes in the Cabal libraries (I see there is a September 2007 thread on libraries@). Unfortunately, I am too busy to maintain it myself; I was probably even too busy to write the patches in the first place, but it is sad to see them go to waste... Frederik P.S. Quan Ta: regarding your July 20, 2006 query, "HSQL-MySQL: non-standard port", my first patch adds the ability to specify a non-standard port with the server host name, if you are still interested I can send it. On Sun, Mar 25, 2007 at 06:55:21PM +0200, Björn Bringert wrote: > I found this very old mail in my inbox, sorry for not responding sooner. > Yeah, a new HSQL release would be a good idea, IMO. Krasimir is the > maintainer, so it's really up to him. > > /Björn > > Frederik Eaton wrote: > > Hi Bjorn, > > > > The sourceforge page for HSQL shows a most recent release date of > > December 15, 2005 - over a year ago, presumably not including the > > timezone patch which I think fixes a rather serious bug, which caused > > a lot of problems for me when I tried to use HSQL for my master's > > project. Perhaps it might be helpful to other users if the fix were > > made available to them as well? > > > > Thanks, > > > > Frederik > > > > On Fri, Aug 25, 2006 at 11:20:55AM +0200, Björn Bringert wrote: > >> Sorry for the delay, I've been on vacation. The timezone patch has been > >> committed. I haven't added the Makefile, since I think that some cabal > >> tools that will come with GHC 6.6 could make it obsolete. > >> > >> /Björn > >> > >> Frederik Eaton wrote: > >>> Hi, > >>> > >>> Oops, sorry, there was a bug. Here's a more-tested patch. Stay tuned! > >>> > >>> (if you've already applied the last one, change HSQL.hsc's > >>> > >>> ctMonth = toEnum $ mon, > >>> > >>> to > >>> > >>> ctMonth = toEnum $ (mon-1), > >>> > >>> ) > >>> > >>> Frederik > >>> > >>> On Thu, Aug 17, 2006 at 01:49:40AM +0100, Frederik Eaton wrote: > >>> > >>>> I'm attaching a patch which I've tested. > >>>> > >>>> I'm also attaching a Makefile derived from Bulat's IIRC, it might be > >>>> good to include it in those directories which contain a .cabal file. > >>>> > >>>> Thanks, > >>>> > >>>> Frederik > >>>> > >>>> On Thu, Aug 17, 2006 at 01:17:11AM +0100, Frederik Eaton wrote: > >>>> > >>>>> Hi, in > >>>>> > >>>>> HSQL/Database/HSQL.hsc > >>>>> > >>>>> instance SqlBind ClockTime where > >>>>> ... > >>>>> > >>>>> all of the 'fromSqlValue' functions convert with respect to the local > >>>>> timezone, while 'toSqlValue' converts with respect to UTC. Thus the > >>>>> values being written are not the same, it would seem, as those being > >>>>> read. Certainly that is the case for me. > >>>>> > >>>>> If the three appearances of currTZ within the body of that instance > >>>>> (and not elsewhere) are changed to 0, it should fix the problem (but I > >>>>> haven't tested). > >>>>> > >>>>> Also, I don't understand why a custom function 'mkClockTime' has been > >>>>> written in that module (it calls 'mktime' via FFI), rather than > >>>>> calling the standard library function toClockTime which seems to do > >>>>> the same thing. > >>>>> > >>>>> Frederik > >>>>> > >>>>> On Thu, Aug 10, 2006 at 04:19:36PM +0100, Frederik Eaton wrote: > >>>>> > >>>>>> In HSQL, why are 'datetime' columns interpreted as being in the local > >>>>>> timezone? > >>>>>> > >>>>>> I would think that UTC would be more useful. If the local timezone is > >>>>>> used, then it makes it more difficult to move a server between > >>>>>> timezones, or to have multiple servers in multiple timezones. > >>>>>> > >>>>>> Frederik > >>>>>> > >>>>>> On Sat, Aug 05, 2006 at 06:59:21PM +0100, Frederik Eaton wrote: > >>>>>> > >>>>>>> Thanks! > >>>>>>> > >>>>>>> On Sat, Aug 05, 2006 at 05:27:04PM +0200, Bjorn Bringert wrote: > >>>>>>> > >>>>>>>> On Aug 5, 2006, at 3:34 PM, Frederik Eaton wrote: > >>>>>>>> > >>>>>>>> > >>>>>>>>> On Sat, Aug 05, 2006 at 11:15:19AM +0200, Bjorn Bringert wrote: > >>>>>>>>> > >>>>>>>>>> On Jul 31, 2006, at 12:55 PM, Frederik Eaton wrote: > >>>>>>>>>> > >>>>>>>>>> > >>>>>>>>>>> Hi, > >>>>>>>>>>> > >>>>>>>>>>> I'm attaching another HSQL patch. > >>>>>>>>>>> > >>>>>>>>>>> It allows users to specify a port number with the server name in the > >>>>>>>>>>> MySQL client. > >>>>>>>>>>> > >>>>>>>>>>> Frederik > >>>>>>>>>> Hi Frederik, > >>>>>>>>>> > >>>>>>>>>> the patch doesn't apply cleanly against the current CVS version. Is > >>>>>>>>>> there any chance you could make a patch against the CVS version > >>>>>>>>>> instead? > >>>>>>>>> It's a very simple modification, can't you just copy and paste from > >>>>>>>>> the patch? > >>>>>>>> It appears to have been a simple whitespace mismatch problem (tabs in the original HSQL source). I > >>>>>>>> have committed your patch to CVS. > >>>>>>>> > >>>>>>>> Thanks! > >>>>>>>> > >>>>>>>> /Björn > >>>> -- > >>>> http://ofb.net/~frederik/ > >>> > >>>> diff -ur HSQL-1.7-modified/HSQL/Database/HSQL.hsc HSQL-1.7-modified-2/HSQL/Database/HSQL.hsc > >>>> --- HSQL-1.7-modified/HSQL/Database/HSQL.hsc 2005-12-15 20:57:18.000000000 +0000 > >>>> +++ HSQL-1.7-modified-2/HSQL/Database/HSQL.hsc 2006-08-17 01:43:19.000000000 +0100 > >>>> @@ -304,28 +304,17 @@ > >>>> > >>>> toSqlValue d = show d > >>>> > >>>> -mkClockTime :: Int -> Int -> Int -> Int -> Int -> Int -> Int -> ClockTime > >>>> mkClockTime year mon mday hour min sec tz = > >>>> - unsafePerformIO $ do > >>>> - allocaBytes (#const sizeof(struct tm)) $ \ p_tm -> do > >>>> - (#poke struct tm,tm_sec ) p_tm (fromIntegral sec :: CInt) > >>>> - (#poke struct tm,tm_min ) p_tm (fromIntegral min :: CInt) > >>>> - (#poke struct tm,tm_hour ) p_tm (fromIntegral hour :: CInt) > >>>> - (#poke struct tm,tm_mday ) p_tm (fromIntegral mday :: CInt) > >>>> - (#poke struct tm,tm_mon ) p_tm (fromIntegral (mon-1) :: CInt) > >>>> - (#poke struct tm,tm_year ) p_tm (fromIntegral (year-1900) :: CInt) > >>>> - (#poke struct tm,tm_isdst) p_tm (-1 :: CInt) > >>>> - t <- mktime p_tm > >>>> -#if __GLASGOW_HASKELL__ >= 603 > >>>> - return (TOD (fromIntegral (fromEnum t) + fromIntegral (tz-currTZ)) 0) > >>>> -#else > >>>> - return (TOD (fromIntegral t + fromIntegral (tz-currTZ)) 0) > >>>> -#endif > >>>> -foreign import ccall unsafe mktime :: Ptr () -> IO CTime > >>>> - > >>>> -{-# NOINLINE currTZ #-} > >>>> -currTZ :: Int > >>>> -currTZ = ctTZ (unsafePerformIO (getClockTime >>= toCalendarTime)) -- Hack > >>>> + toClockTime $ CalendarTime { > >>>> + ctYear = year, > >>>> + ctMonth = toEnum $ mon, > >>>> + ctDay = mday, > >>>> + ctHour = hour, > >>>> + ctMin = min, > >>>> + ctSec = sec, > >>>> + ctPicosec = 0, > >>>> + ctTZ = tz > >>>> + } > >>>> > >>>> parseTZ :: ReadP Int > >>>> parseTZ = (char '+' >> readDecP) `mplus` (char '-' >> fmap negate readDecP) > >>>> @@ -373,14 +362,14 @@ > >>>> getTime :: ReadP ClockTime > >>>> getTime = do > >>>> (hour, minutes, seconds) <- readHMS > >>>> - return (mkClockTime 1970 1 1 hour minutes seconds currTZ) > >>>> + return (mkClockTime 1970 1 1 hour minutes seconds 0) > >>>> > >>>> fromSqlValue SqlDate s = f_read getDate s > >>>> where > >>>> getDate :: ReadP ClockTime > >>>> getDate = do > >>>> (year, month, day) <- readYMD > >>>> - return (mkClockTime year month day 0 0 0 currTZ) > >>>> + return (mkClockTime year month day 0 0 0 0) > >>>> > >>>> fromSqlValue SqlDateTimeTZ s = f_read getDateTimeTZ s > >>>> where > >>>> @@ -400,7 +389,7 @@ > >>>> getDateTime :: ReadP ClockTime > >>>> getDateTime = do > >>>> (year, month, day, hour, minutes, seconds) <- readDateTime > >>>> - return (mkClockTime year month day hour minutes seconds currTZ) > >>>> + return (mkClockTime year month day hour minutes seconds 0) > >>>> > >>>> fromSqlValue _ _ = Nothing > >>>> > >>> > >>>> # This Makefile is mostly a wrapper around Setup.hs for people who > >>>> # just want to type make. > >>>> > >>>> CAS=runhaskell Setup.lhs > >>>> CONFOPTS ?= --user --prefix=${HOME} > >>>> INSTOPTS ?= --user > >>>> > >>>> all: build > >>>> > >>>> configure: .setup-config > >>>> .setup-config: HSQL.cabal > >>>> ${CAS} configure ${CONFOPTS} > >>>> > >>>> build: configure > >>>> ${CAS} build > >>>> > >>>> install: build > >>>> ${CAS} install ${INSTOPTS} > >>>> > >>>> uninstall: unregister > >>>> unregister: configure > >>>> ${CAS} unregister > >>>> > >>>> clean: > >>>> -runhaskell Setup.hs clean > >>>> -rm -rf dist > >>>> -rm -f .setup-config > >>>> > >>>> .PHONY: all configure build install uninstall unregister clean > >>> > >>> > >>> > >>> ------------------------------------------------------------------------ > >>> > >>> diff -ur HSQL-1.7-modified/HSQL/Database/HSQL.hsc HSQL-1.7-modified-2/HSQL/Database/HSQL.hsc > >>> --- HSQL-1.7-modified/HSQL/Database/HSQL.hsc 2005-12-15 20:57:18.000000000 +0000 > >>> +++ HSQL-1.7-modified-2/HSQL/Database/HSQL.hsc 2006-08-17 22:04:23.000000000 +0100 > >>> @@ -304,28 +304,17 @@ > >>> > >>> toSqlValue d = show d > >>> > >>> -mkClockTime :: Int -> Int -> Int -> Int -> Int -> Int -> Int -> ClockTime > >>> mkClockTime year mon mday hour min sec tz = > >>> - unsafePerformIO $ do > >>> - allocaBytes (#const sizeof(struct tm)) $ \ p_tm -> do > >>> - (#poke struct tm,tm_sec ) p_tm (fromIntegral sec :: CInt) > >>> - (#poke struct tm,tm_min ) p_tm (fromIntegral min :: CInt) > >>> - (#poke struct tm,tm_hour ) p_tm (fromIntegral hour :: CInt) > >>> - (#poke struct tm,tm_mday ) p_tm (fromIntegral mday :: CInt) > >>> - (#poke struct tm,tm_mon ) p_tm (fromIntegral (mon-1) :: CInt) > >>> - (#poke struct tm,tm_year ) p_tm (fromIntegral (year-1900) :: CInt) > >>> - (#poke struct tm,tm_isdst) p_tm (-1 :: CInt) > >>> - t <- mktime p_tm > >>> -#if __GLASGOW_HASKELL__ >= 603 > >>> - return (TOD (fromIntegral (fromEnum t) + fromIntegral (tz-currTZ)) 0) > >>> -#else > >>> - return (TOD (fromIntegral t + fromIntegral (tz-currTZ)) 0) > >>> -#endif > >>> -foreign import ccall unsafe mktime :: Ptr () -> IO CTime > >>> - > >>> -{-# NOINLINE currTZ #-} > >>> -currTZ :: Int > >>> -currTZ = ctTZ (unsafePerformIO (getClockTime >>= toCalendarTime)) -- Hack > >>> + toClockTime $ CalendarTime { > >>> + ctYear = year, > >>> + ctMonth = toEnum $ (mon-1), > >>> + ctDay = mday, > >>> + ctHour = hour, > >>> + ctMin = min, > >>> + ctSec = sec, > >>> + ctPicosec = 0, > >>> + ctTZ = tz > >>> + } > >>> > >>> parseTZ :: ReadP Int > >>> parseTZ = (char '+' >> readDecP) `mplus` (char '-' >> fmap negate readDecP) > >>> @@ -373,14 +362,14 @@ > >>> getTime :: ReadP ClockTime > >>> getTime = do > >>> (hour, minutes, seconds) <- readHMS > >>> - return (mkClockTime 1970 1 1 hour minutes seconds currTZ) > >>> + return (mkClockTime 1970 1 1 hour minutes seconds 0) > >>> > >>> fromSqlValue SqlDate s = f_read getDate s > >>> where > >>> getDate :: ReadP ClockTime > >>> getDate = do > >>> (year, month, day) <- readYMD > >>> - return (mkClockTime year month day 0 0 0 currTZ) > >>> + return (mkClockTime year month day 0 0 0 0) > >>> > >>> fromSqlValue SqlDateTimeTZ s = f_read getDateTimeTZ s > >>> where > >>> @@ -400,7 +389,7 @@ > >>> getDateTime :: ReadP ClockTime > >>> getDateTime = do > >>> (year, month, day, hour, minutes, seconds) <- readDateTime > >>> - return (mkClockTime year month day hour minutes seconds currTZ) > >>> + return (mkClockTime year month day hour minutes seconds 0) > >>> > >>> fromSqlValue _ _ = Nothing > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > htoolkit-users mailing list > htoolkit-users(a)lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/htoolkit-users > -- http://ofb.net/~frederik/

4 8

Re: Working with multiple time zones
by Ashley Yakeley 18 Feb '08

18 Feb '08

Don Stewart wrote: > Perhaps we should get a binding to tzset in the unix library? That's probably preferable to calling tzset() before every localtime_r. But perhaps we want a call that combines the putenv and the tzset, just so it exposes fewer implementation details. This is essentially an interface to the zoneinfo database rather than time functionality as such. Ideally there would be a better C interface to zoneinfo that wouldn't involve mucking around with (global) environment variables. -- Ashley Yakeley Seattle, WA

1 0

[Alex] Feature request: Create a customizable state monad AlexState
by Alain Cremieux 17 Feb '08

17 Feb '08

Hi, In Alex, very often I need to include elements in the basic monad of the wrapper ALEX_MONAD. For instance if I want to record the level of a comment, or a list of the comments. Several years ago I hacked the file wrapper.hs to include my own elements, by defining another entry point: #ifdef ALEX_MONAD data AlexState = AlexState { alex_pos :: !AlexPosn, -- position at current input location alex_inp :: String, -- the current input alex_chr :: !Char, -- the character before the input alex_scd :: !Int -- the current startcode */#ifdef ALEX_MONAD_USER_STATE , alex_ust :: AlexUserState -- defined in the user program #endif /* } and runAlex :: String -> Alex a -> Either String a runAlex input (Alex f) = case f (AlexState {alex_pos = alexStartPos, alex_inp = input, alex_chr = '\n', */#ifdef ALEX_MONAD_USER_STATE alex_ust = alexInitUserState, #endif /* alex_scd = 0}) of Left msg -> Left msg Right ( _, a ) -> Right a This way I can define my own elements, and initialize them. Now I'd like to have this change integrated in the development tree. Has anyone another idea as how to obtain the same result (knowing that now there are 2 variants, whith normal chars and with bytestrings) ? How would other Alex users do that ? Thanks, Alain

1 0

Proposal: Add Text.Read.maybeRead :: Read a => String -> Maybe a
by Don Stewart 16 Feb '08

16 Feb '08

This function is typically defined once per project. So its about time this safe variant of 'read' made it into the base. maybeRead :: Read a => String -> Maybe a maybeRead s = case reads s of [(x, "")] -> Just x _ -> Nothing Consideration period: 1 week. Patch to Text.Read attached.

18 31

Pretty class as alternative for Show class
by Henning Thielemann 15 Feb '08

15 Feb '08

There seems to be the need for (at least) two classes for showing values in Haskell: One class that outputs values in a way that can be copied and pasted into Haskell programs. This is useful in GHCi. Another class (say Pretty) that shows values in a prettily formatted way. Current libraries use the Show class for both applications depending on the taste of the author. A third class would be useful for outputting values with markup, say HTML or LaTeX. The Show class converts to String, the Pretty class could convert to a pretty printer data type like Doc. The Show class should output all data that is necessary to reconstruct the value, but not necessarily in the form of the underlying data structure, but maybe using appropriate generating functions. Thus e.g. 'show' for Doubles does not need a parameter for precision, it just emits all available digits. On the other hand a Pretty class should have such parameters (maybe then requiring a multi-parameter type class with functional dependency from value type to formatting style type). A nice infix operator like (//) like in the GSL wrapper could flatten the Pretty data to String, and thus is easy to use in GHCi. Matrix> fiboMatrix matrix [[1,0], [1,1]] Matrix> fiboMatrix // 3 /1.000 0.000\ \1.000 1.000/ where (//) :: Pretty style value => value -> style -> IO String x // p = putStr (Doc.toString (Pretty.toDoc p x)) class Pretty style value | value -> style where toDoc :: style -> value -> Doc If such a separation would become consensus then libraries could better separate the concerns of pretty output and re-usable output.

5 10

Labelled graphs in fgl
by Henning Thielemann 14 Feb '08

14 Feb '08

Is there a reason why graphs in fgl are labeled by default, and unlabelled graphs must be constructed by labeling with () ? In my opinion the more natural design would be to number the nodes and edges with an Enum type instead of Int and to retrieve the labels from an array. This would support the "from simple to complex" design.

2 5

Re: Data.Ord and Heaps
by Ross Paterson 14 Feb '08

14 Feb '08

[redirecting to libraries] On Mon, Feb 04, 2008 at 01:09:00PM +0100, Stephan Friedrichs wrote: > apfelmus(a)quantentunnel.de wrote: >> [newtype Ord a => Reverse a = Reverse { unReverse :: a }] >> >> Yes. I mean "non-standard" in the software-reuse sense, i.e. Ord is for >> user-defined orderings and should be the only such mechanism in order >> to enable reuse. In fact, Data.Heap clearly shows that Data.Ord is >> currently missing functionality. > > Ah, now I see. The entire ordering policy mechanism - no matter how it > is going to be solved - belongs into Data.Ord and not in Data.Heap. As > soon as Data.Ord provides a solution, I'll use it in Data.Heap. Unfortunately the Data.Ord module is constrained to use only Haskell 98 features, because it is used by all implementations. So it could include the above Reverse (also mentioned in the group-by HW paper), but not the various OrdPolicy proposals. They could go in a separate module, though. Off on a tangent: the containers package, which collects portable implementations of general purpose data structures, could use a good priority queue implementation. However it is also constrained to use only Haskell 98 features, which would rule out using policies. Also, using weight-biased leftist trees (Cho & Sahni, 1996) instead of Knuth's rank-biased leftist trees would offer O(1) size for free. And they come out slightly faster in my test runs.

2 1