Stefan, Data.Derive is a most awesome piece of code! Is there soemething in DrIFT that you did not like that made you write it? Thanks a lot! On Apr 5, 2007, at 12:48 AM, Stefan O'Rear wrote:

Data.Derive can do this. In an attempt to avoid munging the relevent files they are attached.

-- http://wagerlabs.com/

Installed derive, trying to load it with ghci -package derive Loading package base ... linking ... done. Loading package template-haskell ... linking ... done. Loading package FilePath-0.11 ... linking ... done. ghc-6.6: unknown symbol `_derivezm0zi1_DataziDeriveziPeephole_zdf7_closure' Loading package derive-0.1 ... linking ... ghc-6.6: unable to load package `derive-0.1' What am I doing wrong? Thanks, Joel -- http://wagerlabs.com/

This is the exposed modules portion of derive.cabal. I had to remove the empty lines since Cabal was complaining about them. I suspect one of these lines had Data.Derive.Peephole in it. Exposed-Modules: Data.Derive Data.Derive.FixedPpr Data.Derive.SYB Data.Derive.TH Data.Derive.Binary Data.Derive.BinaryDefer Data.Derive.Eq Data.Derive.Play On Apr 5, 2007, at 12:46 PM, Joel Reymont wrote:

FunParser.hs:4:7: Could not find module `Data.Derive.Peephole': Use -v to see a list of the files searched for.

-- http://wagerlabs.com/

This is in Language.Haskell.TH.Syntax which is imported at the top of Data/Derive/TH.hs so I don't understand the cause of the error instance Functor Q where fmap f (Q x) = Q (fmap f x) Copying the above into TH.hs gives me Preprocessing library derive-0.1... Preprocessing executables for derive-0.1... Building derive-0.1... [4 of 9] Compiling Data.Derive.TH ( Data/Derive/TH.hs, dist/build/ Data/Derive/TH.o ) Data/Derive/TH.hs:23:10: Not in scope: data constructor `Q' Data/Derive/TH.hs:23:17: Not in scope: data constructor `Q' But Q is exported by Languave.Haskell.TH.Syntax !!! Any suggestions? Thanks, Joel -- http://wagerlabs.com/

Joel Reymont wrote:

This is in Language.Haskell.TH.Syntax which is imported at the top of Data/Derive/TH.hs so I don't understand the cause of the error

instance Functor Q where fmap f (Q x) = Q (fmap f x)

...

Any suggestions?

Since Q is a Monad, you can make the instance

instance Functor Q where fmap = liftM

But Q is exported by Languave.Haskell.TH.Syntax !!!

Only the type constructor is exported, not the data constructor. Twan

With derive compiled and installed I thought I would change the code a bit and try it... ghci -fth -v0 -e '$( _derive_print_instance makeFunParser '"''"'Foo )' baz.hs baz.hs:30:3: Not in scope: `a1' Any help is appreciated! Thanks, Joel --- FunParser.hs: module FunParser where import Data.Derive import Data.Derive.Peephole import Data.List import Text.ParserCombinators.Parsec ( CharParser ) makeFunParser = Derivation drv "FunParser" drv dat@(DataDef name arity ctors) = simple_instance "FunParser" dat [funN "parse" [ sclause [] body ] ] where body = l1 "choice" $ lst [ clause con | con <- ctors ] clause con = l1 "reserved" (lit (trim (ctorName con))) >>: args con (ctorArity con) trim = reverse . takeWhile (/= '.') . reverse args ct 0 = return' (ctp ct 'a') args ct k = l1 "char" (lit '(') >>: args' ct k 0 args' ct remn seen = l0 "parse" >>=: (('a' : show seen) ->: args'' ct (remn-1) (seen+1)) args'' ct 0 seen = l1 "char" (lit ')') >>: return' (ctp ct 'a') args'' ct k seen = l1 "char" (lit ',') >>: args' ct k seen class FunParser a where parse :: CharParser s a baz.hs: import Text.ParserCombinators.Parsec hiding ( parse ) import qualified Text.ParserCombinators.Parsec.Token as T import Text.ParserCombinators.Parsec.Language( emptyDef ) import Data.Derive.TH import FunParser data NumExpr = Int Integer | Num Double instance FunParser NumExpr where parse = numExpr data Foo = Foo NumExpr lexer = T.makeTokenParser emptyDef identifier = T.identifier lexer reserved = T.reserved lexer integer = T.integer lexer float = T.float lexer numExpr :: GenParser Char a NumExpr numExpr = choice [ integer >>= return . Int , float >>= return . Num ] $( derive makeFunParser ''Foo ) -- http://wagerlabs.com/

On Thu, Apr 05, 2007 at 02:47:21PM +0100, Joel Reymont wrote:

Here's the output from -ddump-splices (thanks Saizan for the tip).

It's returning a1 instead of a0.

ghci -fth -e '$( _derive_print_instance makeFunParser '"''"'Foo )' baz.hs -ddump-splices baz.hs:1:0: baz.hs:1:0: Splicing declarations derive makeFunParser 'Foo ======> baz.hs:30:3-28 instance {FunParser Main.Foo} where [] { parse = choice [(>>) (reserved ['F', 'o', 'o']) ((>>) (char '(') ((>>=) parse (\ a0 -> (>>) (char ')') (return (Main.Foo a1)))))] }

baz.hs:30:3: Not in scope: `a1'

Sorry for the late multiple reply, I just spent seven hours sleeping... I am not the maintainer of Data.Derive, nor did I write the majority of the nice code; Neil Mitchell did it, you can ask him "why replace DrIFT". However, using abstract syntax trees WAS my idea. First, _derive_print_instance will never give you a TH splice error, since it always evaluates to an empty list of declarations. It uses the TH 'runIO' facility such that type-checking a file using _derive_print_instance will emit the instances to standard output as a side effect. So the error is coming from the $(derive) in baz.hs, if you have more errors try commenting it out. (you'll get bogus code on stdout, but at least it will be completly haskell!) _derive_print_instance was not intended to be a debugging aid, although it certainly works well in that capacity. The intent is that it will be used when the standalone driver is rewritten to use TH, which I intend to do not long after I can (Neil is out of communication for a week with intent to continue hacking Derive; I'm taking this as a repository lock). Yes, we do use type classes to implement recursion across types. This seems to be a very standard idiom in Haskell, used by Show, Read, Eq, Ord, NFData, Arbitrary, and doubtless many more.

Here's a complete working example. There seems to be an error in the parser but everything is derived fine. *Main> run fooParser "Foo(10)" Foo (Int 10) *Main> run fooParser "Foo(10.5)" parse error at (line 1, column 7): unexpected "." expecting digit or ")" FunParser.hs: Replace "show seen" with "show (seen + 1)" in args' baz.hs: import Text.ParserCombinators.Parsec hiding ( parse ) import qualified Text.ParserCombinators.Parsec as P import qualified Text.ParserCombinators.Parsec.Token as T import Text.ParserCombinators.Parsec.Language( emptyDef ) import Data.Derive.TH import FunParser data NumExpr = Int Integer | Num Double deriving Show instance FunParser NumExpr where parse = numExpr instance FunParser Integer where parse = T.integer lexer instance FunParser Int where parse = T.integer lexer >>= return . fromInteger data Foo = Foo NumExpr | Bar Int NumExpr deriving Show lexer = T.makeTokenParser emptyDef identifier = T.identifier lexer reserved = T.reserved lexer integer = T.integer lexer float = T.float lexer numExpr :: GenParser Char a NumExpr numExpr = choice [ integer >>= return . Int , float >>= return . Num ] $( derive makeFunParser ''Foo ) fooParser :: GenParser Char a Foo fooParser = parse run p input = case (P.parse p "" input) of Left err -> putStr "parse error at " >> print err Right x -> print x

Shouldn't this work just as well? numExpr = choice [ try $ float >>= return . Num , integer >>= return . Int ] It works on "Foo(10.345)" but not on "Bar(10, 103.34)". On Apr 5, 2007, at 4:09 PM, Stefan O'Rear wrote:

...

numExpr :: GenParser Char a NumExpr numExpr = do sg <- lexeme sign nf <- natOrFloat return $ either (Int . sg) (Nat . sg) nf

It seems silly that there is no signed version of natOrFloat predefined, any Parsec experts?

Stefan

-- http://wagerlabs.com/

John Meacham wrote:

You might want to note that DrIFT used to be called derive before it (amicably) changed its name due to a conflict with a product of the same name.

DrIFT might consider changing back, since TI has now officially stopped support for said product [1]. Jacques [1] http://scientificcomputing.blogspot.com/2006/11/derive-to-be-discontinued.ht....

Jules Bean wrote:

data paramType = JNum | JBool | JStr

paramParser JNum = numExpr paramParser JBool = boolExpr paramParser JStr = strExpr

unary x pt = reserved (quasiShow (x undefined)) >> parens (paramParser pt) >>= return . x

strCall = choice ( map unary [ELDateToString,TextGetString,LowerStr,UpperStr,Spaces] )

Oops. unary (x,pt) = reserved (quasiShow (x undefined)) >> parens (paramParser pt) >>= return . x strCall = choice ( map unary [(ELDateToString,JNum),(TextGetString,JNum), (LowerStr,JStr),(UpperStr,JStr),(Spaces,JStr)] ) (have to specify the param types) Jules