January 2004 - Glasgow-haskell-users

GHC-6.2 and Template Haskell on OS X
by Gregory Wright 29 Jan '04

29 Jan '04

Hi, I have a question about ghc-6.2 and template haskell on Mac OS X. I think it's a ghc configuration or build issue, so I'm posting here instead of the TH list. I have ghc-6.2 built under darwinports. This uses a 6.0.1 binary to bootstrap the 6.2 build. I haven't updated the binary bootstrap compiler because it seems to work under both Jaguar (10.2.x) and Panther (10.3.x). The portfile builds ghc-6.2 with the command 'make all'. The resulting ghc-6.2 has worked fine until I tried to build a program using TH. I tried to build the simplest example--the one in the User's guide-- and it failed with: > ghc --make -fth main.hs -o main Chasing modules from: main.hs Compiling Printf ( ./Printf.hs, ./Printf.o ) ./Printf.hs:23: Type constructor or class not in scope: `Expr' ./Printf.hs:26: Variable not in scope: `string' ./Printf.hs:30: Type constructor or class not in scope: `Expr' > This must be a simple problem. The other thing to note about the darwinports build is that it stages the installation into a temporary directory so it can do an inventory of the files that will be installed. If a path in the staging directory gets compiled in, instead of the final installation directory, all manner of unhappiness can result. Any one have any ideas? The answer is probably staring me in the face but I'm not able to see it. Best Wishes, Greg Gregory Wright Antiope Associates LLC 18 Clay Street Fair Haven, New Jersey 07704 gwright(a)antiope.com

1 0

RE: Problem with ghc on Windows ME
by Richard Black 29 Jan '04

29 Jan '04

If you want to know what it does just refer to the source; it is and always has been distributed with Visual C. See C:\Program Files\Microsoft Visual Studio .NET 2003\vc7\crt\src\stdargv.c Richard -----Original Message----- From: Simon Peyton-Jones Sent: 29 January 2004 15:18 To: Simon Marlow; 'Claus Reinke'; 'glasgow-haskell-users(a)haskell.org' Cc: Richard Black; Simon Peyton-Jones Subject: RE: Problem with ghc on Windows ME Well found! Still, I found I could not understand the rules. But I think the story this Convert \" to \\\" otherwise convert " to \" otherwise do no conversion I think it's probably easiest to express this directly: translate :: String -> String translate str = '"' : go str where go ('\\' : '"' : s) = '\\' : '\\' : '\\' : '"' : go s go ('"' : s) = '\\' : '"' : go s go (c:s) = c : go s go [] = ['"'] Simon | -----Original Message----- | From: Simon Marlow | Sent: 29 January 2004 14:53 | To: Claus Reinke; Simon Peyton-Jones; glasgow-haskell-users(a)haskell.org | Subject: RE: Problem with ghc on Windows ME | | After Googling around a bit, I found this description of exactly how Windows interprets command lines | in the C runtime: | | http://msdn.microsoft.com/library/default.asp?url=/library/en-us/vccelng /htm/progs_12.asp | | As you can see, the rules are indeed very strange, but they are invertible. I think this code should do | the trick: | | translate :: String -> String | translate str = '"' : snd (foldr escape (True,"\"") str) | where escape '"' (_, str) = (True, '\\' : '"' : str) | escape '\\' (True, str) = (True, '\\' : '\\' : str) | escape '\\' (False, str) = (False, '\\' : str) | escape c (_, str) = (False, c : str) | | Cheers, | Simon | | > -----Original Message----- | > From: glasgow-haskell-users-bounces(a)haskell.org | > [mailto:glasgow-haskell-users-bounces@haskell.org] On Behalf | > Of Simon Marlow | > Sent: 29 January 2004 11:31 | > To: Claus Reinke; Simon Peyton-Jones; | > glasgow-haskell-users(a)haskell.org | > Subject: RE: Problem with ghc on Windows ME | > | > | > > So does that mean functionality has been lost in the move from | > > ghc-6.0.1? | > | > No, the behaviour is improved in 6.2 because we're now | > careful to escape | > quotes before passing the command-line to the operating system. | > | > I think confusion has arisen because Simon pointed out that there's no | > way, on Windows, to pass this sequence of characters as an | > argument to a | > command: | > | > \" | > | > This is not entirely true, I just discovered that the | > sequence \\\" will | > turn into \". But in general, the sequence \\ does not turn into \. | > Windows' command-line quoting rules are highly mysterious, to say the | > least. | > | > So the patch I posted is not complete. It looks like rawSystem should | > additionally translate \" as \\\". | > | > > Because that (on win98) happily accepts such useful | > > contraptions as | > > | > > System.Cmd.rawSystem "\"c:\\silly | > > space\\Vim\\vim62\\gvim.exe\" --servername GVIM --remote-send | > > \":echo \\\"hi \\\\\\\" there\\\"<cr>\"" >>= print | > | > The idea is that you should pass to rawSystem *exactly* the | > command and | > arguments you want to run, with no quoting except that required by the | > Haskell lexical syntax. You'll notice that rawSystem now takes the | > command and list of arguments separately in 6.2. The | > implementation of | > rawSystem will attempt to compensate for the internal translation that | > Windows does on the command-line; on Unix no translation is necessary. | > So clients of rawSystem should be more portable, because they | > don't have | > to know about quoting that happens under the hood. | > | > Your example will probably work once it's reformulated for | > 6.2's version | > of rawSystem, and once we fix rawSystem in light of the \\\" hack I | > discovered above. | > | > Cheers, | > Simon | > _______________________________________________ | > Glasgow-haskell-users mailing list | > Glasgow-haskell-users(a)haskell.org | > http://www.haskell.org/mailman/listinfo/glasgow-haskell-users | >

1 0

RE: Problem with ghc on Windows ME
by Simon Peyton-Jones 29 Jan '04

29 Jan '04

Well found! Still, I found I could not understand the rules. But I think the story this Convert \" to \\\" otherwise convert " to \" otherwise do no conversion I think it's probably easiest to express this directly: translate :: String -> String translate str = '"' : go str where go ('\\' : '"' : s) = '\\' : '\\' : '\\' : '"' : go s go ('"' : s) = '\\' : '"' : go s go (c:s) = c : go s go [] = ['"'] Simon | -----Original Message----- | From: Simon Marlow | Sent: 29 January 2004 14:53 | To: Claus Reinke; Simon Peyton-Jones; glasgow-haskell-users(a)haskell.org | Subject: RE: Problem with ghc on Windows ME | | After Googling around a bit, I found this description of exactly how Windows interprets command lines | in the C runtime: | | http://msdn.microsoft.com/library/default.asp?url=/library/en-us/vccelng /htm/progs_12.asp | | As you can see, the rules are indeed very strange, but they are invertible. I think this code should do | the trick: | | translate :: String -> String | translate str = '"' : snd (foldr escape (True,"\"") str) | where escape '"' (_, str) = (True, '\\' : '"' : str) | escape '\\' (True, str) = (True, '\\' : '\\' : str) | escape '\\' (False, str) = (False, '\\' : str) | escape c (_, str) = (False, c : str) | | Cheers, | Simon | | > -----Original Message----- | > From: glasgow-haskell-users-bounces(a)haskell.org | > [mailto:glasgow-haskell-users-bounces@haskell.org] On Behalf | > Of Simon Marlow | > Sent: 29 January 2004 11:31 | > To: Claus Reinke; Simon Peyton-Jones; | > glasgow-haskell-users(a)haskell.org | > Subject: RE: Problem with ghc on Windows ME | > | > | > > So does that mean functionality has been lost in the move from | > > ghc-6.0.1? | > | > No, the behaviour is improved in 6.2 because we're now | > careful to escape | > quotes before passing the command-line to the operating system. | > | > I think confusion has arisen because Simon pointed out that there's no | > way, on Windows, to pass this sequence of characters as an | > argument to a | > command: | > | > \" | > | > This is not entirely true, I just discovered that the | > sequence \\\" will | > turn into \". But in general, the sequence \\ does not turn into \. | > Windows' command-line quoting rules are highly mysterious, to say the | > least. | > | > So the patch I posted is not complete. It looks like rawSystem should | > additionally translate \" as \\\". | > | > > Because that (on win98) happily accepts such useful | > > contraptions as | > > | > > System.Cmd.rawSystem "\"c:\\silly | > > space\\Vim\\vim62\\gvim.exe\" --servername GVIM --remote-send | > > \":echo \\\"hi \\\\\\\" there\\\"<cr>\"" >>= print | > | > The idea is that you should pass to rawSystem *exactly* the | > command and | > arguments you want to run, with no quoting except that required by the | > Haskell lexical syntax. You'll notice that rawSystem now takes the | > command and list of arguments separately in 6.2. The | > implementation of | > rawSystem will attempt to compensate for the internal translation that | > Windows does on the command-line; on Unix no translation is necessary. | > So clients of rawSystem should be more portable, because they | > don't have | > to know about quoting that happens under the hood. | > | > Your example will probably work once it's reformulated for | > 6.2's version | > of rawSystem, and once we fix rawSystem in light of the \\\" hack I | > discovered above. | > | > Cheers, | > Simon | > _______________________________________________ | > Glasgow-haskell-users mailing list | > Glasgow-haskell-users(a)haskell.org | > http://www.haskell.org/mailman/listinfo/glasgow-haskell-users | >

1 0

RE: Problem with ghc on Windows ME
by Simon Peyton-Jones 29 Jan '04

29 Jan '04

It's called 'raw' because it is supposed to get the arguments through *unmodified* to the called program. No file globbing, no escape stuff, nothing. Plain 'system' invokes the shell, and therefore does file globbing. Anything about escaping in rawSystem is an internal implementation matter S | -----Original Message----- | From: glasgow-haskell-users-bounces(a)haskell.org [mailto:glasgow-haskell-users- | bounces(a)haskell.org] On Behalf Of Malcolm Wallace | Sent: 29 January 2004 13:48 | To: glasgow-haskell-users(a)haskell.org | Subject: Re: Problem with ghc on Windows ME | | "Simon Marlow" <simonmar(a)microsoft.com> writes: | | > The implementation of | > rawSystem will attempt to compensate for the internal translation that | > Windows does on the command-line; on Unix no translation is necessary. | > So clients of rawSystem should be more portable, because they don't have | > to know about quoting that happens under the hood. | | This is probably a silly naming question, but aren't the names of | 'System.system' and 'rawSystem' the wrong way round? On the face | of things, it would seem sensible for System.system to have the more | portable behaviour, and rawSystem to be dependent on local conventions. | After all, System.system is defined in the official Libraries Report | making no mention of any portability problems, whereas the name 'raw' | suggests that no interpretation/quoting is going on, and hence it | may be highly dependent on the underlying system. | | Regards, | Malcolm | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users(a)haskell.org | http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

1 0

RE: Problem with ghc on Windows ME
by Simon Marlow 29 Jan '04

29 Jan '04

> > The > implementation of > > rawSystem will attempt to compensate for the internal > translation that > > Windows does on the command-line; on Unix no translation is > necessary. > > So clients of rawSystem should be more portable, because > they don't have > > to know about quoting that happens under the hood. > > This is probably a silly naming question, but aren't the names of > 'System.system' and 'rawSystem' the wrong way round? On the face > of things, it would seem sensible for System.system to have the more > portable behaviour, and rawSystem to be dependent on local > conventions. > After all, System.system is defined in the official Libraries Report > making no mention of any portability problems, whereas the name 'raw' > suggests that no interpretation/quoting is going on, and hence it > may be highly dependent on the underlying system. Indeed. The existing behaviour for System.system is (I assume) derived from the POSIX sytem(), which also executes a command using the shell. The naming is sensible in one respect: system goes via the shell, whereas rawSystem bypasses the shell. Unfortunately we can't easily change System.system to be the raw version, because they have different types and to do so would break lots of code. System.Process will have a more reasonable interface, when I get around to finishing it. Cheers, Simon

1 0

detecting memory size?
by Ketil Malde 29 Jan '04

29 Jan '04

Hi, Is it possible to detect memory (i.e. RAM, not virtual memory) size from inside a Haskell program (so that I can keep my program from growing too large with consequent thrashing)? And if so, to what degree of portability? -kzm -- If I haven't seen further, it is by standing in the footprints of giants

4 7

Re: Network problem with ghc on WinXP
by robin abraham 29 Jan '04

29 Jan '04

Works like a charm now :) Thank you for the insight. Robin. >Hi there, > >looks like a network byte-order vs host byte-order gotcha. >Never use the PortNum constructor, but declare 'portnum' >to have type PortNumber and simply drop the use of PortNum >in your code alltogether. Alternatively, use intToPortNumber >to translate between Int and PortNumber. > >hth >--sigbjorn > >----- Original Message ----- >From: "robin abraham" <abraharo(a)cs.orst.edu> >To: <glasgow-haskell-users(a)haskell.org> >Sent: Wednesday, January 28, 2004 21:57 >Subject: Network problem with ghc on WinXP > > >> Hi, >> >> I have ghc-6.0.1 on WinXP and Solaris. I have a simple echo server >(server.hs >> given below) and client (client.hs given below) and I encounter the >following: >> 1) server.hs compiled and running on Solaris: >> a) client.hs (Solaris) can connect. >> b) client.hs (WinXP) cannot connect. >> c) telnet (WinXP) can connect. >> d) telnet (Solaris) can connect. >> e) Scan of port 3000 shows server is listening. >> >> 2) server.hs compiled and running on WinXP: >> a) client.hs (Solaris) cannot connect. >> b) client.hs (WinXP) can connect. >> c) telnet (WinXP) cannot connect. >> d) telnet (Solaris) cannot connect. >> e) Scan of ports does not show server. >> >> Basically, when the server is running on WinXP, only the Haskell program >> client.hs (also running on the same WinXP machine) can "converse" with it >> through the socket connection. To verify this, I wrote a client in C# - >from the >> WinXP machine, it can connect to server running on my Solaris machine but >not to >> server on the same WinXP machine. >> >> Why isn't the server program on WinXP not behaving itself? Any >help/guidance >> would be highly appreciated. >> >> Thank you. >> Robin. >> >> -- server.hs >> module Main where >> >> import SocketPrim >> import Concurrent >> import System (getArgs,exitFailure) >> import Exception(finally) >> import MVar >> import IO >> >> server_sock :: IO (Socket) >> server_sock = do >> s <- socket AF_INET Stream 6 >> setSocketOption s ReuseAddr 1 >> bindSocket s (SockAddrInet (PortNum portnum) iNADDR_ANY) >> listen s 2 >> return s >> >> echo_server :: Socket -> IO () >> echo_server s = do >> (s', clientAddr) <- accept s >> h <- socketToHandle s' ReadWriteMode >> proc <- read_data s' 0 >> putStrLn ("server processed "++(show proc)++" bytes") >> sClose s' >> where >> read_data sock totalbytes = do >> str <- recv sock 18 >> putStrLn ("Server recv: " ++ str) >> if ((length str) >= 18) >> then do >> putStrLn ("Server read: " ++ str) >> writ <- send sock str >> putStrLn ("Server wrote: " ++ str) >> read_data sock $! (totalbytes+(length $! str)) >> else do >> putStrLn ("server read: " ++ str) >> return totalbytes >> >> message = "Hello there sailor" >> portnum = 3000 >> >> main = withSocketsDo $ do { >> ~[n] <- getArgs; >> ssock <- server_sock; >> s <- myForkIO (echo_server ssock); >> join s; >> putStrLn "join s"; >> } >> >> myForkIO :: IO () -> IO (MVar ()) >> myForkIO io = do >> mvar <- newEmptyMVar >> forkIO (io `finally` putMVar mvar ()) >> return mvar >> >> join :: MVar () -> IO () >> join mvar = readMVar mvar >> >> -- end of server.hs >> >> -- *********************************************** >> >> -- client.hs >> >> module Main where >> >> import SocketPrim >> import Concurrent >> import System (getArgs,exitFailure) >> import Exception(finally) >> import MVar >> import IO >> >> local = "128.193.39.108" >> message = "Hello there sailor" >> portnum = 3000 >> >> client_sock = do >> s <- socket AF_INET Stream 6 >> ia <- inet_addr local >> connect s (SockAddrInet (PortNum portnum) ia) >> return s >> >> echo_client n = do >> s <- client_sock >> drop <- server_echo s n >> sClose s >> where >> server_echo sock n = if n > 0 >> then do >> send sock message >> putStrLn ("Client wrote: " ++ message) >> str <- recv sock 19 >> if (str /= message) >> then do >> putStrLn ("Client read error: " ++ str) >> exitFailure >> else do >> putStrLn ("Client read success") >> server_echo sock (n-1) >> else do >> putStrLn "Client read nil" >> return [] >> >> main = withSocketsDo $ do >> ~[n] <- getArgs >> c <- myForkIO (echo_client (read n::Int)) >> join c >> putStrLn "join c" >> >> myForkIO :: IO () -> IO (MVar ()) >> myForkIO io = do >> mvar <- newEmptyMVar >> forkIO (io `finally` putMVar mvar ()) >> return mvar >> >> join :: MVar () -> IO () >> join mvar = readMVar mvar >> >> -- end of client.hs >>

1 0

Network problem with ghc on WinXP
by robin abraham 29 Jan '04

29 Jan '04

Hi, I have ghc-6.0.1 on WinXP and Solaris. I have a simple echo server (server.hs given below) and client (client.hs given below) and I encounter the following: 1) server.hs compiled and running on Solaris: a) client.hs (Solaris) can connect. b) client.hs (WinXP) cannot connect. c) telnet (WinXP) can connect. d) telnet (Solaris) can connect. e) Scan of port 3000 shows server is listening. 2) server.hs compiled and running on WinXP: a) client.hs (Solaris) cannot connect. b) client.hs (WinXP) can connect. c) telnet (WinXP) cannot connect. d) telnet (Solaris) cannot connect. e) Scan of ports does not show server. Basically, when the server is running on WinXP, only the Haskell program client.hs (also running on the same WinXP machine) can "converse" with it through the socket connection. To verify this, I wrote a client in C# - from the WinXP machine, it can connect to server running on my Solaris machine but not to server on the same WinXP machine. Why isn't the server program on WinXP not behaving itself? Any help/guidance would be highly appreciated. Thank you. Robin. -- server.hs module Main where import SocketPrim import Concurrent import System (getArgs,exitFailure) import Exception(finally) import MVar import IO server_sock :: IO (Socket) server_sock = do s <- socket AF_INET Stream 6 setSocketOption s ReuseAddr 1 bindSocket s (SockAddrInet (PortNum portnum) iNADDR_ANY) listen s 2 return s echo_server :: Socket -> IO () echo_server s = do (s', clientAddr) <- accept s h <- socketToHandle s' ReadWriteMode proc <- read_data s' 0 putStrLn ("server processed "++(show proc)++" bytes") sClose s' where read_data sock totalbytes = do str <- recv sock 18 putStrLn ("Server recv: " ++ str) if ((length str) >= 18) then do putStrLn ("Server read: " ++ str) writ <- send sock str putStrLn ("Server wrote: " ++ str) read_data sock $! (totalbytes+(length $! str)) else do putStrLn ("server read: " ++ str) return totalbytes message = "Hello there sailor" portnum = 3000 main = withSocketsDo $ do { ~[n] <- getArgs; ssock <- server_sock; s <- myForkIO (echo_server ssock); join s; putStrLn "join s"; } myForkIO :: IO () -> IO (MVar ()) myForkIO io = do mvar <- newEmptyMVar forkIO (io `finally` putMVar mvar ()) return mvar join :: MVar () -> IO () join mvar = readMVar mvar -- end of server.hs -- *********************************************** -- client.hs module Main where import SocketPrim import Concurrent import System (getArgs,exitFailure) import Exception(finally) import MVar import IO local = "128.193.39.108" message = "Hello there sailor" portnum = 3000 client_sock = do s <- socket AF_INET Stream 6 ia <- inet_addr local connect s (SockAddrInet (PortNum portnum) ia) return s echo_client n = do s <- client_sock drop <- server_echo s n sClose s where server_echo sock n = if n > 0 then do send sock message putStrLn ("Client wrote: " ++ message) str <- recv sock 19 if (str /= message) then do putStrLn ("Client read error: " ++ str) exitFailure else do putStrLn ("Client read success") server_echo sock (n-1) else do putStrLn "Client read nil" return [] main = withSocketsDo $ do ~[n] <- getArgs c <- myForkIO (echo_client (read n::Int)) join c putStrLn "join c" myForkIO :: IO () -> IO (MVar ()) myForkIO io = do mvar <- newEmptyMVar forkIO (io `finally` putMVar mvar ()) return mvar join :: MVar () -> IO () join mvar = readMVar mvar -- end of client.hs

2 1

RE: Problem with ghc on Windows ME
by Simon Peyton-Jones 28 Jan '04

28 Jan '04

| For system, I do need doublequotes in arguments (e.g., when passing | complex option flags to the programs invoked, or when the arguments | contain strings). Can that really always be avoided for rawSystem? Perhaps the comment is misleading. You can indeed put quotes in arguments (they get escaped). It's backslash followed by quote that I don't know how to get into an argument Simon

2 1

RE: Generics...(ish)
by MR K P SCHUPKE 28 Jan '04

28 Jan '04

I have read the "Scrap Your Boilerplate: A Practical Design Pattern for Generic Programming" and have worked out how to do it. This seems much better than using Template Haskell for this kind of problem. I used mkQ and extQ to convert from various types that are allowed in the record structure, to an algebraic enumeration type that codes the valid types. This allows using "everything (++)" to convert a record (tuple type) to a List of type-names in an enumeration. This is a very neet and compact way of doing it. This enables me to define a database table in haskell (where a record represents a table with columns of the given types), for example: data TestTable = TestTable { column1 :: ColumnType Int, column2 :: ColumnType Float, column3 :: ColumnType String } type ColumnType a = String -- use phantom types to statically type check sql... testTable :: TestTable testTable = TestTable { column1 = "id", column2 = "aFloatValue" column3 = "aString" } Using generics I can now generate a list: [("id",SqlInt),("aFloatValue",SqlFloat),("aString",SqlString)] which can be used to check the Haskell version of the table against the version in the database, or create the table if none exists... This is for a kind of port of the HaskellDB stuff to ghc, although it is more of a complete reimplementation using the same idea. My aim is to use a monad transformer, so that db operations can be interleaved with other IO. Regards, Keean Schupke.

1 0