- Glasgow-haskell-users

Question about error message
by Seth Kurtzberg 17 Aug '02

17 Aug '02

Hello, all There is an error message that I get from GHCI that I do not understand. Usually it is caused by a syntax error or an indentation error, but I have occasionally seen it in other contexts and I'd like to understand what it is trying to tell me. The message is that the last statement in a do construct must be an expression (exact text below). I believe this occurs when the last statement in the do construct has a type other than "IO ()". The documentation (or perhaps Simon Thompson's or Paul Hudak's book; they get jumbled together in my mind) states that all statements in a do construct have type "IO ()"; but this is clearly not correct, as in: do line <- getLine putStrLn line where the first statement has type IO String. Here is an example. This code is useless but it is not obvious to me why it is incorrect: main = do line <- getLine producing the error: <interactive>:1: The last statement in a 'do' construct must be an expression (This also limits the ability to enter monadic programming to the interpreter.) TIA, -- Seth Kurtzberg M. I. S. Corp seth(a)cql.com 1-480-661-1849 (GMT-7)

3 2

RE: (not) catching ^C in ghc-built application
by Simon Marlow 16 Aug '02

16 Aug '02

> > I have a program, written in a mixture of Haskell and C, in > which the > > C part is supposed to handle a ^C interrupt, accomplished > through the > > usual signal() call. But when it is compiled by ghc, for > some reason > > the ghc RTS appears to trap the interrupt itself, and halts the > > program with the message > > prog: interrupted > > instead of allowing the newly installed C signal handler to run. > > I finally found the cause of this. The ghc RTS has a global variable > `interrupted', which is set by its interrupt handler, and tested by > various other parts of the scheduler. > > Guess what - my C routines also have a global variable called > `interrupted', and my own ^C interrupt handler was also setting > this variable. D'oh. If I change the name of my global variable, > then everything works as it ought to. > > I know it's a pain for you, but can I file this as a bug report > in ghc? Stealing simple and obvious global identifiers from the > user's namespace is rather dangerous, and even more so when it is > undocumented. These global "common" variables are a pain in the neck. There are 44 of these in GHC's RTS - we should really at the least initialise them so that you'll get a conflict if you define a global with the same name in your own code. Thanks for the report. Cheers, Simon

1 0

RE: intermediate hc files, where
by Simon Marlow 16 Aug '02

16 Aug '02

> I want to build ghc-5.02.3 from scratch, where can I find the > bootstrap files please? The website says they should be listed among > the downloads for the release, but this isn't so. Bootstrapping from .hc is a particularly tricky process at the moment. Unless you've seen it already, take a look at http://www.haskell.org/ghc/docs/latest/html/building/sec-porting-ghc.htm l What platform are you trying to build on? (.hc files are architecture-specific, except for "unregisterised" .hc files). Cheers, Simon

1 0

ghc 5.04 -- make -prof -auto-all under WinNT
by andreas.marth＠daimlerchrysler.com 15 Aug '02

15 Aug '02

Hallo! I tried to compile the following little program (hallo.hs) with ghc-5.04 --make -prof -auto: import System (getArgs) main :: IO () main= do args <- getArgs if args == [] then error "Aufruf: Filename Dimension1 Dimension2 ..." else prog (head args) (tail args) prog :: FilePath -> [String] -> IO () prog file dimZ = do cont <- readFile file putStr (file ++ "\n") putStr (show dimZ ++ "\n") let result = "Hello world!" writeFile (file++".out") ( result) The system is WinNT and compiling succeeded. But I got an a.out file while I should get a file hallo.exe (according to ghc --help). I renamed the program to hallo.exe and tried it. When I don't give any arguments it behaves like expected. If I give it one or more arguments it crashes. When I compile it only with ghc --make hallo.hs I also get an a.out file but this behaves well when renamed to hallo.exe. What is the problem? And what can I do to use the profiling? Thank, Andreas

1 0

RE: GHCI and Concurrent Haskell?
by Simon Marlow 15 Aug '02

15 Aug '02

> Is it possible to test code using the concurrency features with GHCI? There's one problem: everything stops when the prompt appears in GHCi. This is because our readline interface isn't thread-friendly. Otherwise, concurrency works fine in GHCi. Cheers, Simon

1 0

GHCI and Concurrent Haskell?
by Seth Kurtzberg 15 Aug '02

15 Aug '02

Is it possible to test code using the concurrency features with GHCI? TIA, -- Seth Kurtzberg M. I. S. Corp seth(a)cql.com 1-480-661-1849 (GMT-7)

1 0

intermediate hc files, where
by Ian Zimmerman 14 Aug '02

14 Aug '02

I want to build ghc-5.02.3 from scratch, where can I find the bootstrap files please? The website says they should be listed among the downloads for the release, but this isn't so. Thanks. -- Ian Zimmerman, Oakland, California, U.S.A. GPG: 433BA087 9C0F 194F 203A 63F7 B1B8 6E5A 8CA3 27DB 433B A087 EngSoc adopts market economy: cheap is wasteful, efficient is expensive.

1 0

RE: duplicate symbols in a package?
by Simon Marlow 14 Aug '02

14 Aug '02

> GHCi runtime linker: fatal error: I found a duplicate > definition for symbol > r1p_closure > whilst processing object file > /grp/haskell/lib/ghc-5.02.2/HSHaXml.o > This could be caused by: > * Loading two different object files which export the > same symbol > * Specifying the same object file twice on the GHCi > command line > * An incorrect `package.conf' entry, causing some object to be > loaded twice. > GHCi cannot safely continue in this situation. Exiting > now. Sorry. This bug was fixed in 5.02.3, I believe. Cheers, Simon

1 0

duplicate symbols in a package?
by Malcolm Wallace 14 Aug '02

14 Aug '02

Hi GHC users, I am trying to build a package (HaXml) for ghc that will also work in GHCi. I followed the instructions in the manual, to convert the .a file to a .o: ld -r --whole-archive -o HSHaXml.o libHSHaXml.a and installed, ran ghc-pkg, etc. However, now when I start up GHCi, it throws an error: $ ghci -package HaXml ___ ___ _ / _ \ /\ /\/ __(_) / /_\// /_/ / / | | GHC Interactive, version 5.02.2, for Haskell 98. / /_\\/ __ / /___| | http://www.haskell.org/ghc/ \____/\/ /_/\____/|_| Type :? for help. Loading package std ... linking ... done. Loading package lang ... linking ... done. Loading package concurrent ... linking ... done. Loading package posix ... linking ... done. Loading package util ... linking ... done. Loading package data ... linking ... done. Loading package HaXml ... GHCi runtime linker: fatal error: I found a duplicate definition for symbol r1p_closure whilst processing object file /grp/haskell/lib/ghc-5.02.2/HSHaXml.o This could be caused by: * Loading two different object files which export the same symbol * Specifying the same object file twice on the GHCi command line * An incorrect `package.conf' entry, causing some object to be loaded twice. GHCi cannot safely continue in this situation. Exiting now. Sorry. Indeed, running nm on the HSHaXml.o file shows that there are numerous symbols with multiple definitions. Here are just some of them: r13_closure, r13_entry, r13_info, r13_srt, r19_closure, r19_entry, r19_info, r19_srt, r1B_closure, r1D_closure, r1D_entry, r1D_info, r1D_srt, r1IZ_closure, r1J1_closure, r1J1_entry, r1J1_info, r1J1_srt, r1b_closure, r1b_entry, r1b_info, r1b_srt, r1l_closure, r1l_entry, r1l_info, r1p_closure, r1p_entry, r1p_info, s1JM_entry, s1JM_info, s1JZ_entry, s1JZ_info, s1Le_entry, s1Le_info, Any ideas what these are, and how I can avoid them? Regards, Malcolm

1 0

RE: (not) catching ^C in ghc-built application
by Simon Marlow 13 Aug '02

13 Aug '02

> > For a quick workaround, if you install your own handler > *after* calling > > startupHaskell(), you should be able to override GHC's. > > The thing is, that's what I'm already doing. At least, Haskell is > in charge of the execution, and does quite a lot of things before > calling out to an I/O action written in C, which then sets > the handler. > > My best guess is that the use of signal() is incompatible with the > POSIX sigaction() used by the RTS. That's possible I guess. Could you try using sigaction() instead? Cheers, Simon

5 5