Poly/ML

polyml@lists.polyml.org

2 participants
711 discussions

Re: Interactive behaviour and ctrl-c
by thomas.ridge＠cl.cam.ac.uk 13 May '08

13 May '08

Just to follow up my previous post, the first problem also appears under mosml. The behaviour would result if the Interrupt exception were being caught during the proof step and not propagated. The tactic I am using is SIMP. Grepping the HOL sources reveals many places where an arbitrary exception is caught and not propagated. This would prevent Interrupts from being propagated. There has also been some effort to ensure that general exceptions are not caught (e.g. HOL_ERR is caught rather than a general exception). Given that there are many potential places in HOL where Interrupt could be caught and not thrown, is there any reason to believe that the HOL code is not wrong in these places? The second problem, where polyml loses the ability to call system functions, remains. T 2008/5/12 Tom Ridge <thomas.ridge(a)cl.cam.ac.uk>: > Dear All, > > I have been using HOL with Scott Owens' new port to polyml. To > interrupt a proof that takes a long time, I hit ctrl-c. I get the > polyml #> prompt straightaway. However, if I type f (raise an > exception) then nothing happens for a long while. This time seems to > increase the longer the interrupted proof step has been running. > > Eventually the system seems to return, but it is not clear whether an > exception has been raised or not. Certainly the goalstate is now > different. On the other hand, the goal state should not result if the > proof step completed (but this is likely some erroneous behaviour at > the HOL level). > > There is another problem though. The HOL/polyml top-level is usable > for cut-and-pasting. However, I can't invoke any commands that involve > system calls, otherwise I get a "Function system failed" message (from > libpolyml/process_env.cpp). This is a problem because the xemacs sml > mode writes commands to a temporary file and "use"s them into the > top-level. > > Does anyone have any ideas what might be causing these two problems? > > Thanks > > Tom >

2 1

timing issue with polyml 5 on windows
by rdharrison2＠hotmail.co.uk 12 May '08

12 May '08

Hi all, = I've been running polyml 5.1 on windows from a C# application as a child pr= ocess, redirecting standard input/output via the startupinfo and am having = a timing performance problem... = I have found that when sending small SML commands to polyml (that take 10s = of milliseconds to execute) the output only comes back after about 500ms. T= his seems to be because threads in polyml on windows wait for 500ms = in Processes::ThreadPauseForIO before re-polling for input. Is this correct= , and if so can I get around this? (looking at the source code suggests I m= ay be able to send polyml a message to kick it into action at the appropria= te time?) = regards, Richard Harrison. = _________________________________________________________________ Be a Hero and Win with Iron Man http://clk.atdmt.com/UKM/go/msnnkmgl0010000009ukm/direct/01/-------------- = next part -------------- An HTML attachment was scrubbed... URL: http://lists.inf.ed.ac.uk/mailman/private/polyml/attachments/20080501/= d1d4d5a7/attachment.html

3 2

Exception- Subscript unexpectedly raised while compiling
by ldixon＠inf.ed.ac.uk 11 May '08

11 May '08

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 While compiling a large program I came across the following rather strange error message: Exception- Subscript unexpectedly raised while compiling Exception- Fail "Exception- Subscript unexpectedly raised while compiling" raised After a bit of hacking I managed to reduce it to the following program: (which uses Isabelle's pretty printing library, and which I tested using Isabelle2007) fun foo l s x i = [] @ map (fn (i,s) => Pretty.block [(case x of NONE => Pretty.str "a" | SOME openi => (if (openi = i) then Pretty.str "b" else Pretty.str "c")), Pretty.str "d"]) l; This is running PolyML 5.1... is this a known issue fixed in more recent cvs snapshots? My workaround was to move the case statement into it's own function. Then it compiles ok. cheers, lucas -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFIIJ7ciUn2r+81A2gRAlswAJ4/zXeHsGEGYlghC5EwGiFoj3ikxACgpl/l ntaCRqc04AreddbBV/2yYZ8= =kQQi -----END PGP SIGNATURE-----

2 1

Building a console PolyML application on windows
by r.harrison＠eris.qinetiq.com 16 Apr '08

16 Apr '08

Hi David, I've been testing PolyML 5.1 on windows and was wondering how I should go about creating a console PolyML application. I've built polyml using Visual Studio 2005 and the Visual Studio 6 workspace. This worked with hardly any problems. Visual studio 2005 converted the .dsp to a solution without any warnings. Building the solution failed however, as two of the custom build commands needed the $(InputPath) variables double-quoting to protect spaces in the filenames. After this the build succeeded and I was able to export a simple function (fib.obj) and link it using: link /OUT:fib.exe /SUBSYSTEM:WINDOWS fib.obj PolyLib.lib PolyMainLib.lib libcmtd.lib This produces a windows executable using your console GUI (using WinMain as entry point which calls PolyWinMain). What I would really like is an executable or DLL I can use from another application. I don't have much experience with windows programming and in particular DDE which you use for the console GUI so I realise I may be asking a silly question. I would be happy with a simple console application with interaction through stdin, stdout and stderr, so my question is - what's the best way of building such an application? Am I right in thinking that the main runtime is entered through polymain() in PolyLib? Could this be exported along with PolyWinMain from PolyLib.dll and a main() stub be provided in PolyMain.lib. (I've noticed that if you call WinMain with a null pointer for the hInstance parameter you get an (apparently working) console application but this is surely a bad idea and unnecessary! ;-)) regards, Richard.

2 1

compilation problem on x86_64/Debian
by vlaja＠ida.liu.se 08 Apr '08

08 Apr '08

Hello, everyone! I have the following problem when I try to compile PolyML: gmake[2]: Entering directory `/home/vlaja/soft/polyml.5.1/libpolyml' /bin/sh ../libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I. -I.. -Wall -O3 -MT x86_dep.lo -MD -MP -MF .deps/x86_dep.Tpo -c -o x86_dep.lo x86_dep.cpp g++ -DHAVE_CONFIG_H -I. -I.. -Wall -O3 -MT x86_dep.lo -MD -MP -MF ..deps/x86_dep.Tpo -c x86_dep.cpp -fPIC -DPIC -o .libs/x86_dep.o x86_dep.cpp: In member function `virtual bool X86Dependent::GetPCandSPFromContext(TaskData*, ucontext_t*, PolyWord*&, byte*&)': x86_dep.cpp:769: error: `REG_RIP' undeclared (first use this function) x86_dep.cpp:769: error: (Each undeclared identifier is reported only once for each function it appears in.) x86_dep.cpp:770: error: `REG_RSP' undeclared (first use this function) gmake[2]: *** [x86_dep.lo] Error 1 gmake[2]: Leaving directory `/home/vlaja/soft/polyml.5.1/libpolyml' Additional information about the system: brutalix <308> uname -a Linux brutalix 2.6.8-12-amd64-k8-smp #1 SMP Tue Dec 5 23:23:41 UTC 2006 x86_64 GNU/Linux brutalix <302> gcc --version gcc (GCC) 3.3.5 (Debian 1:3.3.5-13) I suspect that gcc version is too old. In particular, there is no GNU Standard C++ Library v3 installed (that is why I cannot use the pre-compiled version provided with Isabelle). Is that the source of the problem? Thanks for any help, Vlad

1 0

Controlling printing of exceptions.
by rda＠lemma-one.com 26 Mar '08

26 Mar '08

David, I've also sent this to the mailing list in case any other users have comments. If you write your own read-eval-print loop using PolyML.compiler, you may want to treat exceptions raised during the eval stage specially. However you can't currently stop Poly/ML printing out the exception. It seems appropriate to me to use negative values of the error depth control to do this. I get the effect I want with the following patch: --- mlsource/MLCompiler/COMPILER_BODY.ML- 2006-11-06 13:20:50.000000000 +0000 +++ mlsource/MLCompiler/COMPILER_BODY.ML 2008-02-12 13:22:56.000000000 +0000 @@ -641,14 +641,17 @@ val pstream = prettyPrint (77, lexPrint lex); in - ppBeginBlock pstream (3, false); - ppAddString pstream "Exception-"; - ppBreak pstream (1, 0); - VALUEOPS.printStruct (exceptionData, - TYPETREE.exnType, ! errorDepth, pstream, printEnv); - ppBreak pstream (1, 0); - ppAddString pstream "raised"; - ppEndBlock pstream (); + if ! errorDepth >= 0 + then ( + ppBeginBlock pstream (3, false); + ppAddString pstream "Exception- "; + ppBreak pstream (1, 0); + VALUEOPS.printStruct (exceptionData, + TYPETREE.exnType, ! errorDepth, pstream, printEnv); + ppBreak pstream (1, 0); + ppAddString pstream "raised"; + ppEndBlock pstream () + ) else (); (* print out timing info *) printTimes (parseTime,pass2Time,treeTime,codeTime,endRunTime) I note that the error depth also influences syntax-checking and type-checking error messages, but I can't see why anyone would want to suppress those. Regards, Rob.

3 7

[SPAM] Calling Poly/ML functions from C
by jstanley＠galois.com 12 Mar '08

12 Mar '08

Hi folks, I've seen the ForeignTest.{c,sml} example of passing callbacks to C functions using Poly/ML's C FFI (ie, via things like the FUNCTION2 conversion). Is there a way to do this without calling C from Poly/ML? Asked another way, is there a straightforward way to start the Poly/ML runtime from a C main() function, and then call Poly/ML functions (perhaps only those which have been registered via some kind of callback registration mechanism) from C? If it's possible, pointers to documentation and/or example code would be immensely appreciated. Thanks, Joel

2 1

builds on mingw + msys
by Michael.Norrish＠nicta.com.au 12 Mar '08

12 Mar '08

Does the latest version of Poly build on Windows if you have minw and msys to hand? Thanks, Michael.

2 3

Windows programming in Poly/ML
by kyleyuty＠gmail.com 05 Mar '08

05 Mar '08

Hello, everyone I'm a student from Ireland and I'm working on my final year project with Poly/ML recently. Here is a question: dlgProc(dial, WM_COMMAND{notifyCode =3D 0, wId=3D1 (* OK button *), ...}, (= )) =3D (* When the OK button is pressed we end the dialogue. *) let val _=3DEndDialog(dial, 1); val tempValue =3D "Hello"; in (LRESINT 0, ()) end I have made a dialog box and I want to define a variable "tempValue" assigned by value "Hello" when I close the dialog box. But the variable could not be defined after I closed the dialog box. Please help me out. Thank you very much. Regards, Yuting Zhong

2 1

REAL signature
by fluet＠tti-c.org 29 Feb '08

29 Feb '08

The implementation of the REAL signature in the Basis Library omits the "where type real = real" constraint on the Math : MATH sub-structure. Without the constraint, when writing a functor with a R : REAL argument, the body of the functor cannot "see" that R.real and R.Math.real are the same type, so the type checker does not allow R.+(R.Math.e, R.Math.pi).

2 1

← Newer
1
...
55
56
57
58
59
60
61
...
72
Older →

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Poly/ML