Poly/ML September 2015

polyml@lists.polyml.org

9 participants
12 discussions

linking polyML modules to C
by dtopham＠gmail.com 09 Nov '18

09 Nov '18

After more experimenting with GUI in SML, I would like to now try calling an SML module from a C program. That way, the GUI could be built using traditional imperative techniques and the program logic could be written in SML. I see how to call a C function from SML, but and not so clear on how to call an SML function from a C program. I want to build code using polyc and then link with a C main program, but could anyone show me a simple example of calling that function and getting its return value? for example, I have defined: fun sum i n f = if i > n then 0.0 else f i + sum (i+1) n f ; fun int2real x = real x ; in linkToC.sml and #include <iostream> #include <fstream> std::ofstream ofs; int main() { std::cout << sum(1,10,int2real) << std::endl; ofs.open("table.dat"); for(int i = 0; i < 20; i++) ofs << i << '\t' << sum(1,i,int2real) << std::endl; } in linkToC.cpp Looking in the polyc script, I thought adding the -c option might allow me to build the object code which I could then link to the sml functions, but it complains about not having a main. Could someone point me in the right direction? Thanks, Dave On Mon, Dec 15, 2014 at 11:44 AM, David Topham <dtopham at gmail.com> wrote: > David, Thank you so much! I really appreciate your help with this. Your > sample below builds fine with polyc on my system (PolyML 5.5.2). I am using > a minimal Linux distribution named TinyCore ( > http://distro.ibiblio.org/tinycorelinux/) which comes with OpenMotif > 2.3.3 It runs in a virtual machine within another computer only taking up > a few hundred MBs! > > I am going to use it to help my students explore Discrete Math next > semester (this textbook uses SML for that purpose: > http://cs.wheaton.edu/~tvandrun/dmfp/). > > I also found a reasonably current project using PolyML and Motif here: > http://www.lemma-one.com/ProofPower/index/ > > ...however it was built on PolyML 4 and has not yet been ported to the > newer version (e.g. depends on PolyML.commit which no longer seems to be > supported). > > Happy Holidays to everyone in the PolyML community! > > -David Topham > > >> Re: GUI Interface (David Matthews) >> >> I've experimented with building a stand-alone executable with Motif and >> succeeded in getting it to work for me using polyc. (It required >> building Poly/ML with --with-x). I took the example from >> http://www.polyml.org/docs/Motif.html and wrapped it up in a function. >> I did find a problem, though. It looks as though the function has to >> suspend itself with something like Posix.Process.pause otherwise nothing >> happens. I seem to recall that the Motif stuff is handled on a separate >> thread to allow the REPL to continue to accept commands. With a >> stand-alone application there isn't a REPL so without the "pause" it >> terminates immediately. >> >> David >> >> open XWindows ; >> open Motif ; >> >> fun main() = >> let >> val shell = >> XtAppInitialise "" "xed" "Editor" [] [XmNwidth 400, XmNheight 400] ; >> >> val main = XmCreateMainWindow shell "main" [] ; >> >> val bar = XmCreateMenuBar main "bar" [] ; >> >> val fileMenu = XmCreateCascadeButton bar "file" [XmNlabelString "File"] ; >> val editMenu = XmCreateCascadeButton bar "edit" [XmNlabelString "Edit"] ; >> val viewMenu = XmCreateCascadeButton bar "view" [XmNlabelString "View"] ; >> val helpMenu = XmCreateCascadeButton bar "help" [XmNlabelString "Help"] ; >> >> val command = XmCreateText main "command" [XmNeditMode >> XmSINGLE_LINE_EDIT] ; >> >> val hscroll = XmCreateScrollBar main "hscroll" [XmNorientation >> XmHORIZONTAL] ; >> val vscroll = XmCreateScrollBar main "vscroll" [XmNorientation >> XmVERTICAL] ; >> >> val work = XmCreateDrawingArea main "work" [] ; >> in >> >> XtManageChildren [fileMenu, editMenu, viewMenu, helpMenu] ; >> XtManageChildren [bar, command, hscroll, vscroll, work] ; >> >> XmMainWindowSetAreas main bar command hscroll vscroll work ; >> >> XtManageChild main ; >> XtRealizeWidget shell; >> Posix.Process.pause() >> end; >> >> >> ------------------------------ >> >> _______________________________________________ >> polyml mailing list >> polyml at inf.ed.ac.uk >> http://lists.inf.ed.ac.uk/mailman/listinfo/polyml >> >> End of polyml Digest, Vol 110, Issue 12 >> *************************************** >> >> -- >> The University of Edinburgh is a charitable body, registered in >> Scotland, with registration number SC005336. >> >>

2 3

Inspecting structures
by ramana.kumar＠gmail.com 27 Jan '16

27 Jan '16

Is there a way to look at the contents (i.e. signature) of a structure without opening it? (I suppose you could just open it then call a lot of PolyML.Compiler.forgetXXX, but that's quite tedious) Also, is it possible to retrieve the long path of an identifier in scope? Something that will, for example, take the rev in scope and say that it came from List. Finally, what about searching for an identifier that would come into scope if you opened some structures?

3 5

Using a finalizer with multiple arguments
by phil.clayton＠lineone.net 12 Oct '15

12 Oct '15

I am trying to create an SML binding to a C function that returns an array that the caller must free. Usually, there is a free function that takes just the array (as a pointer) which can be attached as a finalizer with CInterface.setFinal. I have encountered a case [1] where the caller must also pass the size of the array, returned when the array is created, to the free function. Simplifying the example, we have, for some C type Elem: Elem *new (..., int *n_elems); /* n_elems is an 'out' parameter */ void free (Elem *elems, int n_elems); and want an SML function like val new : ... -> elem vector Unfortunately, the function given to CInterface.setFinal is called with only one argument, the vol that is being finalized. Therefore this free function cannot be used. Does the current FFI architecture allow a variant of setFinal that passes extra arguments to the finalization function? For example: val setFinal1 : sym -> 'a Conversion -> vol -> 'a -> unit This isn't particularly common so is probably not a show-stopper. Another benefit could be enabling use of functions g_slice_alloc and g_slice_free1 that needs the number of bytes to free: https://developer.gnome.org/glib/stable/glib-Memory-Slices.html Thanks, Phil 1. The C function in question is gtk_target_table_new_from_list that returns an array and its size. The array should be freed with gtk_target_table_free which should be passed the size. See: - https://developer.gnome.org/gtk3/stable/gtk3-Selections.html#gtk-target-tab… - https://developer.gnome.org/gtk3/stable/gtk3-Selections.html#gtk-target-tab…

4 21

InternalError: AllocStore found but last allocation not complete
by phil.clayton＠lineone.net 08 Oct '15

08 Oct '15

After integrating the new Finalizable module in a library, I am getting an internal error from the compiler: Exception- InternalError: AllocStore found but last allocation not complete raised while compiling This affects Poly/ML 5.5.2 and the very latest commit with the compiler rebuilt (twice). An example is attached showing this. I have reduced this as far as I can but still a few small library files are required. The file 'polyml.sml' is the top-level file that uses the other files. Regards, Phil

2 3

polyc with different base executable
by Michael.Norrish＠nicta.com.au 24 Sep '15

24 Sep '15

It would be useful for HOL4 if polyc could be used with a different base executable. At the moment, the shell-script is hardcoded to use e.g., /usr/local/bin/poly. If HOL has created a different heap H with a lot of preloaded context, but for which the main function is still PolyML.rootFunction, creating a new heap H' that is a delta on top of H seems to become difficult: my impression is that all of the source files that went into generating H have to be re-used before the new stuff making up H' can be looked at. Or is there another way of getting what I need? I'd like to dispense with explicit calls in our own code to PolyML.export, and then needing to call the linker, but we do want to be able to build heaps in this way, layer by layer. Thanks, Michael ________________________________ The information in this e-mail may be confidential and subject to legal professional privilege and/or copyright. National ICT Australia Limited accepts no liability for any damage caused by this email or its attachments.

3 3

FFI overhead
by artella.coding＠googlemail.com 21 Sep '15

21 Sep '15

Hi, I thought that I would try to speed up the SML code at http://stackoverflow.com/questions/32425267/how-to-improving-array-benchmar… by using the FFI, but this results in significant slowdown. Non ffi code : ************************************************************************* ************************************************************************* ************************************************************************* val size:int = 50000; val loops:int = 30; val cap:int = 50000; val data = IntArray.array(size,0); fun loop () = let fun loopI i = if i = size then let val _ = () in IntArray.update(data,0,IntArray.sub(data,size-1)); () end else let val previous = IntArray.sub(data,i-1) val use = if previous > cap then 0 else previous in IntArray.update(data,i,use+1); loopI (i+1) end in loopI 1 end fun benchmarkRun () = let fun bench i = if i = loops then () else let val _ = () in loop (); bench (i+1) end in bench 1 end fun sum (i,value) = if i = size then value else sum(i+1,value+Array.sub(data,i)) fun main () = let val _ = () in benchmarkRun(); print (Int.toString (sum (0,0))); print "\n" end (*val _ = main ()*) ************************************************************************* ************************************************************************* ************************************************************************* FFI code : c code : ************************************************************************* ************************************************************************* ************************************************************************* //intArray.c #include <stdlib.h> #include <stdio.h> typedef struct _intArray { int size; int* arr; } intArray; intArray* createIntArray(int size){ int i; intArray* p = (intArray*) malloc (sizeof(intArray)); p->arr = (int*) malloc (size*sizeof(int)); for(i=0; i<size; i++){ p->arr[i] = 0; } p->size = size; return p; } void destroyIntArray(intArray* p){ free (p->arr); free (p); } void setIntArray(intArray* p, int elem, int val){ p->arr[elem] = val; } int getIntArray(intArray *p, int elem){ return p->arr[elem]; } int getSumIntArray(intArray* p){ int sum = 0; int i; int size = p->size; for(i=0; i<size; i++){ sum += p->arr[i]; } return sum; } ************************************************************************* ************************************************************************* ************************************************************************* ml code : ************************************************************************* ************************************************************************* ************************************************************************* open CInterface; val lib = load_lib "./intArray.so"; val get = get_sym "./intArray.so"; val PINTARR = POINTER; val c1 = call1 (get "createIntArray") INT PINTARR val c2 = call3 (get "setIntArray") (PINTARR,INT,INT) VOID val c3 = call2 (get "getIntArray") (PINTARR,INT) INT val c4 = call1 (get "getSumIntArray") (PINTARR) INT fun c_createIntArray (size) = c1 (size); fun c_setIntArray (p,elem,value) = c2 (p,elem,value); fun c_getIntArray (p,elem) = c3 (p,elem); fun c_getSumIntArray (p) = c4 (p); val size:int = 50000; val loops:int = 30; val cap:int = 50000; fun loop (pData2) = let fun loopI i = if i = size then let val _ = () in c_setIntArray(pData2,0,c_getIntArray(pData2,size-1)); () end else let val previous = c_getIntArray(pData2,i-1); val use = if previous > cap then 0 else previous in c_setIntArray(pData2,i,use+1); loopI (i+1) end in loopI 1 end fun benchmarkRun (pData2) = let fun bench i = if i = loops then () else let val _ = () in loop (pData2); bench (i+1) end in bench 1 end fun main () = let val pData = c_createIntArray(size); val final = load_sym lib "destroyIntArray"; in setFinal final pData; benchmarkRun(pData); print (Int.toString (c_getSumIntArray (pData))); print "\n" end ************************************************************************* ************************************************************************* ************************************************************************* The times are : a)for non ffi sml : 0.09s b)for ffi sml : 11.8s Is there any way I can improve the speeds on the ffi code? Thanks

3 4

Windows GUI and console
by David.Matthews＠prolingua.co.uk 15 Sep '15

15 Sep '15

The issue of the Windows console window has come up from time to time and I've had another look at the code and made a few minor changes. The first change is to provide an option to "./configure" to allow Poly/ML to be built as a "console" application rather than a "windows" application. Setting "--disable-windows-gui" passes -mconsole to the linker rather than -mwindows. This sets the appropriate bits in the EXE header. As well as affecting the linking of "poly" itself it also affects executables built with "polyc". The default remains "--enable-windows-gui" to build as a "windows" application. The second change is that when Poly is built as a Windows application, the default, it puts up a GUI window only if standard output has not been provided. Previously it put up a GUI window unless both standard input and standard output were provided. If it creates a GUI window for output it also uses it for input unless there is an explicit standard input provided. It may be helpful to provide a bit of background. I have the impression that there are two different classes of users when it comes to Poly/ML on Windows. Users from a Windows background expect the application to come with an installer, a Windows GUI and to support the standard copy-and-paste, in particular with the standard control-C/control-V key bindings. Users from a Unix background are much more used to running applications in a console window and are more likely to build applications from source. The problem is to satisfy both classes of user. Executables in Windows can be either "windows" or "console" applications depending on the "subsystem" setting in the EXE header. "console" applications are always started with a cmd.exe-style console unless they're already in a console. "windows" applications are expected to create their own GUI. My impression is that "console" applications are regarded as a legacy of DOS and not well supported. That's why I've made "windows" mode the default. Nevertheless there may be cases where someone would like to build "poly" as a console application which is the reason for adding the option. David

2 5

Fwd: Course programming language
by David.Matthews＠prolingua.co.uk 15 Sep '15

15 Sep '15

Well of course there is a Windows installer for the latest release version of Poly/ML at http://sourceforge.net/projects/polyml/files/polyml/5.5.2/ . The source for the installer is included in the source files but it needs Visual Studio and Wix ( http://wixtoolset.org/ ). David On 15/09/2015 15:00, Artella Coding wrote: > I saw your comment at > http://lists.inf.ed.ac.uk/pipermail/polyml/2015-September/001620.html > > "I have the impression that there are two different classes of users when > it comes to Poly/ML on Windows. Users from a Windows background expect the > application to come with an installer, a Windows GUI and to support the > standard copy-and-paste, in particular with the standard > control-C/control-V key bindings." > > and thought I would forward you the below, because it confirms what you say. > > ---------- Forwarded message ---------- > From: Andrej Bauer <andrej.bauer at andrej.com> > Date: Mon, Sep 14, 2015 at 7:20 PM > Subject: Re: Course programming language > To: Artella Coding <artella.coding at googlemail.com> > > > This is a completely different thing: until there is a Windows-style > installer where students just click through things it's a show > stopper. That is the reality of today's stdents (of math). > > On Mon, Sep 14, 2015 at 7:45 PM, Artella Coding > <artella.coding at googlemail.com> wrote: >> https://twitter.com/andrejbauer/status/643158821083418625 >> >> p.s. if you were also considering Standard ML you could try the > instructions >> David Matthew gave to build from source. >> >> http://lists.inf.ed.ac.uk/pipermail/polyml/2015-August/001593.html >> http://lists.inf.ed.ac.uk/pipermail/polyml/2015-August/001602.html >> >> All you need is msys2 and the rest is pretty seamless. >> >> >

1 0

Program seems to run instead of just compiling
by artella.coding＠googlemail.com 14 Sep '15

14 Sep '15

Hi, Suppose that I have the following code : (* Test.sml *) fun f() = print "-------test------"; val test = f(); Then if I compile this via "polyc -c Test.sml" the program actually runs before emitting an error message : -------test------poly: : error: Value or constructor (main) has not been declared Found near PolyML.export (List.nth (CommandLine.arguments (), 3), main) Static Errors Why does the program attempt to run instead of merely compiling? Thanks

2 1

The History of Standard ML: Ideas, Principles, Culture
by makarius＠sketis.net 12 Sep '15

12 Sep '15

Here is an interesting talk at the ML Workshop http://www.mlworkshop.org/ml2015 at ICFP 2015: David MacQueen The History of Standard ML: Ideas, Principles, Culture https://www.youtube.com/watch?v=NVEgyJCTee4 It briefly mentions the heroic efforts by David Matthews to make the first complete SML implementation, based on the existing Poly compiler. Glad to see that this very first SML project still defines the cutting edge until today! Makarius

1 0

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 September 2015