[m-dev.] for review: add MC++ implementation of library and runtime

[Tyson Dowd]

On 10-Dec-2000, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> On 04-Dec-2000, Tyson Dowd <trd at cs.mu.OZ.AU> wrote:
> > +% io__do_open(File, Mode, ResultCode, Stream, IO0, IO1).
> > +%	Attempts to open a file in the specified mode.
> > +%	ResultCode is 0 for success, -1 for failure.
> > +:- pragma foreign_code("MC++",
> > +	io__do_open(FileName::in, Mode::in, ResultCode::out,
> > +			Stream::out, IO0::di, IO::uo),
> > +			[will_not_call_mercury, thread_safe],
> > +"
> > +	MR_MercuryFile mf = mercury_open(FileName, Mode);
> > +	MR_c_pointer_to_word(Stream, mf);
> > +	update_io(IO0, IO);
> > +").
> 
> That needs to set ResultCode.

This is another cvs diff bug -- 
	ResultCode = (mf ? 0 : -1);
was there all along (I promise).

> 
> > +:- pragma foreign_code("MC++",
> > +	io__get_exit_status(ExitStatus::out, IO0::di, IO::uo),
> > +		will_not_call_mercury, "
> > +	ExitStatus = System::Environment::get_ExitCode();
> > +	MR_Runtime::SORRY(""foreign code for this function"");
> > +	update_io(IO0, IO);
> > +").
> 
> Is the call to SORRY() there needed?
> If so, best to put it at the start, rather than in the middle.

No, it shouldn't be there.

> 
> > +:- pragma foreign_code("C",
> > +	library__version(Version::out), will_not_call_mercury,
> > +"
> >  	MR_ConstString version_string = 
> >  		MR_VERSION "", configured for "" MR_FULLARCH;
> >  	/*
> ...
> > +:- pragma foreign_code("MC++",
> > +	library__version(Version::out), will_not_call_mercury,
> > +"
> > +	Version = String::Concat(MR_VERSION,
> > +		"", configured for "", MR_FULLARCH);
> >  ").
> 
> Doesn't ANSI C string literal concatenation work in MC++?

Fixed. 

Although strictly speaking I'm going to have to change this to
use S"stringbody" style string literals (this makes it a System::String
instead of a char *).  But that didn't work straight away.

> 
> library/math.m:
> > +:- pragma foreign_decl("MC++", "
> >  
> > +	// This is not defined in the .NET Frameworks.
> > +
> > +	#define	ML_FLOAT_LN2		0.69314718055994530941
> > +").
> 
> I don't understand the comment here.
> Why is the code here different for MC++ than it is for C?
> The code for C defines ML_FLOAT_E and ML_FLOAT_PI too.

e and pi are defined in the .NET Math class, but ln2 isn't.

I used e and pi from the .NET Math class.

I thought this would be a better choice, since it would be nice if our
pi corresponded exactly with pi in other languages.

> > +    static int MR_SECTAG_NONE				= 0;
> > +    static int MR_SECTAG_LOCAL				= 1;
> > +    static int MR_SECTAG_REMOTE				= 2;
> 
> Please put comments at the definitions of the corresponding
> enums in the runtime directory saying that they are duplicated here.

Ok, done.

> If you prefer I'm happy to commit all these fixes for the MC++ versions
> of all-solutions stuff as a separate change.

It's fine, I've done them now.


-- 
       Tyson Dowd           # 
                            #  Surreal humour isn't everyone's cup of fur.
     trd at cs.mu.oz.au        # 
http://www.cs.mu.oz.au/~trd #
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------