OpenEuphoria

Re: Modified Interpreter

• Posted by Derek Parnell <ddparnell at bigpond.com> Jul 12, 2003
• 478 views

----- Original Message -----=20
From: "Igor Kachan" <kinz at peterlink.ru>
To: "EUforum" <EUforum at topica.com>
Subject: Re: Modified Interpreter


>=20
>=20
> Hello Pete,
>=20
> > From: Pete Lomax <petelomax at blueyonder.co.uk>
> > To: EUforum <EUforum at topica.com>
> > Subject: Re: Modified Interpreter
> > Sent: 12 july 2003 y. 12:13
> >=20
> >=20
> > On Sat, 12 Jul 2003 10:14:07 +0400, Igor Kachan <kinz at peterlink.ru>
> > wrote:
> >=20
> > <snip>
> > >Z.EX:4
> > >A namespace qualifier is needed to resolve z.
> > >z is defined as a global symbol in:
> > >    C:\DOWNLO~1\EXM\a.e
> > >    C:\DOWNLO~1\EXM\b.e
> > >    C:\DOWNLO~1\EXM\c.e
> > That is as expected, z.e needs to include a.e, b.e, and c.e with=20
> > namespaces. Given that z.e is your code, that should not be a=20
> > problem.
>=20
> a.e, b.e, c.e  are all my code, not only z.e.=20
> So, there is no problem with any transformations of this code.
> I wanted b.e, c.e, z.ex can do some job with the global=20
> integer z of a.e.
> But interpreter doesn't see this *GLOBAL* symbol
> inside b.e and c.e, it sees global z only inside z.ex after=20
> the first reference to global z in z.ex, not in b.e and c.e.
> This is a problem. Too late error message.
>=20
> > If having just typed in (your own code) and run
> > include b.e
> > ? z
> > and got a namespace qualifier required error, then immediately
> > changing it to:
> > include b.e as b
> > ? b:z
> > is easy peasy, No?
>=20
> Peasy, not easy. I have tons of my own big libs of different versions=20
> which are dependant one on  another and may have the globals with
> the same names. Libs are *my* and proggy is *my*.
>=20
> > <snip and shuffle:>
>=20
> Ok, I'll just reshuffle :]
>=20
> > >So, v2.4 doesn't run my simple program properly,
> > >it doesn't see global integer z, defined in a.e,
> > >inside b.e and c.e.=20
> > It sees all of them, and does not know which=20
> >  you mean.
>=20
> If it doesn't know which one, I prefer it asks me, doesn't
> ignore *my* globals, doesn't print just any first one.
>=20
> > Do *you* know which z you mean?
>=20
> Yes I do know. The very first global z of a.e,
> then the second global z of b.e and then the third global z of c.e.
>=20
> > The problem being addressed here is if a.e, b.e,=20
> > and c.e are not your code but some third party code,=20
> > you do not want to have to edit them,
> > and certainly not re-edit them each time the author=20
> > releases a new improved version.=20
> > That used to be the case,=20
>=20
> Pete, I know.
>=20
> > it should not be any more.
>=20
> Touch wood Pete.
>=20
> > >If I run z.ex on v2.2, I get:
> > >
> > >C:\EUPHORIA>ex Z.EX
> > >1
> > >C:\EUPHORIA\b.e:1
> > >attempt to redefine z - defined already in C:\EUPHORIA\a.e
> >=20
> > Yes, standard 2.4 is already an improvement over 2.2(!)
>=20
> Ok, viva 2.4, I love it, but it issues it's warning and=20
> aborting too late and prints too many different global zs.=20
> I like 2.2 for that my proggy.
>=20
> > >Global integer z of a.e is invisible in b.e and c.e.
> > Not quite. They would be, only eu 2.4 is automatically=20
> > (and correctly) figuring out which z is meant in each=20
> > case (the local one).
>=20
> Why do you say "correctly" if 2.4 doesn't want to know
> my first global z? Matt too doesn't want to know ... =20
> Is it "correct" and "automatic"?
>=20
> > If you did not define z in say b.e, then it (b.e) would use the one
<snip>

I take it back. Now I'm *sure* you have no understanding of the rules.

It seems that you are saying that Euphoria should issue an error message =
when the b.e file tries to declare a global 'z', because a.e has already =
defined 'z'. Is this the problem that you see?

--=20
Derek