It looks a bit strange to me... for example, compare these two lines of code:

    int[] a;
    ...
    a ~= 1;      // works
    a = a ~ 1;   // doesn't work!

Looks weird... why can't I concatenate an int array and a single interger? Furthermore, I believe that the following should work as well:

    a = 1 ~ a;
    a = 1 ~ 2;

Take a look at the latter - binary ~ should make an array out of its operands if they aren't already arrays. This could be a very useful operation, since it practically replaces array literals:

    double max(double[] n) { ... }
    double a, b, c, d;
    ...
    d = max(a ~ b ~ c);

Using suffices, array of almost any built-in type can be constructed:

      1 ~   2 ~   3        int[]
     1U ~  2U ~  3U        uint[]
     1L ~  2L ~  3L        long[]
    1UL ~ 2UL ~ 3UL        ulong[]
    1.0 ~ 2.0 ~ 3.0        double[]
    ...............................

Only byte and short don't have their suffices (BTW, why
not? "b" and "s" would suite just fine).

What do you think of this, Walter? Is it a good idea?

It does sound like a logical extension. I'll add it in to the list! -Walter

"Pavel Minayev" <evilone@omen.ru> wrote in message news:a5qbcq$1kss$1@digitaldaemon.com...
> It looks a bit strange to me... for example, compare these two lines of code:
>
>     int[] a;
>     ...
>     a ~= 1;      // works
>     a = a ~ 1;   // doesn't work!
>
> Looks weird... why can't I concatenate an int array and a single interger? Furthermore, I believe that the following should work as well:
>
>     a = 1 ~ a;
>     a = 1 ~ 2;
>
> Take a look at the latter - binary ~ should make an array out of its operands if they aren't already arrays. This could be a very useful operation, since it practically replaces array literals:
>
>     double max(double[] n) { ... }
>     double a, b, c, d;
>     ...
>     d = max(a ~ b ~ c);
>
> Using suffices, array of almost any built-in type can be constructed:
>
>       1 ~   2 ~   3        int[]
>      1U ~  2U ~  3U        uint[]
>      1L ~  2L ~  3L        long[]
>     1UL ~ 2UL ~ 3UL        ulong[]
>     1.0 ~ 2.0 ~ 3.0        double[]
>     ...............................
>
> Only byte and short don't have their suffices (BTW, why
> not? "b" and "s" would suite just fine).
>
> What do you think of this, Walter? Is it a good idea?
>
>

"Walter" <walter@digitalmars.com> wrote in message news:a5r2rl$1uiq$1@digitaldaemon.com...

> It does sound like a logical extension. I'll add it in to the
list! -Walter

Great to hear that!

By the way, with all those feature requests around... =)
You've mentioned several times that some things will be implemented
sooner or later, but probably won't get into version 1. Do you
have any "feature list" of what gets there, then? Templates are
out, I guess, what about operator overloading? In other words, what
should we expect by DMD 1.0 release, whenever it happens?

"Pavel Minayev" <evilone@omen.ru> wrote in message news:a5r4r7$1vic$1@digitaldaemon.com...
> "Walter" <walter@digitalmars.com> wrote in message
> news:a5r2rl$1uiq$1@digitaldaemon.com...
> By the way, with all those feature requests around... =)
> You've mentioned several times that some things will be implemented
> sooner or later, but probably won't get into version 1. Do you
> have any "feature list" of what gets there, then? Templates are
> out, I guess, what about operator overloading? In other words, what
> should we expect by DMD 1.0 release, whenever it happens?

What's in the spec right now should be 1.0. -Walter

"Pavel Minayev" <evilone@omen.ru> wrote in message news:a5qbcq$1kss$1@digitaldaemon.com...
> Only byte and short don't have their suffices (BTW, why
> not? "b" and "s" would suite just fine).

Just realized "b" won't work, it's a hex digit. There goes consistency!

"Walter" <walter@digitalmars.com> wrote in message news:a5rr4o$28uv$1@digitaldaemon.com...
>
> "Pavel Minayev" <evilone@omen.ru> wrote in message news:a5qbcq$1kss$1@digitaldaemon.com...
> > Only byte and short don't have their suffices (BTW, why
> > not? "b" and "s" would suite just fine).
>
> Just realized "b" won't work, it's a hex digit. There goes consistency!
>
>

So bt then? Or by?


--
Stijn
OddesE_XYZ@hotmail.com
http://OddesE.cjb.net
__________________________________________
Remove _XYZ from my address when replying by mail

"Walter" <walter@digitalmars.com> wrote in message news:a5rmmi$26rg$1@digitaldaemon.com...

> What's in the spec right now should be 1.0. -Walter

Any (probably approximate) release date? Or "when it's done"?

"Pavel Minayev" <evilone@omen.ru> wrote in message news:a5u2l6$fll$1@digitaldaemon.com...
> "Walter" <walter@digitalmars.com> wrote in message news:a5rmmi$26rg$1@digitaldaemon.com...
> > What's in the spec right now should be 1.0. -Walter
> Any (probably approximate) release date? Or "when it's done"?

The release date was 6 months ago <g>.