Re: Implicit conversions through purity
Jonathan M Davis: Honestly, I would have considered that to be a bug. Converting the return type to a different level of mutability based on purity is one thing. Automatically casting the return value just because the function is pure is another matter entirely. Clearly, it can work, but it seems incredibly sloppy to me. In foo1 D is working as designed, as this was a desired feature, it has passed the Kenji and Walter review, and it was implemented several months ago. It's a very handy way to create immutable data with pure functions and it's safe, it's safer than assumeUnique that is just a convention. Very recently Walter has further improved this feature, allowing more implicit conversion cases. So it's the opposite of a bug, it saves you from bugs in three different ways. But my question was about the successive foo functions :-) Bye, bearophile
Re: Implicit conversions through purity
On Saturday, April 12, 2014 21:26:01 bearophile wrote:
> Is it possible and a good idea to allow code like the function
> foo2?
>
>
> string foo1(in string s) pure nothrow {
> auto s2 = s.dup;
> s2[0] = 'a';
> return s2; // OK.
> }
Honestly, I would have considered that to be a bug. Converting the return type
to a different level of mutability based on purity is one thing. Automatically
casting the return value just because the function is pure is another matter
entirely. Clearly, it can work, but it seems incredibly sloppy to me.
- Jonathan M Davis
Re: string mixin only works once per block?
These are excellent, thanks! I went ahead and swapped out the tags and I think I am going to implement something like that opDispatch method for all outside libraries now. Very useful! That type validation is particularly cool. You could be rendering almost arbitrary data in like 2 steps...
Implicit conversions through purity
Is it possible and a good idea to allow code like the function
foo2?
string foo1(in string s) pure nothrow {
auto s2 = s.dup;
s2[0] = 'a';
return s2; // OK.
}
void foo2(in string s, ref string sOut) pure nothrow {
auto s2 = s.dup;
s2[0] = 'a';
sOut = s2; // Error: cannot implicitly convert
}
void foo3(in string s, out string sOut) pure nothrow {
auto s2 = s.dup;
s2[0] = 'a';
sOut = s2; // Error: cannot implicitly convert
}
void main() {}
Bye,
bearophile
Re: DList bug or programmer error?
On Saturday, 12 April 2014 at 19:34:11 UTC, Daniel K wrote: This block of code fails the second assert, why? Because DList is buggy. It should pass. I have an open pull that fixes this, FYI.
Re: Local function overloading
On 4/12/14, monarch_dodra wrote:
> I know you can workaround it by putting
> your functions a static members of a dummy struct
You can also use a mixin template:
class A {}
class B {}
mixin template M()
{
void func2(A a) { }
void func2(B b) { }
}
void main()
{
mixin M!();
func2(new A);
func2(new B);
}
DList bug or programmer error?
This block of code fails the second assert, why? Using
stableRemove{Front,Back} does not fix it.
import std.range;
import std.container;
unittest {
auto dl2 = DList!int([2,7]);
dl2.removeFront();
//dl2.stableRemoveFront();
assert(dl2[].walkLength == 1);
dl2.removeBack();
//dl2.stableRemoveBack();
assert(dl2.empty, "not empty?!");
}
void main(){
}
rdmd -unittest file.d
Re: MS ODBC encoding issue
Sam you should put th ODBC code to the github.com , let all world to see your hard work! Frank
Re: Associative arrays, multithreading, Rebindable
On Saturday, 12 April 2014 at 09:18:12 UTC, bearophile wrote: Jack Applegame: Should Rebindable to support associative arrays? I guess so. But how to implement it? Bye, bearophile Related: "Rebindable supports arrays (Which makes no sense)" https://issues.dlang.org/show_bug.cgi?id=12046 Quote: "It *might* make sense with associative arrays though... Not sure"
Re: alias this and constructor
On Saturday, 12 April 2014 at 14:00:06 UTC, Jack Applegame wrote: And this doesn't compile too http://dpaste.dzfl.pl/1d48eed7e0fb The same reason? Looks like it's something different. I think I know the cause. Could you file it in the bug tracker? I will fix it.
Re: Local function overloading
On Saturday, 12 April 2014 at 16:45:02 UTC, Philpax wrote:
While trying to overload a function in local/function scope, I
ran into this behaviour: http://dpaste.dzfl.pl/b4e8b9ddf78a and
I was wondering what the cause was.
As far as I can tell, this should be fine in global scope (and
it is), but I'm curious as to why it doesn't work inside a
function.
I *think* it has something to do with how name-mangling is done.
I don't know the details. I know you can workaround it by putting
your functions a static members of a dummy struct:
http://dpaste.dzfl.pl/268e3d2d4427
class A {}
class B {}
void main()
{
static struct Dummy
{
static void func2(A a){}
static void func2(B b){}
}
alias func2 = Dummy.func2;
A a;
func2(a);
}
Local function overloading
While trying to overload a function in local/function scope, I ran into this behaviour: http://dpaste.dzfl.pl/b4e8b9ddf78a and I was wondering what the cause was. As far as I can tell, this should be fine in global scope (and it is), but I'm curious as to why it doesn't work inside a function.
Re: alias this and constructor
And this doesn't compile too http://dpaste.dzfl.pl/1d48eed7e0fb The same reason?
Re: alias this and constructor
On Saturday, 12 April 2014 at 13:08:48 UTC, Jack Applegame wrote: On Saturday, 12 April 2014 at 12:47:02 UTC, MrSmith wrote: Maybe, because compiler cannot distinguish between default constructor which takes int and your one, which also takes int. I don't think so. Without alias this, but with constructor it compiles. The issue is that a call that was essentially supposed to do nothing, accidentally calls your constructor. Because the call "that does nothing" was not expected to do much, it was marked as @safe, pure and nothrow. This conflicts with your constructor, which is none of those. That's the explanation for the behavior you are seeing, it shouldn't be happening. You can work around it by marking your constructor as @safe, pure and nothrow. https://github.com/D-Programming-Language/phobos/pull/1529#discussion_r7175502
Re: alias this and constructor
On Saturday, 12 April 2014 at 12:47:02 UTC, MrSmith wrote: Maybe, because compiler cannot distinguish between default constructor which takes int and your one, which also takes int. I don't think so. Without alias this, but with constructor it compiles.
Re: alias this and constructor
On Saturday, 12 April 2014 at 11:48:36 UTC, Jack Applegame wrote: Why this code doesn't want to compile? Aye! That's my bad! It's a bug I introduced, while fixing more bugs :/ The good news is it only affect 2.065.0. It's already fixed in HEAD. 1000 apologies.
Re: alias this and constructor
On Saturday, 12 April 2014 at 11:48:36 UTC, Jack Applegame wrote:
Why this code doesn't want to compile?
import std.algorithm;
import std.array;
struct Foo {
int a;
this(int v) {}
alias a this;
}
void main() {
immutable(Foo)[] foo;
auto arr = array(foo.filter!(o => true));
}
http://dpaste.dzfl.pl/25572b0f6d0b
Maybe, because compiler cannot distinguish between default
constructor which takes int and your one, which also takes int.
Re: Last element of a forward range
On Saturday, 12 April 2014 at 10:27:00 UTC, bearophile wrote:
Steven Schveighoffer:
Interesting problem.
And it's not an academic one, I have already faced it two or
three times :-)
Given that it is a forward range, a zip between it and a saved
copy that is advanced by one may work.
This compiles and gives the expected output, but how do you
extract the last item with this?
seq.zip(seq.dropOne).writeln;
For starters, you can't in the sense that writeln takes by value,
and your zip has value semantics. So in fact, it is still
unchanged. BTW: That UFCS "seq.zip(someOtherRange)": Ew.
So you'd need to start with a:
auto z = zip(seq, seq.dropOne);
while (!z.empty)
z.popFront();
That's step one. Step two would be actually extracting the
remainaing range. In theory, you can't. In practice, you can hack
around the implementation by:
zip.ranges[1].front;
Turns out "ranges" is not private. I wouldn't use this.
The second, is modifying the stopping policy on the fly:
zip.stoppingPolicy = StoppingPolicy.longest;
auto a = zip.front[1];
I think both are *very* hackish.
I'd do this:
auto bck = seq.save; seq.popFront();
while (!seq.empty)
{
bck.popFront();
seq.popFront();
}
auto last = bck.front;
Re: alias this and constructor
If you just comment out constructor, then it compiles.
alias this and constructor
Why this code doesn't want to compile?
import std.algorithm;
import std.array;
struct Foo {
int a;
this(int v) {}
alias a this;
}
void main() {
immutable(Foo)[] foo;
auto arr = array(foo.filter!(o => true));
}
http://dpaste.dzfl.pl/25572b0f6d0b
Re: Chris E. Miller ToolTip (D1)
will the library build for x64? On Friday, 11 April 2014 at 12:33:27 UTC, FrankLike wrote: On Thursday, 14 November 2013 at 16:17:53 UTC, Heinz wrote: You have to manually set the tooltip's max width to a fixed value using the tooltip handle and Win32 API, by doing this you're telling the tooltip object it is a multiline tooltip and from now on it will accept \r\n as end of line: ttip = new ToolTip; SendMessageA(ttip.handle, TTM_SETMAXTIPWIDTH, 0, 250); ttip.setToolTip(find, "a=n (not approved)\r\no=n (not outlooked)\r\nt>0 (total > 0)); That's it, it works (i tested it). By the way, the DFL version on Chris' site is for D1 but there's this version for D2 (i use both): https://github.com/Rayerd/dfl/tree/master/win32/dfl Now, dfl can use the dmd 2.065 you can see the fork : https://github.com/FrankLIKE/dfl fork(from https://github.com/Rayerd/dfl/) Waiting Miller commit it.
Re: Last element of a forward range
Steven Schveighoffer: Interesting problem. And it's not an academic one, I have already faced it two or three times :-) Given that it is a forward range, a zip between it and a saved copy that is advanced by one may work. This compiles and gives the expected output, but how do you extract the last item with this? seq.zip(seq.dropOne).writeln; I have written I've added an ER to Phobos with some preliminary code: https://issues.dlang.org/show_bug.cgi?id=12564 Bye, bearophile
Re: Associative arrays, multithreading, Rebindable
On Saturday, 12 April 2014 at 09:18:12 UTC, bearophile wrote: Jack Applegame: Should Rebindable to support associative arrays? I guess so. But how to implement it? Bye, bearophile May be just like Rebindable for classes?
Re: Associative arrays, multithreading, Rebindable
Jack Applegame: Should Rebindable to support associative arrays? I guess so. But how to implement it? Bye, bearophile
Associative arrays, multithreading, Rebindable
If I want to pass a dynamic array from one thread to another, then I use immutable(T)[] and it is safe. But what if I need to pass an associative array? I suppose that the use of immutable(T)[U] is unsafe. Because one thread can delete the entry and this will have an impact also on the instance of the array in another thread. So I have to use immutable(T[U]). But the immutability of the entire array makes the array reference immutable too: immutable int[string] foo; foo = assumeUnique(other); // error For immutable classes we can use std.typecons.Rebindable: Rebindable!(immutable Foo) foo; foo = new immutable Foo; // fine But for immutable associative arrays we can't: Rebindable!(immutable int[string]) foo; // error foo = assumeUnique(other); Error: template instance std.typecons.Rebindable!(immutable(int[string])) does not match template declaration Rebindable(T) if (is(T == class) || is(T == interface) || isArray!T) Should Rebindable to support associative arrays?
