digitalmars.D.learn - lvalue - opIndexAssign

digitalmars.D.learn - lvalue - opIndexAssign - Tango

The Anh Tran (30/31) Mar 13 2009 Hi,

Daniel Keep (15/54) Mar 13 2009 You can't. This is a hole in the language at the moment, hopefully

The Anh Tran (55/63) Mar 14 2009 I tried to register. But there is no confirmation mail sent to me for 2

Denis Koroskin (2/34) Mar 13 2009
downs (5/21) Mar 15 2009 Take a look at tools.behave_as (for phobos).

The Anh Tran (4/8) Mar 15 2009 Many thanks.

bearophile (4/6) Mar 15 2009 They share part of the same purposes. But D string mixing are scoped, wh...

The Anh Tran <trtheanh gmail.com> writes:

Hi,

When porting from c++ to D, i encounter this strange discrimination:
1. Built-in AA:
	int[int] arr;
	arr[123] += 12345;
	arr[321]++;

2. Tango HashMap:
	auto hm = new HashMap!(int, int)();
	hm[123] += 12345; // error not lvalue
	hm[123]++;	// error

D document says current opIndexAssign does not work as lvalue. But why 
can builtin AA can that? How can i copy builtin AA behaviour?

--------------

Forgive my noob, where is the place to ask question, report bug for Tango?

1. I can't compile D code using tango hashmap in debug mode:

import tango.util.container.HashMap;
void main()
{
	auto hm = new HashMap!(uint, uint)();
}

 dmd -w -g -debug  hello.d // error

2. Compile D code using Tango Regex by GDC emit lots of link errors.

3. Bug in Tango atomicIncrement, atomicDecrement:

int task_done = 0;
atomicIncrement(task_done);

That function is compiled in asm:
lock inc byte ptr[task_done];
Which is wrong. It'll wrap to 0 at 255.
It should be: lock inc dword ptr[task_done];

4. There is no atomicAdd(ref original, int newvalue) family. GCC 
equivalence is __syn_fetch_and_add ...

Mar 13 2009

Daniel Keep <daniel.keep.lists gmail.com> writes:

The Anh Tran wrote:
 Hi,
 
 When porting from c++ to D, i encounter this strange discrimination:
 1. Built-in AA:
     int[int] arr;
     arr[123] += 12345;
     arr[321]++;
 
 2. Tango HashMap:
     auto hm = new HashMap!(int, int)();
     hm[123] += 12345; // error not lvalue
     hm[123]++;    // error
 
 D document says current opIndexAssign does not work as lvalue. But why
 can builtin AA can that? How can i copy builtin AA behaviour?

You can't.  This is a hole in the language at the moment, hopefully
solved by the introduction of ref returns (but that's in D 2.0 which you
don't want to use at the moment.)

 Forgive my noob, where is the place to ask question, report bug for Tango?

You could try the Tango IRC channel:

irc://irc.freenode.org/#d.tango

That, or the Tango forums:

http://dsource.org/projects/tango/forums

You can report problems with Tango via the ticket system:

http://dsource.org/projects/tango/report ("New Ticket" is down the
bottom of the page.)

 1. I can't compile D code using tango hashmap in debug mode:
 
 import tango.util.container.HashMap;
 void main()
 {
     auto hm = new HashMap!(uint, uint)();
 }
 
 dmd -w -g -debug  hello.d // error


When posting problems with compiling something, it helps to mention the
version of the compiler you're using, your platform, the version of
Tango (in this case) and what the error actually is.

 2. Compile D code using Tango Regex by GDC emit lots of link errors.
 
 3. Bug in Tango atomicIncrement, atomicDecrement:
 
 int task_done = 0;
 atomicIncrement(task_done);
 
 That function is compiled in asm:
 lock inc byte ptr[task_done];
 Which is wrong. It'll wrap to 0 at 255.
 It should be: lock inc dword ptr[task_done];
 
 4. There is no atomicAdd(ref original, int newvalue) family. GCC
 equivalence is __syn_fetch_and_add ...

  -- Daniel

Mar 13 2009

The Anh Tran <trtheanh gmail.com> writes:

Daniel Keep wrote:
 
 You could try the Tango IRC channel:
 That, or the Tango forums:
 http://dsource.org/projects/tango/forums
 
 You can report problems with Tango via the ticket system:
 http://dsource.org/projects/tango/report ("New Ticket" is down the
 bottom of the page.)

I tried to register. But there is no confirmation mail sent to me for 2 
days. That's why i post here :(

-------------------------

Linux Ubuntu 8.10.
DMD v1.041 from digitalmars.com
GDC 4.2, download from ubuntu synaptic.
Tango 0.99.7. Zip package from 
http://www.dsource.org/projects/tango/wiki/SourceDownloads

-----------------------

import tango.io.Stdout;
import tango.util.container.HashMap;
void main()
{
	auto hm = new HashMap!(uint, uint)();
}

dmd -w -g -debug test.d
Error:
warning -
../import/tango/util/container/Slink.d(352): Error: statement is not 
reachable
../import/tango/util/container/HashMap.d(13): template instance 
tango.util.container.HashMap.HashMap!(uint,uint) error instantiating

------------------------------------

GDC tango regex has been fixed myself. It's just a compile flag.

------------------------------------

import tango.io.Stdout;
import tango.core.Atomic;

void main()
{
	uint x = 255;
	uint y = atomicIncrement(x);

	Atomic!(uint) at;
	at.store(255);
	uint c = at.increment();
}


gdc ./test.d -o ./test-tango-gdc -frelease -g -O3 -msse2 -march=native 
-mfpmath=sse -femit-templates=auto -fversion=Posix -fversion=Tango -lgtango

./hello.d:21: template 
tango.core.Atomic.Atomic!(uint).Atomic.store(msync ms = msync.seq) does 
not match any template declaration
./hello.d:21: template 
tango.core.Atomic.Atomic!(uint).Atomic.store(msync ms = msync.seq) 
cannot deduce template function from argument types (int)


dmd -g -release -O -inline ./test.d -of./test-tango-dmd
Here is disasm of atomicIncrement():
	push    ebp
	mov     ebp, esp
	push    eax
	mov     eax, [ebp+var_4]
	lock inc byte ptr [eax]		<--- wrong, must be dword ptr
	mov     eax, [eax]
	mov     esp, ebp
	pop     ebp
	retn

Mar 14 2009

"Denis Koroskin" <2korden gmail.com> writes:

On Sat, 14 Mar 2009 06:41:52 +0300, The Anh Tran <trtheanh gmail.com> wrote:

 Hi,

 When porting from c++ to D, i encounter this strange discrimination:
 1. Built-in AA:
 	int[int] arr;
 	arr[123] += 12345;
 	arr[321]++;

 2. Tango HashMap:
 	auto hm = new HashMap!(int, int)();
 	hm[123] += 12345; // error not lvalue
 	hm[123]++;	// error

 D document says current opIndexAssign does not work as lvalue. But why  
 can builtin AA can that? How can i copy builtin AA behaviour?

 --------------

 Forgive my noob, where is the place to ask question, report bug for  
 Tango?

 1. I can't compile D code using tango hashmap in debug mode:

 import tango.util.container.HashMap;
 void main()
 {
 	auto hm = new HashMap!(uint, uint)();
 }

  > dmd -w -g -debug  hello.d // error

 2. Compile D code using Tango Regex by GDC emit lots of link errors.

 3. Bug in Tango atomicIncrement, atomicDecrement:

 int task_done = 0;
 atomicIncrement(task_done);

 That function is compiled in asm:
 lock inc byte ptr[task_done];
 Which is wrong. It'll wrap to 0 at 255.
 It should be: lock inc dword ptr[task_done];

This one is fixed long ago, try updating your Tango installation.

 4. There is no atomicAdd(ref original, int newvalue) family. GCC  
 equivalence is __syn_fetch_and_add ...

Mar 13 2009

downs <default_357-line yahoo.de> writes:

The Anh Tran wrote:
 Hi,
 
 When porting from c++ to D, i encounter this strange discrimination:
 1. Built-in AA:
     int[int] arr;
     arr[123] += 12345;
     arr[321]++;
 
 2. Tango HashMap:
     auto hm = new HashMap!(int, int)();
     hm[123] += 12345; // error not lvalue
     hm[123]++;    // error
 
 D document says current opIndexAssign does not work as lvalue. But why
 can builtin AA can that? How can i copy builtin AA behaviour?
 

Take a look at tools.behave_as (for phobos).

It only works properly with built-in types though.

http://dsource.org/projects/scrapple/browser/trunk/tools/tools/behave_as.d

It should let you implement a HashMap that supports += and similar, at least
for built-ins.

Mar 15 2009

The Anh Tran <trtheanh gmail.com> writes:

downs wrote:
 Take a look at tools.behave_as (for phobos).
 It only works properly with built-in types though.
 http://dsource.org/projects/scrapple/browser/trunk/tools/tools/behave_as.d
 It should let you implement a HashMap that supports += and similar, at least
for built-ins.

Many thanks.
No offense, but for a beginner like me, those mixin are similar to C 
macro hacks years ago. :)

Mar 15 2009

bearophile <bearophileHUGS lycos.com> writes:

The Anh Tran:
 No offense, but for a beginner like me, those mixin are similar to C 
 macro hacks years ago. :)

They share part of the same purposes. But D string mixing are scoped, while C
macros cut across scopes. This makes mixins quite safer to use.

Bye,
bearophile

Mar 15 2009

D Programming

C/C++ Programming

Other

digitalmars.D.learn - lvalue - opIndexAssign - Tango