[ddbg] "Access violation" and "Invalid val type" with dictionaries

Ary Manzana ary at esperanto.org.ar
Thu Jun 28 12:45:30 PDT 2007


Thanks!

Just a curiosity: why "= dictionary.length" gives me the length in 
hexadecimal?

Another one: if the length of an array or a dictionary is big enough (> 
100) then I'd like to group by 100 the contents (like in JDT). So I'll 
request the length of the array, and work on that. However, currently 
ddbg only allows me to show around 30 entries, the last one being shown 
as "$". That means I'll have to group by 30 or less. Can this number be 
configured somehow?

Jascha Wetzel escribió:
> it's missing support for class types in associative arrays, and a 
> missing warning message about that ;)
> the next release will fix it.
> 
> Ary Manzana wrote:
>> Hi Jascha,
>>
>> Take the following code:
>>
>> ---
>> module main;
>>
>> import std.stdio;
>>
>> class Point {
>>         int x;
>>     int y;
>>         this(int x, int y) {
>>         this.x = x;
>>         this.y = y;
>>     }
>>         public override char[] toString() {
>>         return "(" ~ std.string.toString(x) ~ ", " ~ 
>> std.string.toString(y) ~ ")";
>>     }
>>     }
>>
>> void main() {
>>     Point[int] dictionary;
>>     dictionary[1] = new Point(1, 10);
>>     dictionary[2] = new Point(2, 20);
>>     dictionary[3] = new Point(3, 30);
>>         writefln("%s", dictionary[1]);
ha>> }"
>> ---
>>
>> If you run it, you'll see (1, 10) as the output. However, if you set a 
>> breakpoint in the writefln line and you try to access dictionary[1]:
>>
>> ->= dictionary[1]
>> Access Violation
>> ->= dictionary[cast(int)1]
>> Access Violation
>>
>> Further:
>>
>> ->= dictionary
>> invalid val type PC4main5Point
>>
>> Is this a bug in ddbg?
>>
>> Regards,
>> Ary


More information about the Digitalmars-d-debugger mailing list