dmd crashes with stack overflow, and I finally figured out the problem : Forward references...

It's certainly not the first time I've hit these bugs (or a bunch of other bugs - all known), and it's often quite hell to work around.

I really like D, and have used it a lot recently, but all the bugs are getting on my nerves. My first wish for d is to get a stable compiler. I think this is a lot more important than working on d2

Just needed to get some frustration out before I start refactoring...

"Simen Haugen" <simen@norstat.no> wrote in message news:fosaio$2gk2$1@digitalmars.com...
> dmd crashes with stack overflow, and I finally figured out the problem : Forward references...
>
> It's certainly not the first time I've hit these bugs (or a bunch of other bugs - all known), and it's often quite hell to work around.
>
> I really like D, and have used it a lot recently, but all the bugs are getting on my nerves. My first wish for d is to get a stable compiler. I think this is a lot more important than working on d2
>
> Just needed to get some frustration out before I start refactoring...

Sorry about your bad experiences with compiler bugs.  I do think forward reference bugs are important to fix.  However, I personally am more excited about D 2.0 than ever.  Walter is now working on the very features are necessary for me to port my code to D.  I tried porting my code a while back and ran into a brick wall.  Without more ctor/dtors and copy semantics for structs, I will have to take a huge performance hit, and this is unacceptable.  Thus, the features he is working on now are on the top of my priority list, even higher than the annoying forward references bug.  My case may be atypical though.

"Craig Black" <cblack@ara.com> wrote in message news:fosg79$1vk$1@digitalmars.com...
> "Simen Haugen" <simen@norstat.no> wrote in message news:fosaio$2gk2$1@digitalmars.com...
>> dmd crashes with stack overflow, and I finally figured out the problem : Forward references...
>>
>> It's certainly not the first time I've hit these bugs (or a bunch of other bugs - all known), and it's often quite hell to work around.
>>
>> I really like D, and have used it a lot recently, but all the bugs are getting on my nerves. My first wish for d is to get a stable compiler. I think this is a lot more important than working on d2
>>
>> Just needed to get some frustration out before I start refactoring...
>
> Sorry about your bad experiences with compiler bugs.  I do think forward reference bugs are important to fix.  However, I personally am more excited about D 2.0 than ever.  Walter is now working on the very features are necessary for me to port my code to D.  I tried porting my code a while back and ran into a brick wall.  Without more ctor/dtors and copy semantics for structs, I will have to take a huge performance hit, and this is unacceptable.  Thus, the features he is working on now are on the top of my priority list, even higher than the annoying forward references bug.  My case may be atypical though.

For those of us who are actually using D1 as it is now, though, these kind of (*cough*inexcusable!*cough*) bugs really put a damper on things.

Simen Haugen wrote:
> dmd crashes with stack overflow, and I finally figured out the problem : Forward references...
> 
> It's certainly not the first time I've hit these bugs (or a bunch of other bugs - all known), and it's often quite hell to work around.
> 
> I really like D, and have used it a lot recently, but all the bugs are getting on my nerves. My first wish for d is to get a stable compiler. I think this is a lot more important than working on d2
> 
> Just needed to get some frustration out before I start refactoring... 
> 
> 

Make sure to put in a bugzilla report or comment on an existing bugzilla report. I've ran into a couple of these but I've complained, as Walter has fixed them.

"Simen Haugen" <simen@norstat.no> wrote in message news:fosaio$2gk2$1@digitalmars.com...
> dmd crashes with stack overflow, and I finally figured out the problem : Forward references...

I've encountered a related bug, not sure if it's exactly what you've encountered:

    http://d.puremagic.com/issues/show_bug.cgi?id=1800