> On 6 April 2011 23:47, Sean Kelly <sean at invisibleduck.org> wrote:
> > I was hoping that's what the optimizer commit was for.  The for loop over the __inflight list can be eliminated (it's a stack, doesn't need to act like a list) so I'll take care of that in the next few days.  Very glad that the tests are all passing now though.
> 
> It's awesome that we have exception chaining on all platforms now. Great work, Sean.

Does the exception chaining include printing out the whole chain of exceptions when they escape main? In the past, when chaining an exception manually by passing an exception to Exception or Error's constructors (or any type derived from them), only the one on the end of the chain was printed. Ideally, they'd all be printed.

- Jonathan M Davis

They've always all printed for me. I need to fix toString to only print the current and iterate manually in the catch block though.

Sent from my iPhone

On Apr 6, 2011, at 8:38 PM, Jonathan M Davis <jmdavisProg at gmx.com> wrote:

>> On 6 April 2011 23:47, Sean Kelly <sean at invisibleduck.org> wrote:
>>> I was hoping that's what the optimizer commit was for.  The for loop over the __inflight list can be eliminated (it's a stack, doesn't need to act like a list) so I'll take care of that in the next few days.  Very glad that the tests are all passing now though.
>> 
>> It's awesome that we have exception chaining on all platforms now. Great work, Sean.
> 
> Does the exception chaining include printing out the whole chain of exceptions when they escape main? In the past, when chaining an exception manually by passing an exception to Exception or Error's constructors (or any type derived from them), only the one on the end of the chain was printed. Ideally, they'd all be printed.
> 
> - Jonathan M Davis
> _______________________________________________
> D-runtime mailing list
> D-runtime at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/d-runtime

And we need to normalize the output between the various os' so that there's one D format instead of one per platform.

On 4/7/2011 6:48 AM, Sean Kelly wrote:
> They've always all printed for me. I need to fix toString to only print the current and iterate manually in the catch block though.
> 
> Sent from my iPhone
> 
> On Apr 6, 2011, at 8:38 PM, Jonathan M Davis <jmdavisProg at gmx.com> wrote:
> 
>>> On 6 April 2011 23:47, Sean Kelly <sean at invisibleduck.org> wrote:
>>>> I was hoping that's what the optimizer commit was for.  The for loop over the __inflight list can be eliminated (it's a stack, doesn't need to act like a list) so I'll take care of that in the next few days.  Very glad that the tests are all passing now though.
>>>
>>> It's awesome that we have exception chaining on all platforms now. Great work, Sean.
>>
>> Does the exception chaining include printing out the whole chain of exceptions when they escape main? In the past, when chaining an exception manually by passing an exception to Exception or Error's constructors (or any type derived from them), only the one on the end of the chain was printed. Ideally, they'd all be printed.
>>
>> - Jonathan M Davis
>> _______________________________________________
>> D-runtime mailing list
>> D-runtime at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/d-runtime
> _______________________________________________
> D-runtime mailing list
> D-runtime at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/d-runtime

That should be sorted now, except for the stack trace.  Should I try to normalize that as well?

On Apr 7, 2011, at 10:22 AM, Brad Roberts wrote:

> And we need to normalize the output between the various os' so that there's one D format instead of one per platform.
> 
> On 4/7/2011 6:48 AM, Sean Kelly wrote:
>> They've always all printed for me. I need to fix toString to only print the current and iterate manually in the catch block though.
>> 
>> Sent from my iPhone
>> 
>> On Apr 6, 2011, at 8:38 PM, Jonathan M Davis <jmdavisProg at gmx.com> wrote:
>> 
>>>> On 6 April 2011 23:47, Sean Kelly <sean at invisibleduck.org> wrote:
>>>>> I was hoping that's what the optimizer commit was for.  The for loop over the __inflight list can be eliminated (it's a stack, doesn't need to act like a list) so I'll take care of that in the next few days.  Very glad that the tests are all passing now though.
>>>> 
>>>> It's awesome that we have exception chaining on all platforms now. Great work, Sean.
>>> 
>>> Does the exception chaining include printing out the whole chain of exceptions when they escape main? In the past, when chaining an exception manually by passing an exception to Exception or Error's constructors (or any type derived from them), only the one on the end of the chain was printed. Ideally, they'd all be printed.
>>> 
>>> - Jonathan M Davis
>>> _______________________________________________
>>> D-runtime mailing list
>>> D-runtime at puremagic.com
>>> http://lists.puremagic.com/mailman/listinfo/d-runtime
>> _______________________________________________
>> D-runtime mailing list
>> D-runtime at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/d-runtime
> 
> _______________________________________________
> D-runtime mailing list
> D-runtime at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/d-runtime

I think so, yes.  It's needlessly inconsistent right now.

Also, still on th table, the linux/64 traces aren't working nor are freebsd, at least last time I looked.  It's been on my mental todo list to look at why, but finding time has been rather hard lately.

On Wed, 13 Apr 2011, Sean Kelly wrote:

> Date: Wed, 13 Apr 2011 16:47:48 -0700
> From: Sean Kelly <sean at invisibleduck.org>
> Reply-To: D's runtime library developers list <d-runtime at puremagic.com>
> To: D's runtime library developers list <d-runtime at puremagic.com>
> Subject: Re: [D-runtime] [D-Programming-Language/druntime] 1ea100: this should
>      fix the remaining test suite breaks
> 
> That should be sorted now, except for the stack trace.  Should I try to normalize that as well?
> 
> On Apr 7, 2011, at 10:22 AM, Brad Roberts wrote:
> 
> > And we need to normalize the output between the various os' so that there's one D format instead of one per platform.
> > 
> > On 4/7/2011 6:48 AM, Sean Kelly wrote:
> >> They've always all printed for me. I need to fix toString to only print the current and iterate manually in the catch block though.
> >> 
> >> Sent from my iPhone
> >> 
> >> On Apr 6, 2011, at 8:38 PM, Jonathan M Davis <jmdavisProg at gmx.com> wrote:
> >> 
> >>>> On 6 April 2011 23:47, Sean Kelly <sean at invisibleduck.org> wrote:
> >>>>> I was hoping that's what the optimizer commit was for.  The for loop over the __inflight list can be eliminated (it's a stack, doesn't need to act like a list) so I'll take care of that in the next few days.  Very glad that the tests are all passing now though.
> >>>> 
> >>>> It's awesome that we have exception chaining on all platforms now. Great work, Sean.
> >>> 
> >>> Does the exception chaining include printing out the whole chain of exceptions when they escape main? In the past, when chaining an exception manually by passing an exception to Exception or Error's constructors (or any type derived from them), only the one on the end of the chain was printed. Ideally, they'd all be printed.
> >>> 
> >>> - Jonathan M Davis
> >>> _______________________________________________
> >>> D-runtime mailing list
> >>> D-runtime at puremagic.com
> >>> http://lists.puremagic.com/mailman/listinfo/d-runtime
> >> _______________________________________________
> >> D-runtime mailing list
> >> D-runtime at puremagic.com
> >> http://lists.puremagic.com/mailman/listinfo/d-runtime
> > 
> > _______________________________________________
> > D-runtime mailing list
> > D-runtime at puremagic.com
> > http://lists.puremagic.com/mailman/listinfo/d-runtime
> 
> _______________________________________________
> D-runtime mailing list
> D-runtime at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/d-runtime
> 

I've seen inconsistent results using stacktrace in C for professional stuff, but it was never high enough of a priority that I looked into why.  The app in question was sufficiently buggy that I thought it was just stack corruption and not the tool that was at fault.  I'll try to make time to look into this as well.

On Apr 13, 2011, at 5:08 PM, Brad Roberts wrote:

> I think so, yes.  It's needlessly inconsistent right now.
> 
> Also, still on th table, the linux/64 traces aren't working nor are freebsd, at least last time I looked.  It's been on my mental todo list to look at why, but finding time has been rather hard lately.
> 
> On Wed, 13 Apr 2011, Sean Kelly wrote:
> 
>> Date: Wed, 13 Apr 2011 16:47:48 -0700
>> From: Sean Kelly <sean at invisibleduck.org>
>> Reply-To: D's runtime library developers list <d-runtime at puremagic.com>
>> To: D's runtime library developers list <d-runtime at puremagic.com>
>> Subject: Re: [D-runtime] [D-Programming-Language/druntime] 1ea100: this should
>>     fix the remaining test suite breaks
>> 
>> That should be sorted now, except for the stack trace.  Should I try to normalize that as well?
>> 
>> On Apr 7, 2011, at 10:22 AM, Brad Roberts wrote:
>> 
>>> And we need to normalize the output between the various os' so that there's one D format instead of one per platform.
>>> 
>>> On 4/7/2011 6:48 AM, Sean Kelly wrote:
>>>> They've always all printed for me. I need to fix toString to only print the current and iterate manually in the catch block though.
>>>> 
>>>> Sent from my iPhone
>>>> 
>>>> On Apr 6, 2011, at 8:38 PM, Jonathan M Davis <jmdavisProg at gmx.com> wrote:
>>>> 
>>>>>> On 6 April 2011 23:47, Sean Kelly <sean at invisibleduck.org> wrote:
>>>>>>> I was hoping that's what the optimizer commit was for.  The for loop over the __inflight list can be eliminated (it's a stack, doesn't need to act like a list) so I'll take care of that in the next few days.  Very glad that the tests are all passing now though.
>>>>>> 
>>>>>> It's awesome that we have exception chaining on all platforms now. Great work, Sean.
>>>>> 
>>>>> Does the exception chaining include printing out the whole chain of exceptions when they escape main? In the past, when chaining an exception manually by passing an exception to Exception or Error's constructors (or any type derived from them), only the one on the end of the chain was printed. Ideally, they'd all be printed.
>>>>> 
>>>>> - Jonathan M Davis
>>>>> _______________________________________________
>>>>> D-runtime mailing list
>>>>> D-runtime at puremagic.com
>>>>> http://lists.puremagic.com/mailman/listinfo/d-runtime
>>>> _______________________________________________
>>>> D-runtime mailing list
>>>> D-runtime at puremagic.com
>>>> http://lists.puremagic.com/mailman/listinfo/d-runtime
>>> 
>>> _______________________________________________
>>> D-runtime mailing list
>>> D-runtime at puremagic.com
>>> http://lists.puremagic.com/mailman/listinfo/d-runtime
>> 
>> _______________________________________________
>> D-runtime mailing list
>> D-runtime at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/d-runtime
>> 
> _______________________________________________
> D-runtime mailing list
> D-runtime at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/d-runtime