www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - What the hell is going on with exceptions in DMD 2.054?

reply Andrej Mitrovic <andrej.mitrovich gmail.com> writes:
This is one of the worst heisenbug-related DMD releases ever.

I have two completely equivalent directories (minus the name):
C:\Documents and Settings\Andrej\Desktop\bug
D:\dev\projects\DNeonHelium\Samples\cpp

They have a collection of subfolders and inside those subfolders are
Readme files and various other files. I'm trying to read those readme
files and copy the contents to a single readme. One of the files has
an invalid UTF sequence which I didn't know about due to the heisenbug
causing the app to exit with no error whatsoever.

Ok, I run this script in both folders (note that the catch handler is
commented out):

http://codepad.org/PDX0Gywj

Folder 1 results: http://codepad.org/AzymUqZy
Folder 2 results: http://codepad.org/rVDTRvRK

Where's the exception in folder 2??

Here's another one, although not related to folders:
http://d.puremagic.com/issues/show_bug.cgi?id=6329

There's also a related one I found with regards to floating-point
exceptions not being thrown based on the order of declarations,
although I don't know if that's a regression or not and I have to
investigate.

DMD 2.053 does not have these issues, so this is definitely some kind
of regression. And this is not related to RDMD, I've tried compiling
via DMD separately as well.

Here's the folder contents (heavily compressed with 7zip LZMA, hope
you can open this):
https://github.com/AndrejMitrovic/temporary/blob/master/exception_bug/cpp.7z
Jul 21 2011
parent reply "Vladimir Panteleev" <vladimir thecybershadow.net> writes:
On Fri, 22 Jul 2011 00:30:07 +0300, Andrej Mitrovic  
<andrej.mitrovich gmail.com> wrote:


 This is one of the worst heisenbug-related DMD releases ever.


This is the same problem we've discussed in your previous thread. Please  
see the links I posted there for more information.

It sounds like standard crash dialogs are disabled on your system (or the  
problem somehow manifests differently for you). But the cause is the same  
- see the issue I filed.

-- 
Best regards,
  Vladimir                            mailto:vladimir thecybershadow.net
Jul 21 2011
parent reply Andrej Mitrovic <andrej.mitrovich gmail.com> writes:
On 7/22/11, Vladimir Panteleev <vladimir thecybershadow.net> wrote:

 It sounds like standard crash dialogs are disabled on your system (or the
 problem somehow manifests differently for you). But the cause is the same
 - see the issue I filed.


Yes I have them disabled, but what does that have to do with anything?
I'm not compiling with subsystem windows, I should get an error to
stdout as I did with 2.053.
Jul 21 2011
parent reply "Vladimir Panteleev" <vladimir thecybershadow.net> writes:
On Fri, 22 Jul 2011 02:36:50 +0300, Andrej Mitrovic  
<andrej.mitrovich gmail.com> wrote:


 On 7/22/11, Vladimir Panteleev <vladimir thecybershadow.net> wrote:

 It sounds like standard crash dialogs are disabled on your system (or  
 the
 problem somehow manifests differently for you). But the cause is the  
 same
 - see the issue I filed.


 Yes I have them disabled, but what does that have to do with anything?
 I'm not compiling with subsystem windows, I should get an error to
 stdout as I did with 2.053.


Yes. I'm pointing out that this is exactly the same problem as with rdmd  
(so filing a new issue and perhaps creating a new thread was unnecessary).  
The subsystem doesn't matter.

-- 
Best regards,
  Vladimir                            mailto:vladimir thecybershadow.net
Jul 21 2011
parent Andrej Mitrovic <andrej.mitrovich gmail.com> writes:
Ah ok. Well good luck with a minimal test case, this looks hard to catch. :/
Jul 21 2011