[IronPython] Debugging

[Martin Maly]

Jacques, thank you for more details about the issues you are having with IronPython,

The reason you see the 'report an error to MS' dialog is our change in exception handling that I described earlier. We are going to fix it for the next release and display the exceptions coming from running scripts in a way that standard Python does.
Once we fix that, we should get file and line information for the exceptions, and if not, we'll make sure to fix that too.

As for the error below, it is a parser bug that will be easy to fix.

Thanks again and I hope that next release of IronPython will work better for you.

Martin



________________________________

From: users-bounces at lists.ironpython.com [mailto:users-bounces at lists.ironpython.com] On Behalf Of J. de Hooge
Sent: Monday, December 05, 2005 3:41 AM
To: 'Discussion of IronPython'
Subject: Re: [IronPython] Debugging


Martin,

Thanks for your reaction.
My problems with error reports seem to go beyond what you mention.
A brief example of what I mean is below:

lotsOfCodeAndImports_dummy = 'allKindsOfInterestingCodeThatMayStillContainBugsIncludingImportsOfModulesThatMayAlsoStillContainBugs'

needleInAHaystack = 3
print needleInAHaystack []

evenMoreCode_dummy = 'evenMoreInterestingCodeThatMayStillContainBugs'


CPython tells me:

  File "C:\activ_dell\prog\fun\src\funTry.py", line 4
    print needleInAHayStack []
                             ^
SyntaxError: invalid syntax


IronPython tells me:

Unhandled Exception: 'int' object has no attribute '__getitem__'

(To be complete it first asks me wether I want to send MS an error report, I choose "Don't send")
I am especially bothered by the fact that neither a line number nor a file name are mentioned.
The last two weeks I have repeatedly just sprinkeled numerous "print <aUniqueNumber> " statements in my sourcecode, using binary search to pin down the error...
And I can't use CPython for many of my modules since most of them heavily relies on .NET.

By the way, I am aware that error reporting on a product < 1.0 may understandably not yet be perfect.
Still I choose to use it because of its obvious potential.

Kind regards
Jacques de Hooge
info at geatec.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ironpython-users/attachments/20051205/c02cf74c/attachment.html>