Discussion:
Error running tests with Opera 10
Giulio Cesare Solaroli
2009-07-11 16:57:52 UTC
Permalink
Hello,

I was having some issues running Clipperz test suite with Opera 10.

Looking closer at the failed tests, I have find out that Opera 10 is
returning a messed up name and description for Exceptions.

This problem is triggered also running the default MochiKit tests; in
MochiKit.Async test actually.
The test expects an error with message "foo"
(test_MochiKit-Async.html, line 300; trunk version 1525), but the
actual value is "foo\r\nstacktrace: n/a; see
opera:config#UserPrefs|Exceptions Have Stacktrace".

Opera 10 is appending some extra values to the exception message,
screwing up the test; in my case, it alters the behaviour of the whole
application, as some execution paths are based on the value of the
returned error name.

Any idea on how to fix this problem? or just report it to Opera so
that they can fix it for the next release?

Regards,

Giulio Cesare

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "MochiKit" group.
To post to this group, send email to ***@googlegroups.com
To unsubscribe from this group, send email to mochikit+***@googlegroups.com
For more options, visit this group at http://groups.google.com/group/mochikit?hl=en
-~----------~----~----~----~------~----~------~--~---
Per Cederberg
2009-07-11 18:57:30 UTC
Permalink
This is the type of stuff one would rather not add to work-arounds in
the JavaScript code. Don't understand why Opera can't add stacktraces
as separate properties to their exception objects instead? Or is it
just that they've modified the toString() implementation?

Cheers,

/Per

On Sat, Jul 11, 2009 at 18:57, Giulio Cesare
Post by Giulio Cesare Solaroli
Hello,
I was having some issues running Clipperz test suite with Opera 10.
Looking closer at the failed tests, I have find out that Opera 10 is
returning a messed up name and description for Exceptions.
This problem is triggered also running the default MochiKit tests; in
MochiKit.Async test actually.
The test expects an error with message "foo"
(test_MochiKit-Async.html, line 300; trunk version 1525), but the
actual value is "foo\r\nstacktrace: n/a; see
opera:config#UserPrefs|Exceptions Have Stacktrace".
Opera 10 is appending some extra values to the exception message,
screwing up the test; in my case, it alters the behaviour of the whole
application, as some execution paths are based on the value of the
returned error name.
Any idea on how to fix this problem? or just report it to Opera so
that they can fix it for the next release?
Regards,
Giulio Cesare
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "MochiKit" group.
To post to this group, send email to ***@googlegroups.com
To unsubscribe from this group, send email to mochikit+***@googlegroups.com
For more options, visit this group at http://groups.google.com/group/mochikit?hl=en
-~----------~----~----~----~------~----~------~--~---
Giulio Cesare Solaroli
2009-07-11 19:30:07 UTC
Permalink
Hello Per,

I agree with you about trying to avoid work-arounds; but I haven't
found any way to report a 'simple' Javascript issue to Opera itself.

I have investigated if the problem is caused by a different toString()
implementation, but I don't think so, as the extra stack info is
appended to the 'name' and 'message' extension property also.

Giulio Cesare
Post by Per Cederberg
This is the type of stuff one would rather not add to work-arounds in
the JavaScript code. Don't understand why Opera can't add stacktraces
as separate properties to their exception objects instead? Or is it
just that they've modified the toString() implementation?
Cheers,
/Per
On Sat, Jul 11, 2009 at 18:57, Giulio Cesare
Post by Giulio Cesare Solaroli
Hello,
I was having some issues running Clipperz test suite with Opera 10.
Looking closer at the failed tests, I have find out that Opera 10 is
returning a messed up name and description for Exceptions.
This problem is triggered also running the default MochiKit tests; in
MochiKit.Async test actually.
The test expects an error with message "foo"
(test_MochiKit-Async.html, line 300; trunk version 1525), but the
actual value is "foo\r\nstacktrace: n/a; see
opera:config#UserPrefs|Exceptions Have Stacktrace".
Opera 10 is appending some extra values to the exception message,
screwing up the test; in my case, it alters the behaviour of the whole
application, as some execution paths are based on the value of the
returned error name.
Any idea on how to fix this problem? or just report it to Opera so
that they can fix it for the next release?
Regards,
Giulio Cesare
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "MochiKit" group.
To post to this group, send email to ***@googlegroups.com
To unsubscribe from this group, send email to mochikit+***@googlegroups.com
For more options, visit this group at http://groups.google.com/group/mochikit?hl=en
-~----------~----~----~----~------~----~------~--~---

Continue reading on narkive:
Loading...