When the stack is deep and you get an exception, you get
something like this in the middle of the stack trace:
... 20 levels ...
Is there any way to get the entire trace?
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Eric Mahurin schrieb:
When the stack is deep and you get an exception, you get
something like this in the middle of the stack trace:... 20 levels ...
Is there any way to get the entire trace?
Catch the exception and print its backtrace
begin
call_deeply_nested_method
rescue Exception => ex
puts ex, ex.backtrace
end
Regards,
Pit
I already figured out that one. Any way to override the
default behavior? Unfortunately, a simple rescue doesn't catch
interrupt (and other signals). I guess I could set up a signal
handler to catch it. It would be a whole lot easier if I could
just redefine the default print stacktrace method.
--- Pit Capitain <pit@capitain.de> wrote:
Eric Mahurin schrieb:
> When the stack is deep and you get an exception, you get
> something like this in the middle of the stack trace:
>
> ... 20 levels ...
>
> Is there any way to get the entire trace?Catch the exception and print its backtrace
begin
call_deeply_nested_method
rescue Exception => ex
puts ex, ex.backtrace
end
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
Pit Capitain wrote:
Catch the exception and print its backtrace
begin
call_deeply_nested_method
rescue Exception => ex
puts ex, ex.backtrace
end
I am continually reminded of how little I know. 
BTW, Exception is not necessary:
begin
call_deeply_nested_method
rescue => ex #defaults to StandardError
puts ex, ex.backtrace
end
Devin
If you encounter this issue inside of a Unit Test for a controller
(functional test), it doesn't work to put the above code into any of
your test methods.
If you're using Rails-generated code, look for the line at the top of
your controller_test class that looks like this:
# Re-raise errors caught by the controller.
class FooController; def rescue_action(e) raise e end; end
and replace it with one that looks like this:
class FooController; def rescue_action(ex) puts ex, ex.backtrace end;
end
Regards,
Tad
Eric Mahurin schrieb:
I already figured out that one. Any way to override the
default behavior? Unfortunately, a simple rescue doesn't catch
interrupt (and other signals). I guess I could set up a signal
handler to catch it. It would be a whole lot easier if I could
just redefine the default print stacktrace method.
Works for me on Windows 2000. From looking at the source code, I don't think you can change the default behaviour, sorry.
Regards,
Pit