Error report emails

[J. Landman Gay]

Sarah Reichelt wrote:
>> > That line of script does not appear in any of my code, so I assume  
>> there
>> > is an error in the error reporting.

It looks a little like the error reporting script can't identify the 
object that failed.


> I have now tested a stack with an error built using with 2.6.1, 2.7.0,
> 2.8.0 & 2.8.1 and none of them have any trouble with the error
> reporting email. They all show the exact error.
> 
> So either this is Windows only (I'm using Mac and my app only runs on
> Mac & Windows) or it depends on the error.

I'd suspect the error. I've been using the email-myself error reporting 
feature ever since you clued us onto it, it's been invaluable. I haven't 
ever seen any problems, so it's working okay here.

>> > Related to this, it would be really good if there was a way to add some
>> > more data to these reports e.g. the version number of the
>> > software, the platform, the engine number etc. If I bugzilla the 
>> erroneous
>> > error report, I might add an enhancement request for this.
>> > Does anyone have any other suggestions that they would like added at
>> > the same time?

Make it do logging, so if we don't want to send ourselves an email we 
can just read a log somewhere. Even better, allow a global property that 
turns logging on and off. Then I could write a startup handler that 
checks for a keypress or something ("if the shiftkey is down then...") 
that would set the global logging variable to true. Ergo, startup with 
shiftkey down makes a log.

Some of us have written this stuff, but I'd love to see it just built in.

>>
>> With 2.6.1 it exits the stand alone after showing the error.  It'd be 
>> nice
>> if it did not do this, though this may well have been fixed in later
>> versions of Rev. Probably something I should check with the latest beta.
> 
> I don't get this either. The standalone still runs fine after
> displaying the error dialog, whether you email the report or not.
> Again, maybe this is platform-specific. What are you using Scott?

Maybe Scott doesn't mean "quit" but means the script exits? That would 
be normal. Handlers abort if there's a scripting error. If it really 
does abnormally quit, that might be a crash and should be reported.

-- 
Jacqueline Landman Gay         |     jacque at hyperactivesw.com
HyperActive Software           |     http://www.hyperactivesw.com