Re: catch replacing useful errorInfo

In article <1188467204.146993.275010@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
Larry W. Virden <lvirden@xxxxxxxxx> wrote:
.
.
.
programmer than a user of a program. Maybe a better way to do it would
be to log, into some file, the stack, as well as more detailed
information about the location , state, etc. into an error file. Then
the user could just be told to mail the contents of log abc to
bugs@xxxxxxxxxxxxxxxxxx or whatever, in the case of an unanticipated
type error. In the case of an anticipated problem (wrong file name
.
.
.
You mean your programs don't already "call home" when they
encounter an exception?

More soberly, yes, Larry: I've been threatening to write
a book on the subject of error-handling, and one of its
themes would be that, as you describe, developers have a
responsibility to code for many of the anticipated condi-
tions that otherwise manifest as "errors".
.