UnicodeError messages lack information about the problem input string

[Martin Packman]

Currently bug reports about unicode problems require psychic debugging skills, because the Python error message is formatted as:

    UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 20: ordinal not in range(128)

The exception instance does have an 'object' attribute that stores the initial string, but only a single character is mentioned in the standard formatting. For issues such as bug 876511, having the content of the line that failed would greatly help working out which of several possible inputs was to blame.

Some care would need to be taken to avoid knock on problems with huge strings or badly constructed exceptions, but users may be more able to resolve certain issues such as bad locale settings if the basic output had more context.