Module diagnostic

source code

Table-based diagnostic message generation.

Diagnostics are loaded from text files.  These files are laid out
according to special rules:

  - Lines beginning with a hash mark are ignored.

  - Each diagnostic begins with a line that contains an at sign (@) and
    a tag used to identify the diagnostic.

  - Subsequent text until the start of the next diagnostic is
    the diagnostic template.

  - Diagnostic templates may contain named-substition tokens as
    used by the Python % operator on a string.

  - Diagnostic messages are interpreted as structured text.

For example:

    # This line is a comment

    @ my first diagnostic
    The command you entered, '$(command)s', is bogus.  Please try again.

    @ my second diagnostic
    The value you specified, '$(value)d', is completely bogus.  Don't
    even bother trying again.