Although I agree with you in theory, in practice we need to be backwards
compatable as well is absolutelly correct.
ParseException also does not nesseserally mean "syntax error", it indicates
an error encountered during parsing.
If you would like we can create 3 exception classes:
ParseException
TokenMgrException extends ParseException
SyntaxException extends ParseException
but that would require significant changes thoughout the code, which IMHO
should not be in this release.
For this release we should just throw a ParseException.
>
> To make a source code analogy: "return if break;" contains a set of
> valid tokens for the Java language, but the syntax is invalid, making a
> ParseException the right kind of error to raise. Conversely, "swAtch
> (c) { }" contains characters which cannot be recognized by the lexer as
> a legal token, so a TokenMgrException is appropriate.
>
> I would strongly urge against blurring the distinction between these
> two classes of error, as they really are not the same thing.
>
> Eric
>
>
> --
> To unsubscribe, e-mail:
<mailto:[EMAIL PROTECTED]>
> For additional commands, e-mail:
<mailto:[EMAIL PROTECTED]>
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
