Re: Lilypond error behaviour

[David Kastrup]

Sharon Rosner <address@hidden> writes:

>> Except that the premise of this thread was that users refuse to look
>> _at_ _all_ at _any_ messages or error status and instead want to be able
>> to deduce the presence of errors from the existence of output files.
>
>> So it pretty much doesn't matter what we write on the console: it's not
>> acceptable for LilyPond to produce any file in case of a syntax error.
>
> Actually the premise of this thread was something altogether different, and
> I'd rather believe it's a simple case of misunderstanding than you trying
> deliberately to make any sort of healthy debate futile:
>
> Andrew started with:
>
>> Wouldn’t a complete syntax error stop processing?
>
> Then Simon wrote:
>
>> To me, the oddity would be in that Lily speaks of a ‘fatal error’ here.
>
> That's the whole premise for you right there. The existence or absence of
> output files is just a corollary.
>
> If we can further pinpoint the discussion, given the current way lilypond
> handles parsing (or other) errors, what *is* the meaning of 'fatal error'?

An error leading to a fatal error message and a non-zero exit status
because LilyPond does not see fit to deal with the input properly.

-- 
David Kastrup