Re: Lilypond error behaviour

[David Kastrup]

Thomas Morley <address@hidden> writes:

> 2016-04-19 8:48 GMT+02:00  <address@hidden>:
>> On Tue, 19 Apr 2016, Noeck wrote:
>>> think that is what confuses most people here. The error was non-fatal
>>> but still after everything is done, an additional fatal error is raised
>>> to draw attention to previous non-fatal errors.
>>
>> Those are fatal errors.  They kill the program.  They're just not
>> instantly fatal.
>
>
>
> My thesis was that even a non_fatal_error will be turned into a fatal
> for alerting purpose finally.
> See:
> \version "1.2.3"
> { c''1 }
>
> According to the lexer it's a "non_fatal_error" (and correctly
> reported in the log), but turned into a fatal later.
>
> If I'm wrong here I'd love to be corrected.

I don't think that it's the non_fatal_error per se that is turned into a
fatal error, but rather syntax errors specifically (at least the lexer
has a flag for keeping track of syntax errors).  At least that's what I
seem to remember.

-- 
David Kastrup