[EMAIL PROTECTED] writes:
> > Your proposal seems clever enough. I've added this, but you have to
> > set barCheckSynchronize to #f for it to work.
>
> Thanks a lot! This new barcheck is just great. It's now harder than I expected
> to fool it into doing O(n) warnings (and I'm not sure it uses exactly my
> suggestion, but anyways it seems to work even better than it would :) ).
>
> I found some barcheck related issues:
> (i). A further improvement (and simplification?) of the barcheck mechanism:
> Whenever a barcheck fails, \skip the rest of the bar in the context where the
Adding a \skip on the fly is definitely not a simplification.
> (iii). While speaking \addlyrics and barchecks: The following example gives a
> warning for thte "foo | baz" barcheck:
>
> \addlyrics \notes {a1 | r4 a4 } \context Lyrics \lyrics { foo | baz }
>
> Thus, the meaning of the barcheck in { foo | baz } in \addlyrics is that baz
> is attached to a note in the very beginning of the measure.
> It would feel more logical if \addlyrics would handle | so that a barcheck
> between two syllables "foo" and "baz" would mean that the note foo is
> attached to, ends before the beginning of the measure that "baz" starts in.
> I.e., whenever \addlyrics inserts a \skip between foo and bar, it would
> attempt to place | in the nicest possible position inside the \skip.
>
> This would make the following work fine:
> \addlyrics \notes {a2 r2 | r2 a2 } \context Lyrics \lyrics { foo | baz }
You have many ideas, but the point of barcheck is to indicate
errors. I think that it is pointless to be pedantically correct if
errors have already been detected. For the last case, I can imagine
your reasoning, but \addlyrics (and its more modern friend
\newaddlyrics) are enough of a bitch to get working without all the
fancy things.
I propose to look at this again once more pressing issues have been resolved.
--
Han-Wen Nienhuys | [EMAIL PROTECTED] | http://www.xs4all.nl/~hanwen
_______________________________________________
Bug-lilypond mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/bug-lilypond
