It looks as if lilypond, which is being called by lilypond-book is
running out of memory because of the large amount of data, which will
happen at a different place on different systems. Any non-trivial
program will do this given a large enough amount of data so the real bug
is that the situation is not being handled gracefully. Re-running
lilypond-book cures the problem because lilypond-book is intelligent
enough to know which files have already been processed and so sends less
data to lilypond the second time. So the "cure" would seem to be to get
lilypond-book to catch the error and re-call lilypond if not all lily
files have successfully been processed. (It would have to stop with an
error message if _none_ had been processed as otherwise we get an
infinite loop.)
Until it is fixed maybe something could be put into the docs.
/Bernard
On Sun, 2005-05-15 at 00:52 +0200, Erik Sandberg wrote:
> The error occurs for me too, but on a different file (Radoejtesja.ly).
>
> There is a very easy workaround: Just re-run lilypond-book. It will continue
> where it broke, and process the rest without errors. (So Han-Wen: I wouldn't
> say that this is a release-critical bug, since the workaround is easy)
>
> The bug looks rather irregular, in your case it crashed during Interpreting
> music, when I tried it, the crash occured a few pieces earlier, right after
> 'layout output to...'.
>
> Erik
>
_______________________________________________
bug-lilypond mailing list
bug-lilypond@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-lilypond
