Re: race condition in writing lilypond-book filenames

[John Mandereau]

Le vendredi 18 décembre 2009 à 00:38 -0200, Han-Wen Nienhuys a écrit :
> We already have a plausible explanation, and a fairly simple solution:
> use flock() in ly:parse-file on the .ly file.

I think it's better to sanitize lilypond-book behaviour instead, namely
fixing relevant_contents and make sure there are no duplicate in snippet
filename lists passed to lilypond.  I'll take care of both things later.


> Overwriting files is not actually a problem, since the .ly files will
> be overwritten by the same content.

Not exactly, as I shown in my other reply in this thread.  More
precisely, we only want to overwrite in case the different fragment
options don't change lilypond processing (these options are used by
lilypond-book and are not read from the generated ly code but directly
from the snippet Python object); maybe we should just strip those
options from the generated ly BTW...

Best,
John

signature.asc
Description: Ceci est une partie de message numériquement signée