Indeed, when I set complevel parameter in the Filters constructor to 0, the problem goes away, ( toggling the shuffle buffer did not effect the problem ). The error appears with either zlib or lzo compression. I will see about duplicating outside our app.
S --- Francesc Altet <[EMAIL PROTECTED]> wrote: > A Dijous 20 Juliol 2006 17:47, Stefan Kuzminski va escriure: > > I reproduced this stack trace with HDF5 v1.6.5 and also with > > v1.8.0-alpha4 ( both with pytables-1.3.2 ). Here is the most > > mysterious part, I can get the program to run successfully if I > round > > the values I am setting in the HDF5 to <10 places of precision, if > I > > leave them unrounded, or round them to say 12 places, then I get > the > > stack trace. The data columns in the Tables are all created as > > FloatCol() instances. > > That's pretty crazy indeed. Can you try disabling the compression > filter > and/or the shuffle filter? > > Of course, the best would be if you can generate a small script > reproducing > the problem. Whatever the issue is, if it is a PyTables problem, I'd > like to > fix this in the forthcoming PyTables public release (due to appear in > a week > or so). > > Cheers, > > -- > >0,0< Francesc Altet http://www.carabos.com/ > V V Cárabos Coop. V. Enjoy Data > "-" > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Pytables-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/pytables-users
