A Tuesday 15 December 2009 00:19:34 Anthony Foglia escrigué: > I have a rather complicated toolset, that opens multiple hdf5 files. > I'm having a problem under certain conditions. The simplest example we > have so far is that I open up a pre-existing file in mode "a". I don't > write any data to the file, and the file was originally created by > PyTables. I open a dataset, then drop all references, open a second, > then call (roughly) > > fyle.root._f_getChild(node_name) > > which leads to a segmentation fault. > > I've tried doing as many steps as possible in my own code, and it looks > like the crashing code is in > > fyle.root._g_getLChildAttr(nodePath,"CLASS") > > The odd thing is that if I call > > [n._v_name for n in fyle.root] > > before the call to fyle.root._f_getChild, the code runs fine. > > Any ideas on how to debug this any further? I will try to make a > simpler case, but I can't promise I will succeed. I'm running PyTables > 2.1.2 that comes with Ubuntu 9.10.
Can you try PyTables from trunk? I've added more protection to some HDF5 calls. It will not solve your problem, but perhaps we can get more clues (i.e. perhaps you won't get the segfault). Cheers, -- Francesc Alted ------------------------------------------------------------------------------ This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev _______________________________________________ Pytables-users mailing list Pytables-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/pytables-users
