On Feb 11, 2009, at 4:28 PM, Kelly F. Hickel wrote:
Hi all, I didn't see a pypy-user list, so I'm posting here.
I came across PyPy while trying to find a way to speed up cvs2svn while converting a large CVS repo. I've pulled the latest via svn, got it built and have translated the interpreter, but cvs2svn using pypy-c fails with the output: debug: WARNING: library path not found, using compiled-in sys.path ERROR: cvs2svn uses the anydbm package, which depends on lower level dbm libraries. Your system has dumbdbm, with which cvs2svn is known to have problems. To use cvs2svn, you must install a Python dbm library other than dumbdbm or dbm. See http://python.org/doc/current/lib/module-anydbm.html for more information.
Now, when we've run into this issue with python, someone just rebuilt python enabling built in libgdbm support, but I haven't found any way to do that or include some other dbm within PyPy.
Since I'm not that familiar with python to start with, I'm hoping that I'm just missing something quite basic, and that some kind soul here can point me in the right direction...
Should I be trying to translate cvs2svn instead of trying using pypy- c ? (this seems to involve building a target file or something).
Am I just out of luck?
Try to use psyco to speed up cvs2svn. I'm afraid PyPy right know would not give you any improvements in performance from CPython for this (maybe even psyco will not help). ps: Isn't cvs2svn a one time thing? -- Leonardo Santagada santagada at gmail.com