Assertion failure on hotshot.stats.load()

Yang fer7msb02 at
Fri Oct 27 10:28:28 CEST 2006

I fell back onto the old profile module, but got the following error
when trying to use zope.interface. I am now without any way to profile
my application.

Traceback (most recent call last):
  File "/home/yang/local/bin/", line 611, in ?
    run('execfile(%r)' % (sys.argv[0],), options.outfile, options.sort)
  File "/home/yang/local/bin/", line 72, in run
    prof =
  File "/home/yang/local/bin/", line 448, in run
    return self.runctx(cmd, dict, dict)
  File "/home/yang/local/bin/", line 454, in runctx
    exec cmd in globals, locals
  File "<string>", line 1, in ?
  File "/.automount/",
line 5, in ?
    from icedb import *
  File "/home/yang/local/lib/python2.4/site-packages/icedb/",
line 4, in ?
    import cafnet
  File "/home/yang/local/lib/python2.4/site-packages/cafnet/",
line 269, in ?
    class Cnl( object ):
  File "/opt/zope/lib/python/zope/interface/", line 132, in advise
    return callback(newClass)
  File "/opt/zope/lib/python/zope/interface/", line
485, in _implements_advice
    classImplements(cls, *interfaces)
  File "/opt/zope/lib/python/zope/interface/", line
462, in classImplements
    spec.declared += tuple(_normalizeargs(interfaces))
  File "/opt/zope/lib/python/zope/interface/", line
1373, in _normalizeargs
    _normalizeargs(v, output)
  File "/opt/zope/lib/python/zope/interface/", line
1372, in _normalizeargs
    for v in sequence:
TypeError: Error when calling the metaclass bases
    iteration over non-sequence

On 10/27/06, Yang |python|
<...> wrote:
> Note: I realize hotshot is obsoleted by cProfile, but 2.5 breaks
> several packages I depend on. I'm using Python 2.4.3.
> I'm getting an AssertionError on "assert not self._stack" when calling
> hotshot.stats.load() on my app's hotshot profile. The app consistently
> causes hotshot to generate such a problematic profile, but I have no
> idea what's causing it. Anybody know what's wrong?
> Here's the profile:
> Potentially relevant bugs:
> Thanks in advance for any help.
> --

More information about the Python-list mailing list