[Biopython-dev] OT:Strange behaviour of traceback

[Andreas Kuntzagk]

Maybe anybody has an explanation for the following:
My Python-Programm is 
-----------------------
from Bio import GenBank

def main():
    p=GenBank.RecordParser()
    n=GenBank.NCBIDictionary(parser=p)
    n["Y07610"]
            
if __name__ == "__main__":
    main()
-----------------------
Test1:
-----------------------
murple/swissprot_new> setenv PYTHONPATH /home/murple/cvs3/biopython/
murple/swissprot_new> python find_problematic_entries2.py
Traceback (most recent call last):
  File "find_problematic_entries2.py", line 10, in ?
    main()
  File "find_problematic_entries2.py", line 7, in main
    n["Y07610"]
  File "/usr/lib/python2.2/site-packages/PIL/__init__.py", line 1563, in
__getitem__
    # ;-)
KeyError: Empty information returned.
-----------------------
Test2:
----------------------
murple/swissprot_new> unsetenv PYTHONPATH
murple/swissprot_new> python find_problematic_entries2.py
Traceback (most recent call last):
  File "find_problematic_entries2.py", line 10, in ?
    main()
  File "find_problematic_entries2.py", line 7, in main
    n["Y07610"]
  File "/usr/lib/python2.2/site-packages/Bio/GenBank/__init__.py", line
1565, in __getitem__
    raise KeyError, "Empty information returned."
KeyError: Empty information returned.
---------------------
where home/murple/cvs3/biopython/ is my directory with a slightly
modified biopython. But _NO_WAY_ im using PIL anywhere! And there is no 
line 1563 in /usr/lib/python2.2/site-packages/PIL/__init__.py (this is a
short file with only comments). So it seems, traceback is broken.
Did anybody see something like that ever before? I skimmed the Python
bug database on sourceforge but where unable to find anything like this.
I would file a bug, but for this I have to strip this down to a
non-biopython bug. At the moment I don't have the time, maybe in
october.

Andreas