[MMTK] MMTK 2.0 prerelease
Konrad Hinsen
hinsen@cnrs-orleans.fr
Mon, 22 Mar 1999 11:01:39 +0100
> There a few import errors finding Scientific.Whatever modules because
> you didn't distribute blank __init__.py files in all the Scientific folders.
Strange. Here's what my tar says about it:
tar tzf /users/ftp/pub/ScientificPython-2.0a1.tar.gz | grep __init__
ScientificPython-2.0/Scientific/__init__.py
ScientificPython-2.0/Scientific/Geometry/__init__.py
ScientificPython-2.0/Scientific/Visualization/__init__.py
ScientificPython-2.0/Scientific/Functions/__init__.py
ScientificPython-2.0/Scientific/Statistics/__init__.py
ScientificPython-2.0/Scientific/Physics/__init__.py
ScientificPython-2.0/Scientific/IO/__init__.py
ScientificPython-2.0/Scientific/TkWidgets/__init__.py
And when I unpack the file, I get all the __init__.py, which are
empty except for those in Statistics and Geometry.
Could it be that your tar program removes empty files? I could easily
add a comment to the empty files if length zero is a problem.
> Also, there is a conflict between the random.py file in the Python Lib
> directory and the Random.py in MMTK on windows.
One is "random", the other "MMTK.Random", so there should not be a
conflict. At least not from outside MMTK... I probably see where your
problem is; MMTK.Random imports random, and without case distinction
it imports Random from the package, i.e. itself. But I don't know what
to do about it; I'll have to scan the Python manuals once more to see
if there is a way to switch off the "search in package first" rule.
Konrad.
--
-------------------------------------------------------------------------------
Konrad Hinsen | E-Mail: hinsen@cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron | Fax: +33-2.38.63.15.17
45071 Orleans Cedex 2 | Deutsch/Esperanto/English/
France | Nederlands/Francais
-------------------------------------------------------------------------------