Hi
We have the following problem when we try to install a Python script as Win32 service:
The Python script is located in a path that contains the substring "_demonstrator". When installing the Python service for this script, we get the error message, the the file "PythonService_d.exe" isn't found. We don't have this file in our Python installation. First investigations led us to the following lines at win32serviceutil.py:
def LocatePythonServiceExe(exeName = None):
# Try and find the specified EXE somewhere. If specifically registered,
# use it. Otherwise look down sys.path, and the global PATH environment.
if exeName is None:
if win32service.__file__.find("_d")>=0:
exeName = "PythonService_d.exe"
else:
exeName = "PythonService.exe"
...
This function is called to find the executable for the Python service to store it at registry when the Python service is installed. The call to find("_d") returns with >0 in our case, because our path contains the substring "_demonstrator". I don't understand what the meaning of this lines is. Why is the decission for correct PythonService exe made on the substring "_d"? This is not very save, I think.
Can anybody explain this?
An easy workaround is to avoid the substring "_d" in the path to the Python script that should be installed as PythonService. But is it realy intended to forbid "_d" substrings in paths for PythonServices when we don't want to run in PythonService_d.exe? How can this be ensured in real world? Maybe there should be another criteria for the decision which PythonService exe is taken.
Regards,
Wolfgang
_______________________________________________ Python-win32 mailing list Python-win32@python.org http://mail.python.org/mailman/listinfo/python-win32
