Graham Dumpleton
grahamd at dscpl.com.au
Tue Jul 5 07:19:06 EDT 2005
On 05/07/2005, at 4:11 PM, Nicolas Lehuen wrote: > > What I suspect in Martin's case is that req.finfo returns None > because of the global scope of his PythonHandler configuration, or > something like that. Therefore, the publisher ends up executing the > last branch which splits the requested file name into directory + > func_path and adds index.py to the directory. I'll try to reproduce > this behaviour and let you know about this. Except that Martin wasn't using your latest version of the publisher, but that in the 3.1.[34] as I understand it, so I don't think you can compare it to what you do now. What you do now may well work fine, I have never sat down to look at it properly. Although it worries me a bit that you use req.finfo as isn't that making an assumption that the file that Apache identifies is actually the Python code file when it may not. Just speculating, but what happens if someone was using: SetHandler mod_python PythonHandler mod_python.publisher | .html and there existed a "xxx.py" and xxx.html". Here req.finfo would refer to the ".html" file and not ".py" file. Will this matter? Anyway, the use of req.finfo seems confusing to me and not sure how one benefits except maybe to try and avoid a stat call. One day I will sit down and look at it properly, but no time at the moment. Graham
|