Clodoaldo wrote: > 2006/12/3, Graham Dumpleton <[EMAIL PROTECTED]>: >> >> Can you and anyone else who is interested, read through the >> documentation >> I have added and comment on whether it is adequate. Ie., is there >> anything >> that you can think is missing based on your own knowledge of the new >> importer, >> or anything that is particularly clear. >> > > This paragraph is not clear for me: > > "In this scheme for maintaining a pseudo package, individual modules > are still placed into a directory, but there cannot be a __init__.py > file in the directory. If such a file does exist, it will simply be > ignored." > > It says there cannot be a __init__.py file in the directory and > immediately after it says that if it exists it will be ignored. What > will be ignored? The file or the pseudo package? > > Thinking about it the only way that the paragraph makes sense is if > the package is ignored but for me it is confusing at first sight. >
I believe he's saying the following: Normally putting an __init__.py, even an empty one, will cause the directory to act as a package and you can therefore of "import <directory name>" or "from <directory name> import <something>". However using the mod_python path invalidates that package behavior, so if you have a package that relies on code in __init__.py to be executed on import, it won't work any more. Please correct me if I'm wrong, but I think that's the behavior I'm seeing.