Hi, I want to start using Python 2.6 and 3000. I have several questions.
What, in your experiences, is a functionally elegant solution to installing 2.6 and 3 from source without breaking package dependencies on your favorite Linux/Unix flavor? Is compiling Python 2.6 and 3.0 on a *nix development machine and having it work seamlessly as simple as choosing a particular flavor or are there always going to be package juggling/mangling/pinning/managing issues affecting the other programs on the operating system? Is it as simple as choosing a flavor that is likely to have a Python 3 package available? Or: Is there a way to make a copy of shared libraries ( under perhaps /usr/ local/py2.6lib/ and /usr/local/py3lib/ ) so that I can use 2.6 and 3 without causing package problems with other programs within my operating system? If this seems like a good solution, where can I find more information about how to implement separate libraries inside the same OS as appropriate for Python? This might be the better solution than simply choosing a development flavor of *nix because I am going to want to install other Python libraries like numpy and matplotlib from source that might depend on other potentially incompatible shared libraries than either versions of Python or my Linux/Unix distro ( I have a feeling I am going to learn how to use ldconfig ). I have a bit of experience with Debian Etch but I recently garbled my package management database while compiling the latest version of zlib for Python 2.6. ( why was I compiling and installing zlib from source? I was learning another lesson in patience and planning ) -- http://mail.python.org/mailman/listinfo/python-list
