Python package similar to checkpoint in R

Question

Is there a python package that can offer similar or atleast some of the capabilities of the checkpoint library in R:

namely, from this blog:

A new version of the checkpoint package for R has just been released on CRAN. With the checkpoint package, you can easily:

• Write R scripts or projects using CRAN package versions from a specific point in time;
• Share R scripts with others that will automatically install the appropriate package versions (no need to manually install CRAN packages);
• Write R scripts that use older versions of packages, or packages that are no longer available on CRAN;
• Install packages (or package versions) visible only to a specific project, without affecting other R projects or R users on the same system;
• Manage multiple projects that use different package versions;
• Write and share code R whose results can be reproduced, even if new (and possibly incompatible) package versions are released later.

Answer 1

Many of the capabilities that you mention above have been a part of the python setuptools & virtualenv libraries for a while now and the ability to create python wheel format, PEP 427, in setuptools setuptools >= 0.8.0, just about completes the list.

With setuptools you have just about always been able specify dependencies on minimum and/or specific versions of packages in pypi that your code depends on, using virtualenv allows you to install packages without affecting the system libraries and distributing you code in python wheel format lets you bundle the libraries that your code uses, at the version that you tested your code, and the installation of your code will not have any external dependencies or impacts on the end users system.

Note that wheels can be pure python in which case they will work anywhere that the required python is present or can include libraries built using other languages such as C or Fortran - in which case they have to be built for each target system.

Just to top off the list the wheel format can be installed by distribute, pip or by the wheel itself as a wheel includes its own installer.