Rejection Notice

On the first day of sprints at US PyCon 2012 we had a long and fruitful discussion about PEP 382 and PEP 402. We ended up rejecting both but a new PEP will be written to carry on in the spirit of PEP 402. Martin von Löwis wrote up a summary: [2].

Abstract

Namespace packages are a mechanism for splitting a single Python package across multiple directories on disk. In current Python versions, an algorithm to compute the packages __path__ must be formulated. With the enhancement proposed here, the import machinery itself will construct the list of directories that make up the package. An implementation of this PEP is available at [1].

Terminology

Within this PEP, the term package refers to Python packages as defined by Python’s import statement. The term distribution refers to separately installable sets of Python modules as stored in the Python package index, and installed by distutils or setuptools. The term vendor package refers to groups of files installed by an operating system’s packaging mechanism (e.g. Debian or Redhat packages install on Linux systems).

The term portion refers to a set of files in a single directory (possibly stored in a zip file) that contribute to a namespace package.

Namespace packages today

Python currently provides the pkgutil.extend_path to denote a package as a namespace package. The recommended way of using it is to put:

from pkgutil import extend_path
__path__ = extend_path(__path__, __name__)

in the package’s __init__.py. Every distribution needs to provide the same contents in its __init__.py, so that extend_path is invoked independent of which portion of the package gets imported first. As a consequence, the package’s __init__.py cannot practically define any names as it depends on the order of the package fragments on sys.path which portion is imported first. As a special feature, extend_path reads files named <packagename>.pkg which allow to declare additional portions.

setuptools provides a similar function pkg_resources.declare_namespace that is used in the form:

import pkg_resources
pkg_resources.declare_namespace(__name__)

In the portion’s __init__.py, no assignment to __path__ is necessary, as declare_namespace modifies the package __path__ through sys.modules. As a special feature, declare_namespace also supports zip files, and registers the package name internally so that future additions to sys.path by setuptools can properly add additional portions to each package.

setuptools allows declaring namespace packages in a distribution’s setup.py, so that distribution developers don’t need to put the magic __path__ modification into __init__.py themselves.

Rationale

The current imperative approach to namespace packages has lead to multiple slightly-incompatible mechanisms for providing namespace packages. For example, pkgutil supports *.pkg files; setuptools doesn’t. Likewise, setuptools supports inspecting zip files, and supports adding portions to its _namespace_packages variable, whereas pkgutil doesn’t.

In addition, the current approach causes problems for system vendors. Vendor packages typically must not provide overlapping files, and an attempt to install a vendor package that has a file already on disk will fail or cause unpredictable behavior. As vendors might chose to package distributions such that they will end up all in a single directory for the namespace package, all portions would contribute conflicting __init__.py files.

Specification

Rather than using an imperative mechanism for importing packages, a declarative approach is proposed here: A directory whose name ends with .pyp (for Python package) contains a portion of a package.

The import statement is extended so that computes the package’s __path__ attribute for a package named P as consisting of optionally a single directory name P containing a file __init__.py, plus all directories named P.pyp, in the order in which they are found in the parent’s package __path__ (or sys.path). If either of these are found, search for additional portions of the package continues.

A directory may contain both a package in the P/__init__.py and the P.pyp form.

No other change to the importing mechanism is made; searching modules (including __init__.py) will continue to stop at the first module encountered. In summary, the process import a package foo works like this:

1. sys.path is searched for directories foo or foo.pyp, or a file foo.<ext>. If a file is found and no directory, it is treated as a module, and imported.
2. If a directory foo is found, a check is made whether it contains __init__.py. If so, the location of the __init__.py is remembered. Otherwise, the directory is skipped. Once an __init__.py is found, further directories called foo are skipped.
3. For both directories foo and foo.pyp, the directories are added to the package’s __path__.
4. If an __init__ module was found, it is imported, with __path__ being initialized to the path computed all .pyp directories.

Discussion

Original versions of this specification proposed the addition of *.pth files, similar to the way those files are used on sys.path. With a wildcard marker (*), a package could indicate that the entire path is derived by looking at the parent path, searching for properly-named subdirectories.

People then observed that the support for the full .pth syntax is inappropriate, and the .pth files were changed to be mere marker files, indicating that a directories is a package. Peter Tröger suggested that .pth is an unsuitable file extension, as all file extensions related to Python should start with .py. Therefore, the marker file was renamed to be .pyp.

Dinu Gherman then observed that using a marker file is not necessary, and that a directory extension could well serve as a such as a marker. This is what this PEP currently proposes.

Phillip Eby designed PEP 402 as an alternative approach to this PEP, after comparing Python’s package syntax with that found in other languages. PEP 402 proposes not to use a marker file at all. At the discussion at PyCon DE 2011, people remarked that having an explicit declaration of a directory as contributing to a package is a desirable property, rather than an obstacle. In particular, Jython developers noticed that Jython could easily mistake a directory that is a Java package as being a Python package, if there is no need to declare Python packages.

Packages can stop filling out the namespace package’s __init__.py. As a consequence, extend_path and declare_namespace become obsolete.

Namespace packages can start providing non-trivial __init__.py implementations; to do so, it is recommended that a single distribution provides a portion with just the namespace package’s __init__.py (and potentially other modules that belong to the namespace package proper).

The mechanism is mostly compatible with the existing namespace mechanisms. extend_path will be adjusted to this specification; any other mechanism might cause portions to get added twice to __path__.

References

Copyright

This document has been placed in the public domain.