Switch branches/tags
Nothing to show
Find file
Fetching contributors…
Cannot retrieve contributors at this time
106 lines (80 sloc) 4.35 KB

PyBayes Development

This document should serve as a reminder to me and other possible PyBayes hackers about PyBayes coding style and conventions.

General Layout and Principles

PyBayes is developed with special dual-mode technique - it is both perfectly valid pure Python library and optimised cython-built binary python module.

PyBayes modules are laid out with following rules:

  • all modules go directly into pybayes/<module>.py (pure Python file) with cython augmentation file in pybayes/module.pxd
  • in future, bigger independent units can form subpackages
  • pybayes/wrappers/ subpackage is special, it is the only package whose modules have different implementation for cython and for python. It is accomplished by .py (Python) and .pyx, .pxd (Cython) files.

Tests and Stress Tests

All methods of all PyBayes classes should have a unit test. Suppose you have a module pybayes/, then unit tests for all classes in should go into pybayes/tests/ You can also write stress test (something that runs considerably longer than a test and perhaps provides a simple benchmark) that would go into pybayes/tests/

Imports and cimports

No internal module can import pybayes! That would result in an infinite recursion. External PyBayes clients can and should, however, only import pybayes (and in future also import pybayes.subpackage). From inside PyBayes just import relevant pybayes modules, e.g. import pdfs. Notable exception from this rule is cimport, where (presumable due to a cython bug) from a.b cimport c sometimes doesn't work and one has to type from pybayes.a.b cimport c.

Imports in *.py files should adhere to following rules:

  • import first system modules (sys, io..), then external modules (matplotlib..) and then pybayes modules.
  • instead of importing numpy directly use import wrappers._numpy as np. This ensures that fast C alternatives are used in compiled mode.
  • instead of importing numpy.linalg directly use import wrappers._linalg as linalg.
  • use import module [as abbrev] or, for commonly used symbols from module import symbol.
  • from module import * shouldn't be used.

Following rules apply to *.pxd (cython augmentation) files:

  • no imports, just cimports.
  • use same import styles as in associated .py file. (from module cimport vs. cimport module [as abbrev])
  • for numpy use cimport pybayes.wrappers._numpy as np
  • for numpy.linalg use cimport pybayes.wrappers._linalg as linalg

Above rules do not apply to pybayes/tests. These modules are considered external and should behave as a client script.

Releasing PyBayes

Things to do when releasing new version (let it be X.Y) of PyBayes:

Before Tagging

  1. Set fallback version to X.Y in (around line 15)
  2. Set version to X.Y in support/python-pybayes.spec
  3. Ensure ChangeLog.rst mentions all important changes
  4. (Optional) update short description in AND support/python-pybayes.spec
  5. (Optional) update long description README.rst AND support/python-pybayes.spec


  1. Check everything, run tests and stresses for Python 2.7, 3.2 in both pure/Cython mode
  2. git tag -s vX.Y
  3. --format tar --prefix PyBayes-X.Y/ dist/PyBayes-X.Y.tar
  4. gzip dist/PyBayes-X.Y.tar
  5. ./ register

(do not use ./ upload, it does not work as some files are not in MANIFEST etc.)


  1. Upload PyBayes-X.Y.tar.gz to and
  2. Build and upload docs: cd ../pybayes-doc && ./
  3. Upload updated python-pybayes.spec file to
  4. If short description of PyBayes changed, update it manually at following places:
  5. If long description of PyBayes changed, update it manually at following places: