qPython is a Python library providing support for interprocess communication between Python and kdb+ processes, it offers:
- Synchronous and asynchronous queries
- Convenient asynchronous callbacks mechanism
- Support for kdb+ protocol and types as of kdb+ v4.0
- Uncompression of the IPC data stream
- Internal representation of data via numpy arrays (lists, complex types) and numpy data types (atoms)
- Supported on Python 3.4/3.5/3.6 and numpy 1.8+
For more details please refer to the documentation.
To install qPython from PyPI:
$ pip install qpython3
Please do not use old PyPI package name: qpython or exxeleron-qpython.
qPython documentation is generated with help of Sphinx document generator.
In order to build the documentation, including the API docs, execute:
make html from the doc directory.
Documentation is built into the: doc/build/html/ directory.
qPython utilizes Cython to tune performance critical parts of the code.
Instructions:
- Execute:
python setup.py build_ext --inplace
Instructions:
- Execute:
python setup.py bdist
qPython uses py.test as a test runner for unit tests.
Instructions:
- Make sure that top directory is included in the
PYTHONPATH - Execute:
py.test
Configured GitHub Action as per the "Python" section of:
qPython requires numpy 1.8 to run.
Optional requirements have to be met to provide additional features:
- tune performance of critical parts of the code:
- Cython 0.20.1
- support serialization/deserialization of
pandas.Seriesandpandas.DataFrame- pandas 0.14.0
- run Twisted sample:
- Twisted 13.2.0
- build documentation via Sphinx:
- Sphinx 1.2.3
- mock 1.0.1
Required libraries can be installed using pip.
To install all the required dependencies, execute:
pip install -r requirements.txt
Minimal set of required dependencies can be installed by executing:
pip install -r requirements-minimal.txt