A Python client for ODLS
The odls.client package mainly provides a script indexer that will be available in some bin/ directory after install (the concrete location will depend on the type of install).
The indexer script is configured by commandline options (try indexer --help) and configuration files. All this is explained in more detail in the package documentation (see below).
This Python implementation of the ODLS client is the successor of the C++ implementation used in the years before.
While odls.client itself is pure Python, some extensions (notably the SQLite support) require C extensions.
This package is in a really early state!!! Some options do simply not work yet and Windows support is yet completely untested.
The full package documentation can be found at:
odls.client is currently tested on Linux only.
- you need Python >= 2.4
- you need pysqlite whih in turn might require the SQLite libraries and header files.
- for full install (including tests etc.) you also need the Python header files and a working C-compiler like gcc.
Installing the Library
Use easy_install to install the library.
Or download the sources and in the root dir of the extracted package do:
$ python setup.py install
You might need superuser permissions to do that.
Installing for Development
After downloading and extracting the sources, in the root dir of the downloaded file tree do:
$ python bootstrap/bootstrap.py
which will configure the package for your system. Then, run:
which will generate all scripts needed for development in the local bin/ directory.
Afterwards you can run the tests by doing:
odls.client comes with some documentation in the docs/ folder. It can be turned into Sphinx based HTML by running:
The docs then can be found in docs/build/html.
Creating Coverage Reports
We try to keep odls.client at a 100%-test-covered level. You can do the coverage report by issuing on the command line:
$ ./bin/coverage-detect $ ./bin/coveragereport
This will create HTML docs of the test coverage of each module. You can find the generated HTML pages in the coveragereport subdir.