Start Elasticsearch using Python.
elasticsearch_test.ElasticsearchTest( port=1234 es_path='/home/maria/elasticsearch-5.2.4' data_dir='/home/maria/es-data' es_java_opts='-Xms512m -Xmx512m' )
port
Default: 9200
es_path
Path to the Elasticsearch directory (home of /bin
). Default: uses $ES_HOME
es_java_opts
Sets $ES_JAVA_OPTS
for process to this value if present.
data_dir
Directory to place /data
and /logs
in. Default: uses tempfile
.
import time import elasticsearch_test instance = elasticsearch_test.ElasticsearchTest() # This blocks until either an error is found # or Elasticsearch has been initialized. instance.start() # or instance.start(block=False) while True: if instance.is_started(): break time.sleep(1)
Use it as a resource::
import elasticsearch_test import elasticsearch with elasticsearch_test.ElasticsearchTest() as instance: connection_info = instance.get_connection_info() client = elasticsearch.Elasticsearch(connection_info)
Use it as a Pytest fixture::
import elasticsearch_test @pytest.fixture(scope="session") def es(): instance = elasticsearch_test.ElasticsearchTest() instance.start() yield instance instance.stop()
When using data_dir
, if there are preexisting data in the directory,
you should also wait for the data to be initialized::
with elasticsearch_test.ElasticsearchTest(data_dir='my_data') as instance: while not instance.is_data_initialized(): time.sleep(1) # do something!
Needed: pyenv
and the plugin pyenv-virtualenv
.
Create a normal virtualenv and activate it
Install extras
pip install .[testing]
Install the needed Python versions using
pyenv
:3.4
,3.5
,3.6
,3.7
Setup virtualenvs for all but the Python version you are using, for example if you use
3.4
$ pyenv virtualenv -p python3.5 3.5.6 py35 $ pyenv virtualenv -p python3.6 3.6.6 py36 $ pyenv virtualenv -p python3.7 3.7.1 py37
Activate everything and run tox:
$ pyenv shell py35 py36 py37 $ source ./venv/bin/activate $ tox