Phoenix database adapter for Python

phoenixdb is a Python library for accessing the Phoenix SQL database using the remote query server. The library implements the standard DB API 2.0 interface, which should be familiar to most Python programmers.

Installation

The easiest way to install the library is using pip:

pip install phoenixdb

You can also download the source code from here, extract the archive and then install it manually:

cd /path/to/apache-phoenix-x.y.z/phoenix
python setup.py install

Usage

The library implements the standard DB API 2.0 interface, so it can be used the same way you would use any other SQL database from Python, for example:

import phoenixdb
import phoenixdb.cursor

database_url = 'http://localhost:8765/'
conn = phoenixdb.connect(database_url, autocommit=True)

cursor = conn.cursor()
cursor.execute("CREATE TABLE users (id INTEGER PRIMARY KEY, username VARCHAR)")
cursor.execute("UPSERT INTO users VALUES (?, ?)", (1, 'admin'))
cursor.execute("SELECT * FROM users")
print(cursor.fetchall())

cursor = conn.cursor(cursor_factory=phoenixdb.cursor.DictCursor)
cursor.execute("SELECT * FROM users WHERE id=1")
print(cursor.fetchone()['USERNAME'])

Setting up a development environment

If you want to quickly try out the included examples, you can set up a local virtualenv with all the necessary requirements:

virtualenv e
source e/bin/activate
pip install -r requirements.txt
python setup.py develop

To create or update the Avatica protobuf classes, change the tag in gen-protobuf.sh and run the script.

If you need a Phoenix query server for experimenting, you can get one running quickly using Docker:

docker-compose up

Or if you need an older version of Phoenix:

PHOENIX_VERSION=4.9 docker-compose up

If you want to use the library without installing the phoenixdb library, you can use the PYTHONPATH environment variable to point to the library directly:

cd $PHOENIX_HOME/python
python setup.py build
cd ~/my_project
PYTHONPATH=$PHOENIX_HOME/build/lib python my_app.py

Interactive SQL shell

There is a Python-based interactive shell include in the examples folder, which can be used to connect to Phoenix and execute queries:

./examples/shell.py http://localhost:8765/
db=> CREATE TABLE test (id INTEGER PRIMARY KEY, name VARCHAR);
no rows affected (1.363 seconds)
db=> UPSERT INTO test (id, name) VALUES (1, 'Lukas');
1 row affected (0.004 seconds)
db=> SELECT * FROM test;
+------+-------+
|   ID | NAME  |
+======+=======+
|    1 | Lukas |
+------+-------+
1 row selected (0.019 seconds)

Running the test suite

The library comes with a test suite for testing Python DB API 2.0 compliance and various Phoenix-specific features. In order to run the test suite, you need a working Phoenix database and set the PHOENIXDB_TEST_DB_URL environment variable:

export PHOENIXDB_TEST_DB_URL='http://localhost:8765/'
nosetests

Similarly, tox can be used to run the test suite against multiple Python versions:

pyenv install 3.5.5
pyenv install 3.6.4
pyenv install 2.7.14
pyenv global 2.7.14 3.5.5 3.6.4
PHOENIXDB_TEST_DB_URL='http://localhost:8765' tox

Known issues

• You can only use the library in autocommit mode. The native Java Phoenix library also implements batched upserts, which can be committed at once, but this is not exposed over the remote server. This was previously unimplemented due to CALCITE-767, however this functionality exists in the server, but is lacking in the driver. (CALCITE-767)
• TIME and DATE columns in Phoenix are stored as full timestamps with a millisecond accuracy, but the remote protocol only exposes the time (hour/minute/second) or date (year/month/day) parts of the columns. (CALCITE-797, CALCITE-798)
• TIMESTAMP columns in Phoenix are stored with a nanosecond accuracy, but the remote protocol truncates them to milliseconds. (CALCITE-796)
• ARRAY columns are not supported. Again, this previously lacked server-side support which has since been built. The driver needs to be updated to support this functionality. (CALCITE-1050, PHOENIX-2585)