Skip to content

Latest commit

 

History

History
165 lines (118 loc) · 5.01 KB

README.md

File metadata and controls

165 lines (118 loc) · 5.01 KB

Logo

XTDB Python: A Python ORM for XTDB

Python Versions Stable Version

Tests Pre-commit License

Check out the documentation for a more complete documentation of the package.

Installation

You can install this project using pip:

$ pip install xtdb

Usage

The following examples assume you have set the XTDB_URI variable in your environment. To start experimenting, you could use the following setup using Docker:

$ docker run -p 3000:3000 -d juxt/xtdb-standalone-rocksdb:1.21.0
$ export XTDB_URI=http://localhost:3000/_xtdb

Using the Client

The XTDBClient supports the full HTTP API spec.

>>> import os
>>> from xtdb.session import XTDBClient, Operation
>>>
>>> client = XTDBClient(os.environ["XTDB_URI"])
>>> client.submit_tx([Operation.put({"xt/id": "123", "name": "fred"})])
>>>
>>> client.query('{:query {:find [(pull ?e [*])] :where [[ ?e :name "fred" ]]}}')
[[{'name': 'fred', 'xt/id': '123'}]]
>>>
>>> client.get_entity("123")
{'name': 'fred', 'xt/id': '123'}

Take a look at the spec to see the full range of functionality that maps directly to the client.

Using the Datalog module

The datalog module also provides a layer to construct queries with more easily. Given the data from the cities example has been seeded:

>>> from xtdb.datalog import Find, Where
>>>
>>> query = Find("(pull Country [*])") & Find("City") & (Where("City", "City/country", "Country") & Where("City", "City/name", '"Rome"'))
>>> str(query)
{:query {:find [ (pull Country [*]) City] :where [ [ City :City/country Country ] [ City :City/name "Rome" ]]}}
>>>
>>> client.query(query)
[[{'type': 'Country', 'Country/name': 'Italy', 'xt/id': 'c095839f-031f-46ad-85e1-097f634ba4f0'}, '33aa7fa6-b752-4982-a772-d2dbaeda58ae']]

To see more datalog query examples, check out the unit tests.

Using the ORM and Session

Below is an example of how to use the ORM functionality.

import os
from dataclasses import dataclass

from xtdb.orm import Base
from xtdb.query import Query
from xtdb.session import XTDBSession


@dataclass
class TestEntity(Base):
    name: str


@dataclass
class SecondEntity(Base):
    age: int
    test_entity: TestEntity

session = XTDBSession(os.environ["XTDB_URI"])
entity = TestEntity(name="test")

with session:
    session.put(entity)

query = Query(TestEntity).where(TestEntity, name="test")
result = session.query(query)

result[0].dict() #  {"TestEntity/name": "test", "type": "TestEntity", "xt/id": "fe2a3ee0-9254-41dc-91cc-74ad9e2a16db"}

To see more examples, check out the examples directory. Don't hesitate to add your own examples!

Using the CLI for querying

This package also comes with an easy CLI tool to query XTDB. Again, set the XTDB_URI variable in your environment first and you will be able to query XTDB as follows:

$ echo '{:query {:find [(pull ?e [*])] :where [[ ?e :name "fred" ]]}}' | python -m xtdb
[[{"name": "fred", "xt/id": "123"}]]

To use a query from a file, run:

$ cat query.txt
{:query {:find [(pull ?e [*])] :where [[ ?e :name "fred" ]]}}
$
$ python -m xtdb < query.txt
[[{"name": "fred", "xt/id": "123"}]]

Contributing

Installation

To get started, clone the repo and create an environment using Poetry

$ git clone https://github.com/Donnype/xtdb-py.git
$ cd xtdb-py
$ poetry install

Now set up XTDB, for instance using Docker

$ docker run -p 3000:3000 -d juxt/xtdb-standalone-rocksdb:1.21.0

Export the XTDB_URI environment variable to be able to use os.environ["XTDB_URI"] to fetch the endpoint

$ export XTDB_URI=http://localhost:3000/_xtdb

Development

The Makefile has several targets that should make development easier:

$ make utest  # Run unit tests
$ make itest  # Run integration tests
$ make check  # Run all linters
$ make done   # Run all of the above

The CI runs these checks as well. Check out the project page for issues and features to work on.