The pyderivationagent package provides a python wrapper for derivation agents as part of TheWorldAvatar project. It is a python equivalent of uk.ac.cam.cares.jps.base.agent.DerivationAgent.java but based on Flask application behind a gunicorn server, inspired by the Example: Python agent. pyderivationagent uses py4jps>=1.0.32 to access derivation operations provided in uk.ac.cam.cares.jps.base.derivation.DerivationClient.java. For technical details, below are a few useful links:
py4jps- python wrapper for jps-base-libDerivationClient- derivation frameworkDerivationAgent.java- derivation agent java class that uses methods provided inDerivationClientDerivationAsynExample- example on directed acyclic graph (DAG) of derivations operated by derivation agents
For development and testing reasons, follow below instructions to get a copy of the project up and running on your local system.
It is highly recommended to install pyderivationagent packages in a virtual environment. The following steps can be taken to build a virtual environment:
(Windows)
$ python -m venv <venv_name>
$ <venv_name>\Scripts\activate.bat
(<venv_name>) $(Linux)
$ python3 -m venv <venv_name>
$ source <venv_name>/bin/activate
(<venv_name>) $The above commands will create and activate the virtual environment <venv_name> in the current directory.
The following command can be used to install the pyderivationagent package.
(<venv_name>) $ pip install pyderivationagentWhen creating a new derivation python agent, it's strongly advised to stick to the folder structure below:
Recommended python agent folder layout
.
├── ... # other project files (README, LICENSE, etc..)
├── youragent # project source files for your agent
│ ├── agent # module that contains the agent logic
│ │ ├── __init__.py
│ │ └── your_agent.py # your derivation agent
│ ├── conf # module that contains the agent logic
│ │ ├── __init__.py
│ │ └── your_conf.py # specific configuration for your agent
│ ├── data_model # module that contains the dataclasses for concepts
│ │ ├── __init__.py
│ │ └── your_onto.py # dataclasses for your ontology concepts
│ ├── kg_operations # module that handles knowledge graph operations
│ │ ├── __init__.py
│ │ └── your_sparql.py # sparql query and update strings for your agent
│ ├── other_modules # other necessary modules for your agent
│ │ ├── __init__.py
│ │ ├── module1.py
│ │ └── module2.py
│ ├── entry_point.py # agent entry point
│ └── __init__.py
├── youragent.env.example # example environment variables for configuration
├── docker-compose.yml # docker compose file
├── Dockerfile # Dockerfile
└── tests # tests files
Example code
your_agent.py
from pyderivationagent import DerivationAgent
from pyderivationagent import DerivationInputs
from pyderivationagent import DerivationOutputs
from youragent.kg_operations import YourSparqlClient
from youragent.data_model import YOUR_CONCEPT
from youragent.data_model import ANOTHER_CONCEPT
from youragent.data_model import OUTPUT_CONCEPT
from rdflib import Graph
# NOTE For any developer to extend the DerivationAgent class, four @abstractmethod MUST be implemented
class YourAgent(DerivationAgent):
##########################################################
## I. __init__ for custom configuration (if applicable) ##
##########################################################
def __init__(self,
your_str_conf: str,
your_int_conf: int,
your_bool_conf: bool,
**kwargs
):
super().__init__(**kwargs) # pass all other parameters to DerivationAgent.__init__, will throw error if unexpected input received
# Below you may want to assign your custom configuration
# How to provide them when instantiating agent will be detailed in your_conf.py, entry_point.py, and youragent.env.example
self.your_str_conf = your_str_conf
self.your_int_conf = your_int_conf
self.your_bool_conf = your_bool_conf
###############################
## II. Derivation agent part ##
###############################
# Firstly, as the agent is designed to register itself in the knowledge graph when it is initialised
# One need to define the agent inputs/outputs by providing the concept IRIs as return values
# The registration is by default, which can be altered by setting flag REGISTER_AGENT=false in the env file
def agent_input_concepts(self) -> list:
# Assume two input concepts are used by the agent, then developer need to provide it in a list
return [YOUR_CONCEPT, ANOTHER_CONCEPT]
def agent_output_concepts(self) -> list:
# Assume one output concept is produced by the agent, then developer need to provide it in a list
# NOTE even ONE concept should be provided as a list
return [OUTPUT_CONCEPT]
def validate_inputs(self, http_request) -> bool:
# You may want to add some specific validation after the generic checks
if super().validate_inputs(http_request):
# do some specific checking
pass
def process_request_parameters(self, derivation_inputs: DerivationInputs, derivation_outputs: DerivationOutputs):
# Provide your agent logic that converts the agent inputs to triples of new created instances
# The derivation_inputs will be in the format of key-value pairs with the concept as key and instance iri as value
# For example:
# {
# "https://www.example.com/triplestore/repository/Ontology.owl#Concept_1": ["https://www.example.com/triplestore/repository/Concept_1/Instance_1"],
# "https://www.example.com/triplestore/repository/Ontology.owl#Concept_2": ["https://www.example.com/triplestore/repository/Concept_2/Instance_2"],
# "https://www.example.com/triplestore/repository/Ontology.owl#Concept_3":
# ["https://www.example.com/triplestore/repository/Concept_3/Instance_3_1", "https://www.example.com/triplestore/repository/Concept_3/Instance_3_2"],
# "https://www.example.com/triplestore/repository/Ontology.owl#Concept_4": ["https://www.example.com/triplestore/repository/Concept_4/Instance_4"]
# }
# The instance IRIs of the interested class (rdf:type) can be accessed via derivation_inputs.getIris(clz)
# e.g. assume we will take the first iri that has rdf:type YOUR_CONCEPT
instance_iri = derivation_inputs.getIris(YOUR_CONCEPT)[0]
# You may want to create instance of YourSparqlClient for specific queries/updates you would like to perform
# YourSparqlClient should be defined in your_sparql.py that will be introduced later in this README.md file
# This client can be initialised with the configuration you already initialised in YourAgent.__init__ method
# A convenient method get_sparql_client is provided to get sparql_client if provided YourSparqlClient as arg
sparql_client = self.get_sparql_client(YourSparqlClient)
# Please note here we are using instance_iri of class YOUR_CONCEPT within the provided derivation_inputs
response = sparql_client.your_sparql_query(instance_iri)
# You may want to log something during agent execution
self.logger.info("YourAgent has done something.")
self.logger.debug("And here are some details...")
# The new created instances should be added to the derivation_outputs in the form of triples
# e.g. <http://example/ExampleClass_UUID> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://example/ExampleClass>.
# <http://example/ExampleClass_UUID> <http://example/hasValue> 5.
derivation_outputs.createNewEntity("http://example/ExampleClass_UUID", "http://example/ExampleClass")
derivation_outputs.addLiteral("http://example/ExampleClass_UUID", "http://example/hasValue", 5)
# Alternatively, you may create an instance of rdflib.Graph and add the whole graph to derivation_outputs
# In which case, the above triples can be added by:
g = Graph()
g.add((URIRef("http://example/ExampleClass_UUID"), RDF.type, URIRef("http://example/ExampleClass")))
g.add((URIRef("http://example/ExampleClass_UUID"), URIRef("http://example/hasValue"), Literal(5)))
derivation_outputs.addGraph(g)
################################################################
## III. Any other periodical job the necessary for your agent ##
################################################################
# Additionally, if you would like to define your own periodical job to be executed, you may do it like blow:
def your_periodical_job(self):
# Here provide the job logic to be executed periodically
pass
@DerivationAgent.periodical_job # This decorator enables the _start_your_periodical_job() to be called when calling start_all_periodical_job()
def _start_your_periodical_job(self):
# You also need to provide a function so that your periodical job can be started on its own
# self.scheduler is an object of APScheduler class
self.scheduler.add_job(
id='your_periodical_job', # the name for the periodical job
func=self.your_periodical_job, # the function for the periodical job
trigger='interval', # trigger type
seconds=timescale_for_your_periodical_job # the time interval you prefer for the job execution
)
# The start of your periodical job can be done in two ways once instantiated YourAgent (assume an object named "your_agent"):
# Option 1:
# your_agent._start_your_periodical_job() # start your periodical job independently
# Option 2:
# your_agent.start_all_periodical_job() # start your periodical job together with all other periodical jobs (e.g. _start_monitoring_derivations)
# An example is also provided in entry_point.py later in this README.md fileyour_conf.py
from pyderivationagent.conf import AgentConfig
from pyderivationagent.conf import config_generic
# Similar to AgentConfig, here you may provide the configurations specific to your agent
class YourConfig(AgentConfig):
YOUR_STR_CONF: str
YOUR_INT_CONF: int
YOUR_BOOL_CONF: bool
def config_your_agent(env_file: str = None) -> YourConfig:
"""Return configurations from either environment variables or env_file."""
# Remember to put YourConfig as the first input argument to config_generic
return config_generic(YourConfig, env_file)your_onto.py
from __future__ import annotations # imported to enable pydantic postponed annotations
import pydantic
from typing import Optional
# Module pyderivationagent.data_model.iris.py provides some IRI of basic/common concepts and relationships
# In case you are defining new concepts in your project, you may put it here
YOUR_CONCEPT = 'https://www.example.com/triplestore/repository/Ontology.owl#YOUR_CONCEPT'
# You may also want to define dataclasses for some concepts you are working on
# Please note normally you need to define the concept before it gets referenced by others, as python is a scripting language
class YourOtherConcept(pydantic.BaseModel):
instance_iri: str
# However, if you have to reference other concepts that will have to appear at a later point in your script, for example, in the situation of self-referencing or circular referencing (both of which are quite common in ontologies), you may do so like:
class AnotherConcept(pydantic.BaseModel):
instance_iri: str
refers_to_oneself: AnotherConcept # Here the AnotherConcept self-reference
refers_to_yourconcept: YourConcept # This object relationship refers to YourConcept which will be defined in next lines
# For both reference to work, one must call update_forward_refs() after the definition of YourConcept to make it resolvable
class YourConcept(pydantic.BaseModel):
instance_iri: str
# Here you may want to declare an object relationship that points to YourOtherConcept
# However, you may want to make it Optional with a default None
example_object_relationship_to: Optional[YourOtherConcept] = None
circular_reference_to: AnotherConcept # Here we complete the circular-reference situation
# Call update_forward_refs() to make AnotherConcept resolvable
# For more details, please see https://pydantic-docs.helpmanual.io/usage/postponed_annotations/
AnotherConcept.update_forward_refs()your_sparql.py
from pyderivationagent import PySparqlClient
from youragent.data_model import YourConcept
class YourSparqlClient(PySparqlClient):
# PySparqlClient class provides a few utility functions that developer can call within their own functions:
# - checkInstanceClass(self, instance, instance_class)
# - getAmountOfTriples(self)
# - performQuery(self, query)
# - performUpdate(self, update)
# - uploadOntology(self, filePath)
# - uploadFile(self, local_file_path)
# - downloadFile(self, remote_file_path, downloaded_file_path)
def your_sparql_query(self, your_instance_iri: str) -> YourConcept:
# Construct your SPARQL query string
query = """SELECT ?target WHERE {?target ?predicate ?object.}"""
# Perform SPARQL query with provided method performQuery in PySparqlClient class
response = self.performQuery(query)
# Instantiate TargetClass instances with query results
new_target_instance = YourConcept(response[0]['target'])
return new_target_instance
def your_sparql_update(self, your_instance_iri: str):
# Construct your SPARQL update string
update = """INSERT DATA {<http://subject> <http://predicate> <http://object>.}"""
# Perform SPARQL update with provided method performUpdate in PySparqlClient class
self.performUpdate(update)entry_point.py
from pyderivationagent.conf import config_derivation_agent
from youragent.conf import config_your_agent
from youragent.agent import YourAgent
def create_app():
# If you would like to deploy your agent within a docker container (using docker-compose.yml and youragent.env which will be introduced later in this README.md file), then you may use:
agent_config = config_your_agent() # here we assume custom config are required, for normal config, you may use:
# agent_config = config_derivation_agent()
# Else, if you would like to create agent to run in your memory, then you may want to provide the path to youragent.env file as argument to function config_your_agent(), i.e.,
# agent_config = config_your_agent("/path/to/youragent.env")
# Again, for normal config, you may use:
# agent_config = config_derivation_agent("/path/to/youragent.env")
# Create agent instance
agent = YourAgent(
your_str_conf = agent_config.YOUR_STR_CONF, # remember to populate custom config if applicable
your_int_conf = agent_config.YOUR_INT_CONF, # remember to populate custom config if applicable
your_bool_conf = agent_config.YOUR_BOOL_CONF, # remember to populate custom config if applicable
agent_iri = agent_config.ONTOAGENT_SERVICE_IRI,
time_interval = agent_config.DERIVATION_PERIODIC_TIMESCALE,
derivation_instance_base_url = agent_config.DERIVATION_INSTANCE_BASE_URL,
kg_url = agent_config.SPARQL_QUERY_ENDPOINT,
kg_update_url = agent_config.SPARQL_UPDATE_ENDPOINT,
kg_user = agent_config.KG_USERNAME,
kg_password = agent_config.KG_PASSWORD,
fs_url = agent_config.FILE_SERVER_ENDPOINT,
fs_user = agent_config.FILE_SERVER_USERNAME,
fs_password = agent_config.FILE_SERVER_PASSWORD,
agent_endpoint = agent_config.ONTOAGENT_OPERATION_HTTP_URL,
app = Flask(__name__)
flask_config = FlaskConfig(),
register_agent = agent_config.REGISTER_AGENT,
logger_name = "dev",
# note that you can set the maximum number of threads to monitor async derivations at the same time
max_thread_monitor_async_derivations = agent_config.MAX_THREAD_MONITOR_ASYNC_DERIVATIONS,
# note that you may choose NOT to supply below parameters if you DO NOT want email notifications
email_recipient = agent_config.EMAIL_RECIPIENT,
email_subject_prefix = agent_config.EMAIL_SUBJECT_PREFIX,
email_username = agent_config.EMAIL_USERNAME,
email_auth_json_path = agent_config.EMAIL_AUTH_JSON_PATH,
email_start_end_async_derivations = agent_config.EMAIL_START_END_ASYNC_DERIVATIONS,
)
# Start listening sync/monitoring async derivations
# There are two ways of doing this, the first way it to start the monitoring process independently by:
agent._start_monitoring_derivations() # Option 1
# Or, you can execute below line, which will start all periodical jobs that decorated with @DerivationAgent.periodical_job
# where _start_monitoring_derivations() will be called as well
# agent.start_all_periodical_job() # Option 2, particularly useful when custom periodical job is defined
# Expose flask app of agent
app = agent.app
return appyouragent.env.example
ONTOAGENT_SERVICE_IRI=http://www.example.com/resources/agents/Service__Example#Service
DERIVATION_PERIODIC_TIMESCALE=60
DERIVATION_INSTANCE_BASE_URL=http://www.example.com/triplestore/repository/
SPARQL_QUERY_ENDPOINT=http://www.example.com/blazegraph/namespace/kb/sparql
SPARQL_UPDATE_ENDPOINT=http://www.example.com/blazegraph/namespace/kb/sparql
KG_USERNAME=
KG_PASSWORD=
FILE_SERVER_ENDPOINT=http://www.example.com/FileServer/
FILE_SERVER_USERNAME=
FILE_SERVER_PASSWORD=
ONTOAGENT_OPERATION_HTTP_URL=http://localhost:7000/Example
REGISTER_AGENT=false
MAX_THREAD_MONITOR_ASYNC_DERIVATIONS=1
[email protected];[email protected]
EMAIL_SUBJECT_PREFIX=ExampleAgent
[email protected]
EMAIL_START_END_ASYNC_DERIVATIONS=false
YOUR_STR_CONF=
YOUR_INT_CONF=
YOUR_BOOL_CONF=
You may want to commit this example file without credentials to git as a template for your agent configuration. At deployment, you can make a copy of this file, rename it to youragent.env and populate the credentials information. It is suggested to add *.env entry to your .gitignore of the agent folder, thus the renamed youragent.env (including credentials) will NOT be committed to git. For the usage of each default configuration, please refer to pyderivationagent.conf.AgentConfig class.
NOTE: you may want to provide SPARQL_QUERY_ENDPOINT and SPARQL_UPDATE_ENDPOINT as the internal port of the triple store (most likely blazegraph) docker container, e.g., http://blazegraph:8080/blazegraph/namespace/kb/sparql (blazegraph:8080 depends on your specification in the docker-compose.yml which will be introduced later in this README.md file), if you would like to deploy your derivation agent and the triple store within the same docker stack. An alternative to this is to add extra_hosts: - host.docker.internal:host-gateway to the your_agent service in docker-compose.yml (as shown in below) - then you can access the blazegraph via http://host.docker.internal:27149/blazegraph/namespace/kb/sparql (host.docker.internal:27149 depends on your specification in the docker-compose.yml). Please also note that the host and port of ONTOAGENT_OPERATION_HTTP_URL (i.e., localhost:7000 in http://localhost:7000/Example) should match the value provided in the docker-compose.yml to ensure it is resolvable for handling synchronous derivations once registered in the knowledge graph. At deployment, configurations in this file will be picked up by config_derivation_agent() when instantiating the agent in create_app() of entry_point.py.
docker-compose.yml
version: '3.8'
services:
# Your derivation agent
your_agent:
image: your_agent:1.0.0
container_name: your_agent
environment:
LOG4J_FORMAT_MSG_NO_LOOKUPS: "true"
# Add email auth json path that to be read by the yagmail service
EMAIL_AUTH_JSON_PATH: /run/secrets/email_auth
build:
context: .
dockerfile: ./Dockerfile
ports:
- 7000:5000
# Note that "host.docker.internal" is only a placeholder string, you can replace it with anything, e.g. "localhost" (HOWEVER, NOTE THAT "localhost" IS NO LONGER WORKING AS OF py4jps 1.0.23, WHEREAS ANY OTHER PLACEHOLDER STRING STILL WORKS, AS DETAILED IN ISSUE https://github.com/cambridge-cares/TheWorldAvatar/issues/347)
# But please be aware that this can be unstable on some versions docker-desktop as noticed by other developers:
# https://github.com/docker/for-win/issues/8861
extra_hosts:
- host.docker.internal:host-gateway
env_file:
- ./youragent.env
# Blazegraph
blazegraph:
image: ghcr.io/cambridge-cares/blazegraph:1.2.0
container_name: "blazegraph_test"
ports:
- 27149:8080
environment:
# Use BLAZEGRAPH_USER and BLAZEGRAPH_PASSWORD_FILE if you would like to add authentication
# Otherwise, you may wish to comment them out
BLAZEGRAPH_USER: bg_user
BLAZEGRAPH_PASSWORD_FILE: /run/secrets/blazegraph_password
# Add a secret to set the password for BASIC authentication
secrets:
- blazegraph_password
# Secrets used to set runtime passwords
secrets:
blazegraph_password:
file: tests/dummy_services_secrets/blazegraph_passwd.txt
email_auth: # You may want to add below file name to your .gitignore
file: tests/dummy_services_secrets/email_auth.json
You may refer to DoEAgent for a concrete implementation of the above suggested folder structure based on pyderivationagent. The design of pyderivationagent is continually evolving, and as the project grows, we hope to make it more accessible to developers and users.
The DerivationAgent class provides the feature to send email notifications to list of recipients specified by the developer. As the agent uses yagmail package, a gmail account is required. The feature relies on OAuth2 for authorisation. A step-by-step instruction can be find here.
The pyderivationagent package also provides dockerised integration tests, following the same context as DerivationAsynExample. Interested developer may refer to the README of the Java example for more context, or TheWorldAvatar/JPS_BASE_LIB/python_derivation_agent/tests for more technical details.
One may execute below commands for each set of dockerised integration test:
-
Agents instantiated and run in memory, operating on a blazegraph docker container with dynamic port decided on deployment
(Linux)$ cd /absolute_path_to/TheWorldAvatar/JPS_BASE_LIB/python_derivation_agent $ pytest -s ./tests/test_derivation_agent.py -
All agents and blazegraph deployed within the same docker stack and they communicate via internal port address
(Linux)$ cd /absolute_path_to/TheWorldAvatar/JPS_BASE_LIB/python_derivation_agent $ pytest -s --docker-compose=./docker-compose.test.yml ./tests/test_docker_integration.py -
ExceptionThrowAgentand blazegraph deployed within the same docker stack and they communicate via internal port address(Linux)$ cd /absolute_path_to/TheWorldAvatar/JPS_BASE_LIB/python_derivation_agent $ pytest -s --docker-compose=./docker-compose.test.yml ./tests/test_exception_throw.py
Ideally, we would like to provide this set of dockerised integration test to demo how one may develop integration test for derivation agents. Any ideas/discussions/issues/PRs on how to make this more standardised and accessible to developers are more than welcome.
Developers who add new features to the python agent wrapper handle the distribution of package pyderivationagent on PyPI and Test-PyPI. If you want to add new features that suit your project and release the package independently, i.e. become a developer/maintainer, please contact the repository's administrator to indicate your interest.
The release procedure is currently semi-automated and requires a few items:
- Your Test-PyPI and PyPI account and password
- The version number x.x.x for the release
- Clone of
TheWorldAvatarrepository on your local machine - Docker-desktop is installed and running on your local machine
For a stable version release, please create and checkout to a new branch from your feature branch once you are happy with the feature and above details are ready. The release process can then be started by using the commands below, depending on the operating system you're using. (REMEMBER TO CHANGE THE CORRECT VALUES FOR <absolute_path_to> IN THE COMMANDS BELOW!)
WARNING: at the moment, releasing package from Windows OS has an issue that the package will not be installed correctly while executing "$PIPPATH --disable-pip-version-check install -e $SPATH"[dev]"" in install_script_pip.sh. Please use Linux to release future versions before the issue is fixed.
(Windows)
$ cd \<absolute_path_to>\TheWorldAvatar\JPS_BASE_LIB\python_derivation_agent
$ release_pyderivationagent_to_pypi.sh -v x.x.x(Linux)
$ cd /<absolute_path_to>/TheWorldAvatar/JPS_BASE_LIB/python_derivation_agent
$ ./release_pyderivationagent_to_pypi.sh -v x.x.xPlease follow the instructions presented in the console once the process has begun. If everything goes well, change the version number in JPS_BASE_LIB/python_derivation_agent/release_pyderivationagent_to_pypi.sh to the one you used for the script release.
echo "./release_pyderivationagent_to_pypi.sh -v 0.0.1 - release version 0.0.1"The changes mentioned above should then be committed with the changes performed automatically during the release process, specifically in python script JPS_BASE_LIB/python_derivation_agent/pyderivationagent/__init__.py
__version__ = "0.0.1"
and JPS_BASE_LIB/python_derivation_agent/setup.py
version='0.0.1',
Finally, merge the release branch back to the feature branch and make a Pull Request for the feature branch to be merged back into the main branch.
For development version release, you may do it in your feature branch. The development package will be released to TestPyPI repository by default. Once you have collected the required information, the release process can then be started by using the commands below. (REMEMBER TO CHANGE THE CORRECT VALUES FOR <absolute_path_to> IN THE COMMANDS BELOW!)
NOTE: The development release requires the version number to contain a, b or rc (alpha, beta or release candidate) to be compatible with the pre-release configuration on TestPyPI. You may provide number at the end to differentiate different development version in the same pre-release stage, e.g., "1.1.0a1". In case no number is provided in the end (i.e., "1.0.0a"), "0" will be appended automatically to make it "1.0.0a0". For example, see prerelease-example.
(Linux)
$ cd /<absolute_path_to>/TheWorldAvatar/JPS_BASE_LIB/python_derivation_agent
$ ./release_pyderivationagent_to_pypi.sh -d x.x.xaPlease follow the instructions presented in the console once the process has begun. If everything goes well, change the version number and commit changes following the same procedure as in the stable version release, but no pull request is required.
Jiaru Bai ([email protected])