task-object-storage 1.4.0

Task Object Storage: Transaction management for Robot Framework RPA using MongoDB.

Transaction management for Robot Framework RPA using MongoDB.

Included are two packages: tos and TOSLibrary. The former is the actual task object storage implementation written in pure Python, without any Robot Framework dependencies. TOSLibrary is a Robot Framework keyword library built on top of tos. It is built as a lightweight dynamic proxy library, so all the changes in tos are immediately available in TOSLibrary keywords.

Quickstart guide

Requirements

• MongoDB (version >4)

• Python (version >= 3.9)

Installation

Latest stable version from PyPi with pip:

pip install task-object-storage

Latest development version from PyPi:

pip install task-object-storage --pre

Usage

You can import TOSLibrary into your Robot Framework suite with

*** Settings ***
Library    TOSLibrary    ${db_server}:${db_port}    ${db_name}

If you are using authentication in MongoDB, you can pass credentials:

*** Settings ***
Library    TOSLibrary    ${db_server}:${db_port}    ${db_name}    ${db_user}    ${db_passw}

By default, task data is written to task_objects collection in the specified database. You can also add prefix or suffix to the collection name: pass valid collection_suffix or collection_prefix arguments to TOSLibrary, to access collection with a name your_prefix.task_objects.your_suffix.

It is also possible to separate payloads of task objects to their own payloads collection, and also optionally set a lifetime for them. The user should have full access to the specified collections and ListCollections access to the database. For example, the following role can be created in MongoDB:

db.createRole( {
    role: "rpa_role",
    privileges: [
        { resource: { db: "rpa_data", collection: "my_robot.payloads" }, actions: [ "find", "update", "insert", "remove", "createIndex"] },
        { resource: { db: "rpa_data", collection: "my_robot.task_objects" }, actions: [ "find", "update", "insert", "remove", "createIndex"] },
        { resource: { db: "rpa_data", collection: ""}, actions:  ["listCollections"] }
    ],
    roles: []
})

*** Settings ***
Library    TOSLibrary    ${db_server}:${db_port}    ${db_name}    ${db_user}    ${db_passw}
...    ${collection_suffix}    ${collection_prefix}    db_auth_source=my_auth_source
...    separate_payloads=${TRUE}    payloads_ttl=30 days

Additional MongoClient options can be passed to the library as a dictionary mongo_client_options=. For further details, see the PyMongo documentation: https://pymongo.readthedocs.io/en/4.11.2/api/pymongo/mongo_client.html

After initialization, the keywords are callable from Robot Framework.

This library instance can also be accessed inside Python keywords with

from robot.libraries.BuiltIn import BuiltIn

toslib = BuiltIn().get_library_instance("TOSLibrary")
toslib.keyword_name

The keywords can be also called with:

BuiltIn().run_keyword("<keyword name>")

To use task object storage in Python without any Robot Framework dependencies you should use tos package directly:

from tos.task_object_storage import TaskObjectStorage

tos = TaskObjectStorage(
        db_server=server,
        db_name=db_name,
        db_user=db_user,
        db_passw=db_passw,
        mongo_client_options=mongo_client_options,
)

If db_user and db_passw are empty strings, TaskObjectStorage will default to using no authentication.

Available keywords

See the full API documentation.

For developers

Release a new version by running:

scripts/build.sh release

You can also install the whl package found in the dist directory with

pip install <package-name>.whl

or put inside requirements.txt:

file:C:/path/to/<package-name>.whl  # on windows
file:/path/to/<package-name>.whl  # on linux