Loading Data from Attio to Azure Cosmos DB Using dlt in Python

tip

We will be using the dlt PostgreSQL destination to connect to Azure Cosmos DB. You can get the connection string for your Azure Cosmos DB database as described in the Azure Cosmos DB Docs.

Need help deploying these pipelines, or figuring out how to run them in your data stack?

Join our Slack community or book a call with our support engineer Violetta.

Loading data from Attio to Azure Cosmos DB can streamline the management of relationships, deals, and work organization for teams. Attio is a collaborative workspace designed for these purposes, while Azure Cosmos DB is a fully managed NoSQL and relational database ideal for modern app development. By using the open-source python library dlt, you can efficiently transfer and manage your data between these two platforms. This documentation provides detailed steps and guidelines on how to achieve this integration. For more information about Attio, visit their website.

dlt Key Features

• Pipeline Metadata: dlt pipelines leverage metadata to provide governance capabilities, including load IDs for tracking data loads and facilitating data lineage and traceability. Learn more
• Schema Enforcement and Curation: Enforce and curate schemas to ensure data consistency and quality, defining the structure of normalized data and guiding the processing and loading of data. Learn more
• Schema Evolution: dlt alerts users to schema changes, allowing stakeholders to review and validate changes, update downstream processes, or perform impact analysis. Learn more
• Scaling and Finetuning: Offers several mechanisms and configuration options to scale up and finetune pipelines, including parallel execution and memory buffer adjustments. Learn more
• Advanced Features and Community Support: dlt is a constantly growing library that supports many features and use cases needed by the community. Join our Slack to find recent releases or discuss what you can build with dlt. Learn more

Getting started with your pipeline locally

OpenAPI Source Generator dlt-init-openapi

This walkthrough makes use of the dlt-init-openapi generator cli tool. You can read more about it here. The code generated by this tool uses the dlt rest_api verified source, docs for this are here.

0. Prerequisites

dlt and dlt-init-openapi requires Python 3.9 or higher. Additionally, you need to have the pip package manager installed, and we recommend using a virtual environment to manage your dependencies. You can learn more about preparing your computer for dlt in our installation reference.

1. Install dlt and dlt-init-openapi

First you need to install the dlt-init-openapi cli tool.

pip install dlt-init-openapi

The dlt-init-openapi cli is a powerful generator which you can use to turn any OpenAPI spec into a dlt source to ingest data from that api. The quality of the generator source is dependent on how well the API is designed and how accurate the OpenAPI spec you are using is. You may need to make tweaks to the generated code, you can learn more about this here.

# generate pipeline
# NOTE: add_limit adds a global limit, you can remove this later
# NOTE: you will need to select which endpoints to render, you 
# can just hit Enter and all will be rendered.
dlt-init-openapi attio --url https://raw.githubusercontent.com/dlt-hub/openapi-specs/main/open_api_specs/Business/attio_api.yaml --global-limit 2
cd attio_pipeline
# install generated requirements
pip install -r requirements.txt

The last command will install the required dependencies for your pipeline. The dependencies are listed in the requirements.txt:

dlt>=0.4.12

You now have the following folder structure in your project:

attio_pipeline/
├── .dlt/
│   ├── config.toml          # configs for your pipeline
│   └── secrets.toml         # secrets for your pipeline
├── rest_api/                # The rest api verified source
│   └── ...
├── attio/                
│   └── __init__.py          # TODO: possibly tweak this file
├── attio_pipeline.py        # your main pipeline script
├── requirements.txt         # dependencies for your pipeline
└── .gitignore               # ignore files for git (not required)

1.1. Tweak attio/__init__.py

This file contains the generated configuration of your rest_api. You can continue with the next steps and leave it as is, but you might want to come back here and make adjustments if you need your rest_api source set up in a different way. The generated file for the attio source will look like this:

Click to view full file (399 lines)

from typing import List

import dlt
from dlt.extract.source import DltResource
from rest_api import rest_api_source
from rest_api.typing import RESTAPIConfig


@dlt.source(name="attio_source", max_table_nesting=2)
def attio_source(
        base_url: str = dlt.config.value,
) -> List[DltResource]:

    # source configuration
    source_config: RESTAPIConfig = {
        "client": {
            "base_url": base_url,
            "paginator": {
                "type":
                    "offset",
                "limit":
                    20,
                "offset_param":
                    "offset",
                "limit_param":
                    "limit",
                "total_path":
                    "",
                "maximum_offset":
                    20,
            },
        },
        "resources": 
        [
        # Lists all attributes defined on a specific object or list. Attributes are returned in the order that they are sorted by in the UI.  Required scopes: `object_configuration:read`.
        {
            "name": "get_v_2_targetidentifierattributes",
            "table_name": "attribute",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/{target}/{identifier}/attributes",
                "params": {
                    "target": "FILL_ME_IN", # TODO: fill in required path parameter
                    "identifier": "FILL_ME_IN", # TODO: fill in required path parameter
                    # the parameters below can optionally be configured
                    # "show_archived": "OPTIONAL_CONFIG",

                },
            }
        },
        # Gets information about a single attribute on either an object or a list.  Required scopes: `object_configuration:read`.
        {
            "name": "get_v_2_targetidentifierattributesattribute",
            "table_name": "attribute",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/{target}/{identifier}/attributes/{attribute}",
                "params": {
                    "target": "FILL_ME_IN", # TODO: fill in required path parameter
                    "identifier": "FILL_ME_IN", # TODO: fill in required path parameter
                    "attribute": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # Get a single comment by ID.  To view comments on records, you will need the `object_configuration:read` and `record_permission:read` scopes.  To view comments on list entries, you will need the `list_configuration:read` and `list_entry:read` scopes.  Required scopes: `comment:read`.
        {
            "name": "get_v_2_commentscomment_id",
            "table_name": "comment",
            "primary_key": "thread_id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/comments/{comment_id}",
                "params": {
                    "comment_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # List all entries, across all lists, for which this record is the parent.  Required scopes: `record_permission:read`, `object_configuration:read`, `list_entry:read`.
        {
            "name": "get_v_2_objectsobjectrecordsrecord_identries",
            "table_name": "entry",
            "primary_key": "entry_id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/objects/{object}/records/{record_id}/entries",
                "params": {
                    "object": "FILL_ME_IN", # TODO: fill in required path parameter
                    "record_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # Gets a single list entry by its `entry_id`.  Required scopes: `list_entry:read`, `list_configuration:read`.
        {
            "name": "get_v_2_listslistentriesentry_id",
            "table_name": "entry",
            "primary_key": "parent_record_id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/lists/{list}/entries/{entry_id}",
                "params": {
                    "list": "FILL_ME_IN", # TODO: fill in required path parameter
                    "entry_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # List all lists that your access token has access to. lists are returned in the order that they are sorted in the sidebar.  Required scopes: `list_configuration:read`.
        {
            "name": "get_v_2_lists",
            "table_name": "list",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/lists",
            }
        },
        # Gets a single list in your workspace that your access token has access to.  Required scopes: `list_configuration:read`.
        {
            "name": "get_v_2_listslist",
            "table_name": "list",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/lists/{list}",
                "params": {
                    "list": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # List notes for all records or for a specific record.  Required scopes: `note:read`, `object_configuration:read`, `record_permission:read`.
        {
            "name": "get_v_2_notes",
            "table_name": "note",
            "primary_key": "parent_record_id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/notes",
                "params": {
                    # the parameters below can optionally be configured
                    # "parent_object": "OPTIONAL_CONFIG",
                    # "parent_record_id": "OPTIONAL_CONFIG",

                },
            }
        },
        # Get a single note by ID.  Required scopes: `note:read`, `object_configuration:read`, `record_permission:read`.
        {
            "name": "get_v_2_notesnote_id",
            "table_name": "note",
            "primary_key": "parent_record_id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/notes/{note_id}",
                "params": {
                    "note_id": {
                        "type": "resolve",
                        "resource": "get_v_2_notes",
                        "field": "parent_record_id",
                    },

                },
            }
        },
        # Lists all system-defined and user-defined objects in your workspace.  Required scopes: `object_configuration:read`.
        {
            "name": "get_v_2_objects",
            "table_name": "object",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/objects",
            }
        },
        # Gets a single object by its `object_id` or slug.  Required scopes: `object_configuration:read`.
        {
            "name": "get_v_2_objectsobject",
            "table_name": "object",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/objects/{object}",
                "params": {
                    "object": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # Gets a single person, company or other record by its `record_id`.  Required scopes: `record_permission:read`, `object_configuration:read`.
        {
            "name": "get_v_2_objectsobjectrecordsrecord_id",
            "table_name": "record",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/objects/{object}/records/{record_id}",
                "params": {
                    "object": "FILL_ME_IN", # TODO: fill in required path parameter
                    "record_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # Lists all select options for a particular attribute on either an object or a list.  Required scopes: `object_configuration:read`.
        {
            "name": "get_v_2_targetidentifierattributesattributeoptions",
            "table_name": "select_option",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/{target}/{identifier}/attributes/{attribute}/options",
                "params": {
                    "target": "FILL_ME_IN", # TODO: fill in required path parameter
                    "identifier": "FILL_ME_IN", # TODO: fill in required path parameter
                    "attribute": "FILL_ME_IN", # TODO: fill in required path parameter
                    # the parameters below can optionally be configured
                    # "show_archived": "OPTIONAL_CONFIG",

                },
            }
        },
        # Identify the current access token, the workspace it is linked to, and any permissions it has.
        {
            "name": "get_v_2_self",
            "table_name": "self",
            "primary_key": "sub",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "$",
                "path": "/v2/self",
            }
        },
        # Lists all statuses for a particular status attribute on either an object or a list.  Required scopes: `object_configuration:read`.
        {
            "name": "get_v_2_targetidentifierattributesattributestatuses",
            "table_name": "status",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/{target}/{identifier}/attributes/{attribute}/statuses",
                "params": {
                    "target": "FILL_ME_IN", # TODO: fill in required path parameter
                    "identifier": "FILL_ME_IN", # TODO: fill in required path parameter
                    "attribute": "FILL_ME_IN", # TODO: fill in required path parameter
                    # the parameters below can optionally be configured
                    # "show_archived": "false",

                },
            }
        },
        # List all tasks. Results are sorted by creation date, from oldest to newest.  Required scopes: `task:read`, `object_configuration:read`, `record_permission:read`, `user_management:read`.
        {
            "name": "get_v_2_tasks",
            "table_name": "task",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/tasks",
                "params": {
                    # the parameters below can optionally be configured
                    # "sort": "OPTIONAL_CONFIG",
                    # "linked_object": "OPTIONAL_CONFIG",
                    # "linked_record_id": "OPTIONAL_CONFIG",
                    # "assignee": "OPTIONAL_CONFIG",
                    # "is_completed": "OPTIONAL_CONFIG",

                },
            }
        },
        # Get a single task by ID.  Required scopes: `task:read`, `object_configuration:read`, `record_permission:read`, `user_management:read`.
        {
            "name": "get_v_2_taskstask_id",
            "table_name": "task",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/tasks/{task_id}",
                "params": {
                    "task_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # List threads of comments on a record or list entry.  To view threads on records, you will need the `object_configuration:read` and `record_permission:read` scopes.  To view threads on list entries, you will need the `list_configuration:read` and `list_entry:read` scopes.  Required scopes: `comment:read`.
        {
            "name": "get_v_2_threads",
            "table_name": "thread",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/threads",
                "params": {
                    # the parameters below can optionally be configured
                    # "record_id": "OPTIONAL_CONFIG",
                    # "object": "OPTIONAL_CONFIG",
                    # "entry_id": "OPTIONAL_CONFIG",
                    # "list": "OPTIONAL_CONFIG",

                },
            }
        },
        # Get all comments in a thread.  To view threads on records, you will need the `object_configuration:read` and `record_permission:read` scopes.  To view threads on list entries, you will need the `list_configuration:read` and `list_entry:read` scopes.  Required scopes: `comment:read`.
        {
            "name": "get_v_2_threadsthread_id",
            "table_name": "thread",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/threads/{thread_id}",
                "params": {
                    "thread_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # Gets all values for a given attribute on a record. If the attribute is historic, this endpoint has the ability to return all historic values using the `show_historic` query param.  Required scopes: `record_permission:read`, `object_configuration:read`.
        {
            "name": "get_v_2_objectsobjectrecordsrecord_idattributesattributevalues",
            "table_name": "value",
            "primary_key": "referenced_actor_id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/objects/{object}/records/{record_id}/attributes/{attribute}/values",
                "params": {
                    "object": "FILL_ME_IN", # TODO: fill in required path parameter
                    "record_id": "FILL_ME_IN", # TODO: fill in required path parameter
                    "attribute": "FILL_ME_IN", # TODO: fill in required path parameter
                    # the parameters below can optionally be configured
                    # "show_historic": "false",

                },
            }
        },
        # Gets all values for a given attribute on a list entry. If the attribute is historic, this endpoint has the ability to return all historic values using the `show_historic` query param.  Required scopes: `list_entry:read`, `list_configuration:read`.
        {
            "name": "get_v_2_listslistentriesentry_idattributesattributevalues",
            "table_name": "value",
            "primary_key": "referenced_actor_id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/lists/{list}/entries/{entry_id}/attributes/{attribute}/values",
                "params": {
                    "list": "FILL_ME_IN", # TODO: fill in required path parameter
                    "entry_id": "FILL_ME_IN", # TODO: fill in required path parameter
                    "attribute": "FILL_ME_IN", # TODO: fill in required path parameter
                    # the parameters below can optionally be configured
                    # "show_historic": "false",

                },
            }
        },
        # Get all of the webhooks in your workspace.  Required scopes: `webhook:read`.
        {
            "name": "get_v_2_webhooks",
            "table_name": "webhook",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/webhooks",
            }
        },
        # Get a single webhook.  Required scopes: `webhook:read`.
        {
            "name": "get_v_2_webhookswebhook_id",
            "table_name": "webhook",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/webhooks/{webhook_id}",
                "params": {
                    "webhook_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        # Lists all workspace members in the workspace.  Required scopes: `user_management:read`.
        {
            "name": "get_v_2_workspace_members",
            "table_name": "workspace_member",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/workspace_members",
            }
        },
        # Gets a single workspace member by ID.  Required scopes: `user_management:read`.
        {
            "name": "get_v_2_workspace_membersworkspace_member_id",
            "table_name": "workspace_member",
            "endpoint": {
                "data_selector": "data",
                "path": "/v2/workspace_members/{workspace_member_id}",
                "params": {
                    "workspace_member_id": "FILL_ME_IN", # TODO: fill in required path parameter

                },
            }
        },
        ]
    }

    return rest_api_source(source_config)

2. Configuring your source and destination credentials

info

dlt-init-openapi will try to detect which authentication mechanism (if any) is used by the API in question and add a placeholder in your secrets.toml.

• If you know your API needs authentication, but none was detected, you can learn more about adding authentication to the rest_api here.
• OAuth detection currently is not supported, but you can supply your own authentication mechanism as outlined here.

The dlt cli will have created a .dlt directory in your project folder. This directory contains a config.toml file and a secrets.toml file that you can use to configure your pipeline. The automatically created version of these files look like this:

generated config.toml

[runtime]
log_level="INFO"

[sources.attio]
# Base URL for the API
# Production
base_url = "https://api.attio.com"

generated secrets.toml

[sources.attio]
# secrets for your attio source
# example_api_key = "example value"

2.1. Adjust the generated code to your usecase

Further help setting up your source and destinations

At this time, the dlt-init-openapi cli tool will always create pipelines that load to a local duckdb instance. Switching to a different destination is trivial, all you need to do is change the destination parameter in attio_pipeline.py to postgres and supply the credentials as outlined in the destination doc linked below.

• Read more about setting up the rest_api source in our docs.
• Read more about setting up the Azure Cosmos DB destination in our docs.

3. Running your pipeline for the first time

The dlt cli has also created a main pipeline script for you at attio_pipeline.py, as well as a folder attio that contains additional python files for your source. These files are your local copies which you can modify to fit your needs. In some cases you may find that you only need to do small changes to your pipelines or add some configurations, in other cases these files can serve as a working starting point for your code, but will need to be adjusted to do what you need them to do.

The main pipeline script will look something like this:

import dlt

from attio import attio_source


if __name__ == "__main__":
    pipeline = dlt.pipeline(
        pipeline_name="attio_pipeline",
        destination='duckdb',
        dataset_name="attio_data",
        progress="log",
        export_schema_path="schemas/export"
    )
    source = attio_source()
    info = pipeline.run(source)
    print(info)

Provided you have set up your credentials, you can run your pipeline like a regular python script with the following command:

python attio_pipeline.py

4. Inspecting your load result

You can now inspect the state of your pipeline with the dlt cli:

dlt pipeline attio_pipeline info

You can also use streamlit to inspect the contents of your Azure Cosmos DB destination for this:

# install streamlit
pip install streamlit
# run the streamlit app for your pipeline with the dlt cli:
dlt pipeline attio_pipeline show

5. Next steps to get your pipeline running in production

One of the beauties of dlt is, that we are just a plain Python library, so you can run your pipeline in any environment that supports Python >= 3.8. We have a couple of helpers and guides in our docs to get you there:

The Deploy section will show you how to deploy your pipeline to

• Deploy with GitHub Actions: Learn how to set up and deploy a dlt pipeline using GitHub Actions for CI/CD. Follow the step-by-step guide here.

• Deploy with Airflow and Google Composer: Discover how to deploy a dlt pipeline using Airflow and Google Composer. Detailed instructions can be found here.

• Deploy with Google Cloud Functions: Explore the process of deploying a dlt pipeline using Google Cloud Functions. Check out the guide here.

• Other Deployment Methods: For more deployment options and methods, visit the comprehensive guide here.

The running in production section will teach you about:

• How to Monitor your pipeline: Learn how to effectively monitor your dlt pipeline to ensure smooth operations and quick identification of issues. How to Monitor your pipeline
• Set up alerts: Configure alerts for your dlt pipeline to get notified of any critical events or failures that need immediate attention. Set up alerts
• Set up tracing: Implement tracing in your dlt pipeline to gain detailed insights into the execution flow and performance metrics. And set up tracing

Available Sources and Resources

For this verified source the following sources and resources are available

Source Attio

Attio source provides CRM data including workspace members, attributes, comments, statuses, and tasks.

Resource Name	Write Disposition	Description
workspace_member	append	Details of members in the workspace
attribute	append	Various attributes associated with records
comment	append	Comments added by users
self	append	Information about the authenticated user
select_option	append	Options for select fields
status	append	Status updates for records
list	append	Lists of records
thread	append	Threads of conversations or activities
record	append	Individual records in the workspace
value	append	Values assigned to attributes
object	append	Various objects in the workspace
webhook	append	Webhooks configured for the workspace
entry	append	Entries in the workspace
note	append	Notes added to records
task	append	Tasks assigned to users

Additional pipeline guides

• Load data from Crypt API to Google Cloud Storage in python with dlt
• Load data from Chess.com to BigQuery in python with dlt
• Load data from Klarna to AlloyDB in python with dlt
• Load data from Microsoft SQL Server to Azure Synapse in python with dlt
• Load data from IBM Db2 to YugabyteDB in python with dlt
• Load data from Soundcloud to BigQuery in python with dlt
• Load data from DigitalOcean to PostgreSQL in python with dlt
• Load data from Microsoft SQL Server to CockroachDB in python with dlt
• Load data from Looker to Redshift in python with dlt
• Load data from GitLab to CockroachDB in python with dlt