Skip to main content

Loading Data from Imgur to Databricks with dlt in Python

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 Imgur to Databricks can be streamlined using the open-source Python library, dlt. Imgur is an online image sharing and hosting service known for viral images and memes, especially those shared on Reddit. Databricks, created by the original developers of Apache Spark™, is a unified data analytics platform that enhances innovation by integrating data science, engineering, and business. This documentation will guide you through the process of using dlt to efficiently transfer data from Imgur to Databricks. For further details on Imgur, visit Imgur's website.

dlt Key Features​

  • Automated Schema Management: dlt automatically handles schema inference and evolution, making it easier to maintain data consistency. Read more
  • Scalable Data Extraction: Leverage iterators, chunking, and parallelization for efficient processing of large datasets. Learn how
  • Governance Support: Utilize pipeline metadata, schema enforcement, and schema change alerts for robust governance. Explore governance features
  • Declarative Interface: User-friendly, declarative interface that removes knowledge obstacles for beginners while empowering senior professionals. Get started
  • Integration with Airflow: Seamlessly deploy dlt pipelines to Airflow, ensuring data consistency and granular loading. Find out 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 imgur --url https://raw.githubusercontent.com/dlt-hub/openapi-specs/main/open_api_specs/Public/imgur.yaml --global-limit 2
cd imgur_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:

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

1.1. Tweak imgur/__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 imgur source will look like this:

Click to view full file (110 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="imgur_source", max_table_nesting=2)
def imgur_source(
        api_key: str = dlt.secrets.value,
        base_url: str = dlt.config.value,
) -> List[DltResource]:

    # source configuration
    source_config: RESTAPIConfig = {
        "client": {
            "base_url": base_url,
            "auth": {
                
            "type": "api_key",
            "api_key": api_key,
            "name": "Authorization",
            "location": "header"

            },
        },
        "resources": 
        [
        {
            "name": "get_account",
            "table_name": "account_response",
            "endpoint": {
                "data_selector": "$",
                "path": "/3/account/{userName}",
                "params": {
                    "userName": "FILL_ME_IN", # TODO: fill in required path parameter

                },
                "paginator": "auto",
            }
        },
        {
            "name": "get_account_images_count",
            "table_name": "basic_int_32_response",
            "endpoint": {
                "data_selector": "$",
                "path": "/3/account/{userName}/images/count",
                "params": {
                    "userName": {
                        "type": "resolve",
                        "resource": "get_account_images",
                        "field": "id",
                    },

                },
                "paginator": "auto",
            }
        },
        {
            "name": "get_account_images",
            "table_name": "image",
            "primary_key": "id",
            "write_disposition": "merge",
            "endpoint": {
                "data_selector": "data",
                "path": "/3/account/{userName}/images",
                "params": {
                    "userName": "FILL_ME_IN", # TODO: fill in required path parameter

                },
                "paginator": "auto",
            }
        },
        {
            "name": "get_account_image",
            "table_name": "image_response",
            "endpoint": {
                "data_selector": "$",
                "path": "/3/account/{userName}/images/{imageHash}",
                "params": {
                    "imageHash": {
                        "type": "resolve",
                        "resource": "get_account_images",
                        "field": "id",
                    },
                    "userName": "FILL_ME_IN", # TODO: fill in required path parameter

                },
                "paginator": "auto",
            }
        },
        {
            "name": "get_image",
            "table_name": "image_response",
            "endpoint": {
                "data_selector": "$",
                "path": "/3/image/{imageHash}",
                "params": {
                    "imageHash": "FILL_ME_IN", # TODO: fill in required path parameter

                },
                "paginator": "auto",
            }
        },
        ]
    }

    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.imgur]
# Base URL for the API
base_url = "https://api.imgur.com"

generated secrets.toml​


[sources.imgur]
# secrets for your imgur source
api_key = "FILL ME OUT"  # TODO: fill in your credentials

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 imgur_pipeline.py to databricks 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 Databricks 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 imgur_pipeline.py, as well as a folder imgur 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 imgur import imgur_source


if __name__ == "__main__":
    pipeline = dlt.pipeline(
        pipeline_name="imgur_pipeline",
        destination='duckdb',
        dataset_name="imgur_data",
        progress="log",
        export_schema_path="schemas/export"
    )
    source = imgur_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 imgur_pipeline.py

4. Inspecting your load result​

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

dlt pipeline imgur_pipeline info

You can also use streamlit to inspect the contents of your Databricks destination for this:

# install streamlit
pip install streamlit
# run the streamlit app for your pipeline with the dlt cli:
dlt pipeline imgur_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: Easily set up CI/CD for your dlt pipeline using GitHub Actions.
  • Deploy with Airflow: Utilize Google Composer or any Airflow instance to manage and schedule your dlt pipelines with Airflow.
  • Deploy with Google Cloud Functions: Leverage the serverless capabilities of Google Cloud to deploy your dlt pipeline using Google Cloud Functions.
  • Explore Other Deployment Options: Discover various other methods and platforms for deploying your dlt pipeline here.

The running in production section will teach you about:​

  • How to Monitor your pipeline: Learn how to effectively monitor your dlt pipeline in production to ensure smooth operation and timely issue detection. How to Monitor your pipeline
  • Set up alerts: Configure alerts to stay informed about critical events and potential issues in your dlt pipeline. Set up alerts
  • Set up tracing: Implement tracing to gain insights into the performance and behavior of your dlt pipeline. And set up tracing

Available Sources and Resources​

For this verified source the following sources and resources are available

Source Imgur​

Fetches image, account, and interaction data from Imgur.

Resource NameWrite DispositionDescription
image_responseappendDetails about individual images hosted on Imgur, including metadata and image statistics.
account_responseappendInformation about user accounts on Imgur, such as account settings and user activity.
basic_int_32_responseappendBasic response containing integer values, potentially used for counters or simple metrics.
imageappendCore data about images uploaded to Imgur, including URLs, upload timestamps, and image properties.

Additional pipeline guides​

This demo works on codespaces. Codespaces is a development environment available for free to anyone with a Github account. You'll be asked to fork the demo repository and from there the README guides you with further steps.
The demo uses the Continue VSCode extension.

Off to codespaces!

DHelp

Ask a question

Welcome to "Codex Central", your next-gen help center, driven by OpenAI's GPT-4 model. It's more than just a forum or a FAQ hub – it's a dynamic knowledge base where coders can find AI-assisted solutions to their pressing problems. With GPT-4's powerful comprehension and predictive abilities, Codex Central provides instantaneous issue resolution, insightful debugging, and personalized guidance. Get your code running smoothly with the unparalleled support at Codex Central - coding help reimagined with AI prowess.