@kapa.ai You have removed

airbyte-ci

at the Airbyte source code level. After I modify the connector code, how should I build the Docker image?

Based on the knowledge sources, there are two main approaches to build a Docker image for a patched connector without `airbyte-ci`: Option 1: Dockerfile based on the existing connector image (recommended for patching) Create a

Dockerfile

in your connector directory based on the latest published image:

FROM airbyte/source-<your-connector>:latest

COPY . ./airbyte/integration_code
RUN pip install ./airbyte/integration_code

# The entrypoint and default env vars are already set in the base image
# ENV AIRBYTE_ENTRYPOINT "python /airbyte/integration_code/main.py"
ENTRYPOINT ["python", "/airbyte/integration_code/main.py"]

Then build and test it:

docker build -t airbyte/source-<your-connector>:dev .
# Verify it works
docker run airbyte/source-<your-connector>:dev spec

[Patching connector image] Option 2: Using airbyte-cdk image build (for JVM/Python connectors) The

airbyte-cdk

CLI provides a lighter alternative that uses the Dockerfiles in the

docker-images/

directory:

cd airbyte-integrations/connectors/source-<your-connector>
airbyte-cdk image build

This automatically builds both

arm64

and

amd64

images. [Docker Images README] Option 3: Gradle (Java connectors only) For JVM-based connectors, use Gradle:

./gradlew :airbyte-integrations:connectors:<connector-name>:assemble

[Docker Images README] Once built, you can push the image to your own registry with

docker tag

and

docker push

, then register it as a custom connector in the Airbyte UI under Settings → Sources/Destinations → Custom Connector. 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.

@kapa.ai What if I modify the code of the Shopify source connector? How do I build the Docker image without

airbyte-ci

?

Based on the knowledge sources, the Shopify connector's documented build method uses

airbyte-ci

. However, the same Dockerfile-based patching approach that applies to other Python connectors works here too. Create a

Dockerfile

in the

source-shopify

directory:

FROM airbyte/source-shopify:latest

COPY . ./airbyte/integration_code
RUN pip install ./airbyte/integration_code

# The entrypoint and default env vars are already set in the base image
# ENV AIRBYTE_ENTRYPOINT "python /airbyte/integration_code/main.py"
ENTRYPOINT ["python", "/airbyte/integration_code/main.py"]

Then build and verify the image:

docker build -t airbyte/source-shopify:dev .
# Verify it works
docker run airbyte/source-shopify:dev spec

[Shopify README] To push to your own repository, retag and push:

docker tag airbyte/source-shopify:dev your-repo/source-shopify:your-tag
docker push your-repo/source-shopify:your-tag

> Note: The Shopify connector's README only documents

airbyte-ci

as the official build method. The Dockerfile approach above is derived from the pattern used by other Python connectors and is noted as functional but not optimized. [Patching connector image] 👍 0 button 👎 0 button ℹ️ Tip: Mention @kapa.ai in the thread for follow-ups or DM the app for 1-to-1 help.