Github Action: DockerHub Fetch Image-Tag Metadata

Github Action for fetching the metadata of a certain Docker image and tag (including the SHA digests), from the DockerHub public registry.

The API endpoint used for fetching the metadata is (with the example of the official Python image): https://registry.hub.docker.com/v2/repositories/library/python/tags?page=1

Action I/O

Inputs

• image: full image name to find, with format author/image:tag, or image:tag for official images (required)
• os: image OS to find (default: linux)
• architecture: image architecture to find (default: amd64)
• pageLimit: how many pages of results to parse, until giving up (default: 50)

Outputs

If the image is not found, the Action will fail!

• digest: found image digest (example: sha256:ec698176504f2f2d0e50e7e627d7db17be0c8c1a36fe34bb5b7c90c79355f7bb)
• size: found image size, in bytes (example: 43640637)
• Full JSON outputs from the API: these are categorized in two outputs:
  • tagMetadata is the whole object for a certain tag (e.g. python:slim-buster). This includes an array of images, being each image a variant for a certain os and architecture
  • finalImageMetadata is the whole object for a concrete image of the found tag, matching the input os and architecture. This object is filtered out from the array of images found on the "tagMetadata" object

Example

An example of the usage can be found on the test workflow. A simplified example of how the Action is declared would be the following:

steps:
- name: Fetch image metadata
  id: metadata
  uses: David-Lor/action-dockerhub-get-tag-metadata@0.1.1
  with:
    image: debian:slim-buster
    os: linux
    architecture: arm/v7
    pageLimit: 5

- name: Print image metadata
  run: |
    echo "Digest: ${{ steps.metadata.outputs.digest }}"
    echo "Size: ${{ steps.metadata.outputs.size}} bytes"
    echo "Tag Metadata JSON: ${{ steps.metadata.outputs.tagMetadata }}"
    echo "Target Image Metadata JSON: ${{ steps.metadata.outputs.finalImageMetadata }}"

TODO

• Compress node_modules