Skip to content

Latest commit

 

History

History
28 lines (16 loc) · 1.35 KB

CONTRIBUTING.md

File metadata and controls

28 lines (16 loc) · 1.35 KB

Contributing

Thanks for your interest in contributing to the dataflow Python examples!

Here is the general guidance:

  • Batch Pipelines should live in batch-examples folder

  • Streaming Pipelines should live in Streaming-examples folder

  • Pipelines written using single python module without any other associated scripts should be added to either batch-examples/cookbook-examples or streaming-examples/cookbook-examples depending on the nature of pipeline.

  • Pipelines written either over multiple modules or has associated scripts (e.g: shell scripts / terraform etc) must live in self contained directory as a sub directory under either batch-examples/ or streaming-examples/ depending on the nature of pipeline.

    Every self contained pipeline should follow below guidelines:

    1. Include requirements.txt specifying dependent python packages with respective versions

    2. Include cloudbuild.yaml file consisting 2 steps:

      1. Building a docker image with beam as base image and python packages required for your pipeline.

      2. Run tests

    3. Add an additional step to root level cloudbuild.yaml referring to cloudbuild.yaml inside your sub directory