Skip to content

pass-job-outputs.md

Latest commit

 

History

History
70 lines (60 loc) · 2.86 KB

pass-job-outputs.md

File metadata and controls

70 lines (60 loc) · 2.86 KB
title Passing information between jobs
shortTitle Pass job outputs
intro You can define outputs to pass information from one job to another.
versions
fpt ghes ghec
*
*
*
redirect_from
/actions/using-jobs/defining-outputs-for-jobs
/actions/writing-workflows/choosing-what-your-workflow-does/defining-outputs-for-jobs
/actions/writing-workflows/choosing-what-your-workflow-does/passing-information-between-jobs
/actions/how-tos/writing-workflows/choosing-what-your-workflow-does/passing-information-between-jobs
category
Write workflows
contentType how-tos

Defining and using job outputs

  1. Open the workflow file containing the job you want to get outputs from.

  2. Use the jobs.<job_id>.outputs syntax to define the outputs for the job. For example, the following job defines the output1 and output2 outputs, which are mapped to the results of step1 and step2 respectively:

    jobs:
  job1:
    runs-on: ubuntu-latest
    outputs:
      output1: {% raw %}${{ steps.step1.outputs.test }}{% endraw %}
      output2: {% raw %}${{ steps.step2.outputs.test }}{% endraw %}
    steps:
      - id: step1
        run: echo "test=hello" >> "$GITHUB_OUTPUT"
      - id: step2
        run: echo "test=world" >> "$GITHUB_OUTPUT"

  3. In a separate job where you want to access those outputs, use the jobs.<job_id>.needs syntax to make it dependent on the original job. For example, the following job checks that job1 is complete before running:

    jobs:
  # Assume job1 is defined as above
  job2:
    runs-on: ubuntu-latest
    needs: job1

  4. To access the outputs in the dependent job, use the needs.<job_id>.outputs.<output_name> syntax. For example, the following job accesses the output1 and output2 outputs defined in job1:

    jobs:
  # Assume job1 is defined as above
  job2:
    runs-on: ubuntu-latest
    needs: job1
    steps:
      - env:
          OUTPUT1: {% raw %}${{needs.job1.outputs.output1}}{% endraw %}
          OUTPUT2: {% raw %}${{needs.job1.outputs.output2}}{% endraw %}
        run: echo "$OUTPUT1 $OUTPUT2"

Next steps

To learn more about job outputs and the needs context, see the following sections of AUTOTITLE:

To learn more about passing job outputs from one workflow to another, see the following section of AUTOTITLE: