This message was deleted.

What will be using the name in the next step? If it's another Pulumi program, you would use a stack reference. If it's anything else, you would invoke

pulumi stack output

and grab the value from there.

it another github action step eg:

- name: upload package to bucket
        run: aws s3 cp accessRequest.zip <s3-bucket-name>

i will try adding pulumi stack output thanks

The Pulumi action doesn't support the

stack

command. If you need it in another step, then you need to run the

pulumi

command directly, and parse its output.

out put will be consumed by another github action step

- uses: pulumi/actions@v3
        with:
          command: up
          stack-name: dev
          #comment-on-pr: true
          upsert: true
          refresh: true
          work-dir: awsResource/s3
        env:
          PULUMI_ACCESS_TOKEN: ${{ secrets.PULUMI_ACCESS_TOKEN }}
          PULUMI_CONFIG_PASSPHRASE: ${{ secrets.PULUMI_CONFIG_PASSPHRASE }}

      - name: upload foo package to bucket
        run: aws s3 cp foo.zip <bucket name from output of pulumi up >

Ah. If that can't be done in the Pulumi program (maybe by using a file asset) then you will want to use

pulumi stack output | grep bucketName

or similar. Unless the Pulumi action can store outputs in a workflow variable or similar... I don't think I've read anything about that.

This worked for me

-  name: get s3 bucket name to push package
         id: pulumiStackOutput
         working-directory: awsResource/s3
         env:
          PULUMI_CONFIG_PASSPHRASE: ${{ secrets.PULUMI_CONFIG_PASSPHRASE }}
         run: echo ::set-output name=bucketname::$(pulumi login ${{ secrets.AWS_BUCKET }}  > /dev/null 2>&1  && pulumi stack output -j | jq -r .bucketName)
      -  name: print bucket name 
         run: echo ${{ steps.pulumiStackOutput.outputs.bucketname }}

Yep, it's a bit hairy, but maybe a request for an option to the "up" Pulumi action that puts outputs (or at least, some named outputs) into a workflow variable would be good...