This action runs Splunk's AppInspect CLI against a provided directory of Splunk App. It fails if the result contains any failures or manual checks are not vetted.

The (json) result will be written to the file specified with result-file. This can be uploaded for later viewing to use in another step/job using actions/upload-artifact@v2.

For a more comprehensive Splunk app testing workflow, visit the splunk/splunk-app-testing which includes a workflow for cypress testing.

Required: The path to directory of the app in the working directory.

The file name to use for the json result.

default: appinspect_result.json

Appinspect tags to include

required: false

Appinspect tags to exclude

required: false

Path to file which contains list of manual checks

required: false default: .appinspect.manualcheck.yaml

Path to file which contains list of expected appinspect failures

required: false default: .appinspect.expect.yaml

Path to generated file with markdown for manual checks

required: false default: manual_check_markdown.txt

Path to generated file with markdown for expected appinspect failures

required: false default: expected_failure_markdown.txt

pass|fail

To see checks to be verified, inspect the result_file from appinspect-cli-action. Verify manual checks and mark them as reviewed by adding them one by one into .appinspect.manualcheck.yaml, ex:

name_of_manual_check_1:
  comment: 'your comment'
name_of_manual_check_2:
  comment: 'your comment'

Please note that names of validated manual checks should be aligned with those from result_file and your comment can't be empty.

To mark Failures as expected, add them into .appinspect.expect.yaml with proper comment containing ticket id of ADDON/APPCERT project associated with the exception, ex:

name_of_exception_1:
  comment: 'ADDON-123: your comment'
name_of_exception_2:
  comment: 'APPCERT-123: your comment'

Please note that your comment can't be empty, it must include ticket id of ADDON/APPCERT project associated with the exception and the names of exceptions should be aligned with those from result_file.

When appinspect-cli-action is called, it scans the package with Splunk's AppInspect CLI. If there are any failures observed then action compares results_file with .appinspect.expect.yaml. If that failure isn't present in .appinspect.expect.yaml or it does not contain an appropriate comment(containing ADDON/APPCERT ticket id associated with the exception) then the job fails with proper failure reason. In the next step, action compares results_file with .appinspect.manualcheck.yaml. If any manual check wasn't reviewed by addon developer and isn't in .appinspect.manualcheck.yaml then the job fails.

- uses: splunk/appinspect-cli-action@v1
  with:
    app_path: 'test'

If the comparison is successful then a markdown consisting a table with check names and comments is generated. It can be uploaded to artifacts.

- uses: actions/checkout@v2
- uses: splunk/appinspect-cli-action@v1.3
  with:
    app_path: 'test'
    included_tags: {appinspect-tags-to-include}
    manual_check_markdown: manual_check_markdown.txt
    expected_failure_markdown: expected_failure_markdown.txt
- name: upload-manual-check-markodown
  uses: actions/upload-artifact@v2
  with:
    name: manual_check_markdown.txt
    path: manual_check_markdown.txt
- name: upload-expected_failure-markodown
  uses: actions/upload-artifact@v2
  with:
    name: expected_failure_markdown.txt
    path: expected_failure_markdown.txt

The markdown is ready to paste into confluence, by: Edit -> Insert more content -> Markup, change insert type to Markdown and paste the contents of the file.