Report

[picandocodigo]

Test report

This is a Ruby script that generates the ../apis_report.md file.

Usage:

Download artifacts

Download the necessary files with rake download_all. This will download and unzip the Elasticsearch JSON API spec and the elasticsearch-specification spec. The files will be downloaded to ./tmp/rest-api-spec and ./tmp/schema.json respectively. The tmp directory is added to .gitignore.

Generate report

Run rake report to check the available endpoints and which ones have been tested. At the time of writing this, the script checks the ../tests folder for ocurrences of each endpoint name. The script will generate a Markdown file in ../apis_report.md.

There are three main files: reporter.rb, Rakefile and template.erb. The first one has the code for the Elastic::Reporter class. When it is initialized, it gathers all the API names from the spec (excluding names starting with _ like _common and _internal). Then it checks for each API endpoint name, if it's being used in any of the yml files in ./tests. The reporter is initialized in the report task in the Rakefile and the result is passed to the ERB template to generate the markdown file and write it to ./apis_report.md.

TO-DO:

• GitHub Action
• Make STACK_VERSION a variable

[picandocodigo]

Shortcut to what the report looks like just now:
https://github.com/elastic/serverless-clients-tests/blob/ba83b228ff0ff2ddf07a928f3a3be1a68946fe7d/apis_report.md

[picandocodigo]

Not sure if the GitHub Action will work 🤔 (Update: it seems to be working fine, there's just no changes on the report)

It should run for every Pull Request and automatically add a commit to the Pull Request if the report has any changes, or open a new Pull Request if there's a push to main that would change the report (which shouldn't happen if we never push directly to main).

[picandocodigo]

I updated it so that every endpoint that is listed as "tested" includes a link to the test file where it's being tested.

[ezimuel]

Very cool, LGTM!