Skip to content

Latest commit

 

History

History
46 lines (26 loc) · 3.18 KB

CONTRIBUTING.md

File metadata and controls

46 lines (26 loc) · 3.18 KB
Raw

Contributing

Requirements

  • go 1.19.5 (or later)

Installation

Clone the repository into your $GOPATH and add that directory as a workflow within Alfred (make sure to disable any other versions of this workflow in Alfred so that the aws keyword doesn't conflict). Then, from the root of this repo, run:

# you must run this whenever you make changes to the go files
./build.sh

You should now be able to invoke the Alfred workflow from your cloned repo.

Adding services or sub-services

To add/update/remove services or sub-services, simply update the .yml config file with your changes. This file is used by the executable to populate entries in Alfred (for a list of valid properties, see the models file). Once you've done that, run ./test.sh to update any snapshots with the data changes you've made.

Bulk Verifying URLs

If you're wanting to verify all sub-service URLs for a particular service, you can use OPEN_ALL as a sub-service term within the Alfred workflow. For example, to open all EC2 sub-service URLs, use aws ec2 OPEN_ALL. This will allow you to just go through the tabs opened in your browser to verify that the links are not broken.

Adding a searcher

Generally, if you're just wanting to add a searcher, you can follow the patterns/examples from already implemented searchers (like the EC2 Instance searcher). Some things to keep in mind:

  • Each new searcher must have tests. The tests in this repo use the following libraries/patterns:
    • go-vcr for recording AWS requests to be re-used deterministically in CI. Ensure that any new AWS fixtures that you add are purged of any sensitive information related to your account. There are sanitizers implemented to scrub some of this data (please feel free to add to them), though this is hard to exhaustively do automatically.
    • cupaloy for snapshot testing to assert results in the Alfred Workflow. The ./test.sh script should automatically update snapshots on every run, so make sure that the snapshots look the way that they should before committing them.
  • Each new searcher should have a test case in this file. This file is an integration-style test that will massage logic across the entire workflow. The new test case should have a query that matches the searcher that you added.

Advanced

For more rapid development, you can also run:

./run.sh -query='ec2 instances' # example query; don't prefix string with `aws` here!

This will build and run a single execution of the workflow and log any relevant output to the console. Keep in that the ./run.sh environment is not 100% the same as running within Alfred, so you should always verify your changes via ./build.sh and invoking this workflow manually in Alfred.