Benchmarks / Contributing Updates

[mikemimik]

What / Why

What

Contains an update to benchmarks.yml workflow file that does the following:

• removes unnecessary checkout of repository
• removes unnecessary setup of nodejs
• adds ability to discern if pull-request event is coming from current repository or a forked repository

Contains an additional workflow file benchmarks-comments.yml file which does the following:

• detects if the triggering comment came from a pull-request
• checks if the comments starts with "test this please ✅"
• dispatches a request to get benchmarks (same action taken as benchmarks.yml file)
• adds a rocket reaction to the comment the first time triggered

Contains an updated CONTRIBUTING.md file which has updated information about contributing to this project. This includes information about the above mentioned benchmarking for pull-requests.

Why

The reason for benchmarks-comments.yml is because when a pull-request from a forked repository triggers an event, the github action that runs doesn't get scoped with secret content (this is for security purposes). Making it impossible to dispatch and trigger the benchmark suite (requires authentication from secrets).

References

• closes: Update ci.yml to send correct payload information to benchmark repo benchmarks#3
• closes: Benchmarks work v2 statusboard#27

[npm-deploy-user]

	angular-quickstart			app-large			app-medium			ember-quickstart			react-app
	prev	current	status	prev	current	status	prev	current	status	prev	current	status	prev	current	status
initial install	41s	36.4s	✅	39.5s	37s	✅	34.5s	30.7s	✅	28.1s	23.2s	✅	33.5s	29.9s	✅
repeat install	9.6s	7.9s	✅	8.6s	7.8s	✅	8.3s	7.7s	✅	7.7s	6.5s	✅	9.1s	7.9s	✅
with warm cache	32.3s	27.8s	✅	33.8s	30.4s	✅	31.3s	27.8s	✅	23.4s	21.7s	✅	29.1s	25.2s	✅
with node_modules	9.2s	7.9s	✅	8.6s	7s	✅	9s	8.2s	✅	7.8s	6.5s	✅	9.4s	8.1s	✅
with lockfile	32.6s	26.1s	✅	31s	29.7s	✅	29s	29.6s	✅🐌	21.7s	19.1s	✅	27.4s	23.7s	✅
with warm cache and node_modules	9.3s	8.2s	✅	7.9s	6.7s	✅	8.4s	8.2s	✅	7.6s	7.5s	✅	9.2s	7.9s	✅
with warm cache and lockfile	24.7s	22.7s	✅	26.3s	25.4s	✅	24.3s	19.9s	✅	17.8s	15.4s	✅	21.3s	20.1s	✅
with node_modules and lockfile	10.1s	9.5s	✅	9.8s	8s	✅	8.6s	7.9s	✅	7.8s	7s	✅	9.7s	8.7s	✅

[isaacs]

This looks fine to me, but I am not familiar enough with the benchmarking code to really be a great judge here.