v7.5.2

This release makes internal changes to upgrade the Action to node20 and it also updates all internal node dependencies with npm update

What's Changed

• Upgrade to node20 by @GrantBirki in #206
• update all node dependencies by @GrantBirki in #207

Full Changelog: v7.5.1...v7.5.2

v7.5.1

v7.5.1

This release adds new documentation and squashes a long living bug related to how merge commits are checked when using the "Merge Commit Strategy" workflow. @chrisgavin lent a hand to squash this tricky bug and now the "Merge Commit Strategy" workflow will continue to run as expected, even if you make merge commits into your branch, update your branch, or resolve merge conflicts 🎉.

What's Changed

• Add multi-environment example by @ncalteen in #203
• Simplify identical commit check using tree SHAs. by @chrisgavin in #204
• Identical commit check fixes by @GrantBirki in #205

New Contributors

• @ncalteen made their first contribution in #203
• @chrisgavin made their first contribution in #204

Full Changelog: v7.5.0...v7.5.1

v7.5.0

v7.5.0

@hubot Style Deploy Locking 🔒

This release introduces Hubot Style Deploy Locking (aka sticky deployment locks). Currently, when you run .deploy it creates a lock during the deployment and then releases the lock when the deployment completes. By using the new input option that this release introduces, you can change the deployment locking behavior so that the lock persists even after the deployment finishes.

New Input Options:

• sticky_locks - By default, this value is set to "false".
• sticky_locks_for_noop - By default, this value is set to "false". You should probably leave it disabled unless you have a significant reason to lock an environment due to a .noop style deployment

Example:

- name: branch-deploy
  id: branch-deploy
  uses: github/[email protected]
    with:
      sticky_locks: "true" # <--- enables sticky deployment lock / hubot style deployment locks
      sticky_locks_for_noop: "true" # <--- enables sticky deployment lock / hubot style deployment locks for noop deployments
      # ... other configuration

This option, combined with the "Unlock on Merge" workflow strategy is highly suggested for mission critical projects using this Action to deploy their code to production.

You can learn more about this new method of deployment locking by reading the new documentation 📚.

This release will be internally tested at GitHub before being set to the latest release

What's Changed v7.4.0

• add base_ref output by @GrantBirki in #194
• update docs by @GrantBirki in #197
• Hubot Style Deploy Locking 🔒 by @GrantBirki in #199
• Update docs by @GrantBirki in #200

Full Changelog: v7.3.1...v7.4.0

What's Changed v7.5.0

• Sticky Locks for noop deployments by @GrantBirki in #202

Full Changelog: v7.4.0...v7.5.0

v7.3.1

v7.3.1

This release fixes a bug where a deployment gets denied if reviews are disabled and no CI checks have been defined

Thanks to @redoz for finding this bug!

What's Changed

• Prechecks Fixes by @GrantBirki in #192

---

Full Changelog: v7.3.0...v7.3.1

v7.3.0

v7.3.0

This new release brings life to this Action's log output though the magic of emojis 🔮 and colors 🌈 !

Example 📸 :

This release also does the following (summarized):

• Moves from core.getInput() to core.getBooleanInput() internally (where applicable)
• Updates all internal node packages with npm update
• Further updates internal node packages via updating package pins in package.json
• Slight code performance improvements
• Slight test cleanup
• Adds new GitHub Action's output variables:
  • issue_number - The issue (PR) number which the Action was triggered on
• Adding the octokit retry plugin to additional functions

What's Changed

The pull requests that are baked into this release

• General Improvements by @GrantBirki in #188
• Additional Octokit Retries & Additional Colors 🌈 by @GrantBirki in #189
• Additional msg formatting by @GrantBirki in #190

Full Changelog: v7.2.2...v7.3.0

v7.2.2

v7.2.2

Minor and rare bug fix with .lock commands

This release fixes a rare edge case with the .lock functionality of this Action. If users --reason <message contained any part of the lockInfoFlag flags, then it would think the request is for lock details, and not to claim the actual lock. This was a wacky bug but it is now resolved by just adding a single leading whitespace character to our lockInfoFlag variable.

This release also adds enhanced debug logging for lock related functions that are internal to this Action

What's Changed

• Lock bug by @GrantBirki in #187

Full Changelog: v7.2.1...v7.2.2

v7.2.1

v7.2.1

This release fixes a minor edge case where the "post deploy" step could fail if the request for lockData returned null. Thanks to @malancas for finding this bug!

What's Changed

• Post Deploy Fixes by @GrantBirki in #183

Full Changelog: v7.2.0...v7.2.1

v7.2.0

v7.2.0

This release introduces retries for all Octokit requests. This issue was raised in #180 which pointed out that potential latency was causing the Action to fail when it was trying to add a file to a branch that didn't quite exist yet. This release adds significant improvements for resiliency when making requests to the GitHub API through Octokit retries.

Users do not need to do anything when upgrading to this release and retries will be handled for you out of the box.

Enjoy! 🚀

What's Changed

• Octokit Retries by @GrantBirki in #181

---

Full Changelog: v7.1.0...v7.2.0

v7.1.0

v7.1.0

This release introduces a new input parameter for this Action:

• deploy_message_path - The path to a markdown file which is used as a template for custom deployment messages

You can read more about this change under the dedicated documentation for custom deployment messages.

This release will give users of this Action even more flexibility to customize what their projects deployment messages look like! All this can now be done in our familiar friend, markdown and it also comes with nunjucks template rendering support 🚀.

What's Changed

• File Based Post Deploy Messages by @GrantBirki in #174
• Code cleanup by @GrantBirki in #175
• node package updates by @GrantBirki in #176
• Documentation Updates by @GrantBirki in #177
• add example screenshot by @GrantBirki in #178

Full Changelog: v7.0.0...v7.1.0

v7.0.0

v7.0.0

⚠️ Breaking Changes ⚠️

This new major release changes the way that "noop" style commands work. Starting with this release, the .deploy noop command will no longer work and it will be immediately replaced with .noop instead. The core reasoning for this, is so that .noop can become its own standalone command and no longer be a subset of the .deploy command. This will lead to less mistakes, user error, and higher flexibility when customizing this Action for future projects.

What do I need to do?

For most users, you don't need to do anything at all. Simply use .noop (the default) instead of .deploy noop. In fact, if you upgrade to v7.0.0 (this release) and run .deploy noop, you will be presented with a comment containing details about the command to use instead and a link to the deprecation docs

You can always customize the noop_trigger input option to be something else besides .noop (you could do .plan if you're doing Terraform things for example) but the "noop" trigger will never again be a subset of the primary deploy command.

Why was this change made?

Here are some suggested readings to learn more about why this change was made if you are curious to learn more:

• discussion issue
• pull request containing changes
• deprecation docs

What's Changed

• noop prefix enhancements by @GrantBirki in #169
• isDeprecated Function Fixes by @GrantBirki in #171
• node package updates by @GrantBirki in #172

Full Changelog: v6.1.0...v7.0.0