Prepare the change log and create the release draft. The change log is created using labelled pull requests.
The change log can have several sections. Each section can have a set of labels used to filter pull requests.
Process for creating the change log is:
- We find the 2 commits associated with tag and prev_tag.
- We find all the commits between the 2 commits.
- We find all the pull requests that are associated with these commits.
- We group the pull requests by label.
- We sort the pull requests in each group by timestamp.
- We create each line in each section using line = line_template(pull_request).
Name | Description | Required |
---|---|---|
token | GitHub API token with push permissions to the repo. Used for creating the release draft. | true |
tag | An existing tag to use for the release draft. | true |
prev_tag | The previous tag to start calculating the change log from. | true |
owner | Owner of the repo to create the release draft for. Default is the owner of the current repo. | false |
repo | Repo to create the release draft for. Default is the current repo. | false |
title | The title for the release draft. Default is 'Release ' + tag. | false |
title_prefix | title = prefix + tag. If both title and title_prefix are specified then title takes priority. | false |
header | The header will be placed before the change log in the release body. Default empty string. | false |
footer | The footer will be placed after the change log in the release body. Default empty string. | false |
no_changes_message | This message will be placed in the release body when the change log is empty. Default is "No changes from the previous release." | false |
draft | Set to false to publish this release draft. Default is true. | false |
prerelease | Set to true if this is a prerelease. By default it parses the tag as a semantic version to see if it is a prerelease. | false |
config | Path to a configuration file. By default no config is used. The settings specified here override those in the config. | false |
The config file is just a .js
javascript file containing the same settings as above and some extras. Example:
module.exports = {
title_prefix: "My App ",
// valid PR types: ['feat', 'fix', 'docs', 'style', 'refactor', 'perf', 'test', 'build', 'ci', 'chore', 'revert']
sections:
[
{ title: "π Features", labels: ["enhancement", "feat", "perf"] },
{ title: "π Bug Fixes", labels: ["bug", "fix", "revert"] },
{ title: "π§Ή Maintenance", labels: ["docs", "style", "refactor", "test", "build", "ci", "chore"] },
],
header: `For more documentation and support please visit https://konveyor.io/move2kube/
# Changelog`,
line_template: x => `- ${x.title} [#${x.number}](${x.html_url})`,
}
For more documentation and support please visit https://konveyor.io/move2kube/
# Changelog
## π Features
- feat: Move fixers to a package and simplify convert logic [#404](https://github.com/konveyor/move2kube/pull/404)
- feat: order containerization options in terms of container build type priority [#422](https://github.com/konveyor/move2kube/pull/422)
- feat: support * as a match all selector in qa keys [#415](https://github.com/konveyor/move2kube/pull/415)
## π Bug Fixes
- fix: parsing of config strings [#416](https://github.com/konveyor/move2kube/pull/416)
## π§Ή Maintenance
- chore: allow install script to install any tag [#424](https://github.com/konveyor/move2kube/pull/424)
- chore: get the new index.yaml from the latest release [#420](https://github.com/konveyor/move2kube/pull/420)
- chore: combine all 3 steps of asking a question into a single function call [#418](https://github.com/konveyor/move2kube/pull/418)
- chore: octokit is not defined in github workflows, use github instead [#413](https://github.com/konveyor/move2kube/pull/413)