Skip to content

feat!: path-based change detection, Java 25 default, speed & debug tu…#3

Merged
TheMeinerLP merged 4 commits into
mainfrom
feat/path-filter-and-speed-tuning
May 11, 2026
Merged

feat!: path-based change detection, Java 25 default, speed & debug tu…#3
TheMeinerLP merged 4 commits into
mainfrom
feat/path-filter-and-speed-tuning

Conversation

@TheMeinerLP
Copy link
Copy Markdown
Contributor

…ning

  • gradle-build-pr: skip build when no Gradle-relevant files changed
    (dorny/paths-filter); add paths-filters and force-build inputs
  • Default Java 25 (was 21)
  • Default runner reduced to ubuntu-latest only (multi-OS via runs-on)
  • Drop clean from default Gradle task to preserve incremental caches
  • Cache-read-only on non-default branches to keep caches lean
  • Concurrency group cancels superseded PR runs
  • Auto-enable --info --stacktrace when re-run with debug logging
    (runner.debug == 1)
  • Upload test reports as artifact on failure (configurable)
  • gradle-publish: same Java 25 default, debug-aware verbosity, publish
    log artifact on failure
  • README updated with new defaults and force-build / paths-filters

BREAKING CHANGE: default java-version changed from 21 to 25,
default runs-on changed from triple-OS matrix to ["ubuntu-latest"],
default gradle-task changed from clean build test to build test.## Overview

Fixes #{issue number}

Description

Submitter Checklist

  • Make sure you are opening from a topic branch (/feature/fix/docs/ branch (right side)) and not your main branch.
  • Ensure that the pull request title represents the desired changelog entry.
  • New public fields and methods are annotated with @since TODO.
  • I read and followed the contribution guidelines.

…ning

- gradle-build-pr: skip build when no Gradle-relevant files changed
  (dorny/paths-filter); add `paths-filters` and `force-build` inputs
- Default Java 25 (was 21)
- Default runner reduced to ubuntu-latest only (multi-OS via `runs-on`)
- Drop `clean` from default Gradle task to preserve incremental caches
- Cache-read-only on non-default branches to keep caches lean
- Concurrency group cancels superseded PR runs
- Auto-enable `--info --stacktrace` when re-run with debug logging
  (`runner.debug == 1`)
- Upload test reports as artifact on failure (configurable)
- gradle-publish: same Java 25 default, debug-aware verbosity, publish
  log artifact on failure
- README updated with new defaults and force-build / paths-filters

BREAKING CHANGE: default `java-version` changed from `21` to `25`,
default `runs-on` changed from triple-OS matrix to `["ubuntu-latest"]`,
default `gradle-task` changed from `clean build test` to `build test`.
…nfig

- gradle-build-pr: restore three-OS matrix default (ubuntu, windows,
  macos) since PRs should always run cross-platform
- Always upload test reports (drop upload-reports input); artifacts are
  attached on success and failure for full traceability
- New `test-report` job aggregates JUnit XML across matrix jobs and
  publishes a unified check + PR comment via
  EnricoMi/publish-unit-test-result-action
- gradle-publish: always upload reports (drop upload-reports input)
- Add renovate.json so this repo (and downstream consumers as template)
  keeps Action versions current; github-actions manager covers reusable
  workflow refs automatically
- Add `.github/workflows/markdown-lint.yml` exposing markdownlint-cli2
  and lychee as a single workflow_call entry. Honours path filter,
  concurrency cancel, force-toggle and repository-owner skip — same
  conventions as the Gradle workflows.
- Self-lint via `.github/workflows/lint-docs.yml` runs on PRs that
  touch markdown.
- Add `.markdownlint.json` and `.lycheeignore` with sensible defaults.
@TheMeinerLP TheMeinerLP merged commit 5400a36 into main May 11, 2026
1 check passed
@TheMeinerLP TheMeinerLP deleted the feat/path-filter-and-speed-tuning branch May 11, 2026 14:35
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant