Skip to content

Docs: refresh stale README/pom notes after the PR #298 merge#299

Merged
bernardladenthin merged 1 commit into
mainfrom
claude/java-llama-cpp-features-l4tl6v
Jul 6, 2026
Merged

Docs: refresh stale README/pom notes after the PR #298 merge#299
bernardladenthin merged 1 commit into
mainfrom
claude/java-llama-cpp-features-l4tl6v

Conversation

@bernardladenthin

Copy link
Copy Markdown
Owner

Summary

  • README "Features" list: the closing bullet still described the pre-classifier era ("CUDA, Metal, and Vulkan supported via local build") with an incomplete platform list. It now states the actual default-JAR platform set (Linux x86-64/aarch64/s390x, macOS incl. Metal, Windows x86-64/x86/arm64, Android arm64/x86-64), points at the classifier table for the six GPU backends, and mentions the llama-android AAR + llama-kotlin façade. Also adds the missing headline bullets for conversation checkpoints, GGUF metadata inspection, and the router/attach server modes (their dedicated sections already existed further down).
  • README "TODO" section: claimed PIT targetClasses was "narrowed to a single class (Pair)" — it has long covered value.*/exception.*/args.* plus four json parsers (295 mutations, 100% killed, hermetic). Now states the actual scope and links TODO.md as the canonical open-items list.
  • llama/pom.xml PIT comment: removed the obsolete "audioFile mutants only killed with the audio fixture present" note — the gate is hermetic via ContentPartTest's @TempDir tests.

Documentation-only follow-up from the post-merge audit of #298; no code, build, or CI behavior changes.

Test plan

  • Affected unit / integration tests pass locally (docs-only; reuse lint clean, pom edit is a comment)
  • CI is green on this branch
  • Docs / CHANGELOG updated where applicable (this PR is the docs update)

Related issues / PRs

Follow-up to #298.

Checklist

  • I have read CONTRIBUTING.md and CODE_OF_CONDUCT.md
  • My commits follow Conventional Commits
  • No security-sensitive changes (if there are, I have notified the maintainer privately per SECURITY.md)

🤖 Generated with Claude Code

https://claude.ai/code/session_01XVMuGj2shABrHWJ9sNqLqX


Generated by Claude Code

Post-merge documentation audit found three spots that no longer matched
reality:

- README "Features" closing bullet still described the pre-classifier
  era ("CUDA, Metal, and Vulkan supported via local build") and an
  incomplete platform list. Now lists the actual default-JAR platform
  set (Linux x86-64/aarch64/s390x, macOS incl. Metal, Windows
  x86-64/x86/arm64, Android arm64/x86-64), points at the classifier
  table for the six GPU backends, and mentions the llama-android AAR +
  llama-kotlin facade. Also adds the missing headline bullets for
  conversation checkpoints, GGUF metadata inspection, and router/attach
  server modes (their dedicated sections already existed).

- README "TODO" section claimed PIT targetClasses was "narrowed to a
  single class (Pair)" — it has long covered value.*/exception.*/args.*
  plus four json parsers (295 mutations, 100% killed, hermetic). Now
  states the actual scope and links TODO.md as the canonical open list.

- llama/pom.xml PIT comment still carried the obsolete "audioFile
  mutants only killed with the audio fixture present" note; the gate is
  hermetic via ContentPartTest's @tempdir tests.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XVMuGj2shABrHWJ9sNqLqX
@bernardladenthin bernardladenthin merged commit 29d2dd2 into main Jul 6, 2026
10 of 12 checks passed
@bernardladenthin bernardladenthin deleted the claude/java-llama-cpp-features-l4tl6v branch July 6, 2026 13:11
@sonarqubecloud

sonarqubecloud Bot commented Jul 6, 2026

Copy link
Copy Markdown

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.

2 participants