feat(code-reviewer): Implement async calls and tests#13

Open

hamiltop wants to merge 108 commits intorelease/stable-with-patchesfrom

feat/code-reviewer-impl

Commits on Apr 25, 2025

fix(resolver): Prioritize env var for runtime image and fix workflow syntax
openhands-agent
committed
feat: Implement code reviewer core logic and tests
openhands-agent
committed

Commits on Apr 26, 2025

feat(code-reviewer): Refactor async calls, add tests, fix errors
openhands-agent
committed
fix(workflow): Remove comment inside multi-line if condition
openhands-agent
committed
fix(workflow): Install openhands from current branch
openhands-agent
committed
Revert "fix(workflow): Install openhands from current branch"
openhands-agent
committed
fix(workflow): Install openhands from feat/code-reviewer-impl branch
openhands-agent
committed
fix: Handle SecretStr token and pass output file path
openhands-agent
committed
fix: Ensure output dir exists early and add job permissions
openhands-agent
committed
fix: Use correct inputs for App token permissions
openhands-agent
committed
feat: Add debug logging for GraphQL request
openhands-agent
committed
fix: Set log level to DEBUG in review_pr
openhands-agent
committed
fix: Force logger reconfiguration to DEBUG
openhands-agent
committed
fix: Log response body on HTTPStatusError
openhands-agent
committed
chore: Remove debug logging for token tracing
openhands-agent
committed
fix(workflow): Run review script from workspace and set output dir
openhands-agent
committed
fix(workflow): Clean up /tmp/repo before cloning
openhands-agent
committed
fix(workflow): Use GITHUB_WORKSPACE instead of cloning repo in script
openhands-agent
committed
fix(workflow): Ensure SANDBOX_BASE_CONTAINER_IMAGE has fallback value
openhands-agent
committed
fix(workflow): Prioritize input then env var for SANDBOX_BASE_CONTAINER_IMAGE
openhands-agent
committed
fix(workflow): Pass SANDBOX_BASE_CONTAINER_IMAGE to review step env
openhands-agent
committed
refactor(workflow): Use step outputs for SANDBOX_BASE_CONTAINER_IMAGE
openhands-agent
committed
fix(workflow): Correctly check workflow env var in set_vars step
openhands-agent
committed
chore(workflow): Add step to debug base image variable values
openhands-agent
committed
fix(workflow): Directly use vars context for base image fallback
openhands-agent
committed
fix(review): Use stricter check before closing runtime
openhands-agent
committed
chore(workflow): Add step to dump runtime Docker logs
openhands-agent
committed
fix(review): Catch TypeError during runtime cleanup
openhands-agent
committed
fix(review): Fix post-comment args and add loop logging
openhands-agent
committed
fix(review): Remove fake_user_response_fn for non-interactive agent
openhands-agent
committed
fix(review): Treat AWAITING_USER_INPUT as success if review generated
openhands-agent
committed
fix(review): Temporarily disable MAIN event subscribers during agent run
openhands-agent
committed
refactor(review): Align review_pr structure with resolve_issue
openhands-agent
committed
refactor: Align review_pr and resolve_issue structure
openhands-agent
committed
Refactor: Simplify event stream handling in review_pr.py
openhands-agent
committed
Fix: Add missing parse_pr_url method to GithubPRHandler
openhands-agent
committed
Fix: Use get_converted_issues instead of get_issue_info in review_pr.py
openhands-agent
committed
fix(review): Implement checkout_pr and fix JSON serialization
openhands-agent
committed
fix(review): Improve JSON serialization robustness
openhands-agent
committed
fix(review): Correct prompt path and instruct agent to finish
openhands-agent
committed
fix(review): Handle AgentFinishAction and Issue serialization
openhands-agent
committed
fix(review): Update prompt to prevent markdown in JSON output
openhands-agent
committed
feat(code-reviewer): Modify prompt and parsing for git diff and finish action
openhands-agent
committed
fix(code-reviewer): Use dict access for issue in prompt
openhands-agent
committed
fix(code-reviewer): Improve error serialization and add logging for issue data
openhands-agent
committed
fix(code-review): Fix type errors and pre-commit issues
openhands-agent
committed
fix(code-review): Add git fetch instruction to prompt
openhands-agent
committed
fix(code-review): Reinforce JSON output format in prompt
openhands-agent
committed
fix(code-review): Parse review JSON from final_thought instead of message
openhands-agent
committed
fix(code-review): Remove await from synchronous runtime.close() call
openhands-agent
committed
fix(code-review): Use json_default serializer for ReviewerOutput
openhands-agent
committed
fix(reviewer): Write output to file and fix workflow args
openhands-agent
committed
fix(ci): Pass --output-file arg to review_pr.py
openhands-agent
committed
fix(reviewer): Handle escaped JSON in final_thought and multi-line output file
openhands-agent
committed
feat: add debug logging for github token in post_review
openhands-agent
committed
fix: correct type hint for token in GithubPRHandler
openhands-agent
committed
fix: use get_secret_value() for Authorization header
openhands-agent
committed
Revert "fix: use get_secret_value() for Authorization header"
openhands-agent
committed
Revert "fix: correct type hint for token in GithubPRHandler"
openhands-agent
committed
refactor: remove SecretStr usage for token in post_review_comments
openhands-agent
committed
chore: remove debug logging for token auth
openhands-agent
committed
refactor: remove redundant check in post_comments
openhands-agent
committed
refactor(code-reviewer): Remove GitLab support and fix pre-commit errors
openhands-agent
committed
feat(code-reviewer): Set default review level=line, depth=deep
openhands-agent
committed
fix(workflow): Remove explicit review level/depth args to use defaults
openhands-agent
committed
fix(workflow): Set default review level/depth in workflow inputs
openhands-agent
committed
fix(workflow): Correct fallbacks for REVIEW_LEVEL/DEPTH env vars
openhands-agent
committed
fix(workflow): Restore explicit args for review level/depth
openhands-agent
committed
feat(code-review): Emphasize adherence to review level/depth in prompt
openhands-agent
committed
refactor(code-review): Remove argparse defaults for level/depth
openhands-agent
committed
fix(code-review): Pass workflow inputs directly to review_pr.py args
openhands-agent
committed
fix(workflow): Add missing line continuation for review_pr args
openhands-agent
committed
fix(code-review): Instruct agent to use AgentFinishAction.final_thought
openhands-agent
committed
fix(code-review): Refine prompt for final_thought and truncate cmd logs
openhands-agent
committed
fix(code-review): Correct prompt to use finish tool message arg
openhands-agent
committed
fix(code-review): Remove fake_user_response_fn for non-interactive run
openhands-agent
committed
fix(code-review): Log full thoughts and finish action message
openhands-agent
committed
fix(code-review): Import AgentThinkAction for logging
openhands-agent
committed
refactor(code-review): Revert custom logging for thoughts/finish actions
openhands-agent
committed
feat(code-review): Add detailed instructions for git diff parsing to prompt
openhands-agent
committed
feat(code-review): Require agent to verify line numbers against checked-out files
openhands-agent
committed
feat(code-review): Improve line verification logic in prompt
openhands-agent
committed
fix(review): Refactor review_pr.py structure and fix pre-commit errors
openhands-agent
committed
feat(review): Update prompt to improve line number accuracy and prevent hanging
openhands-agent
committed
feat(review): Add --llm-num-retries argument
openhands-agent
committed
chore: Upgrade litellm to 1.67.2
openhands-agent
committed
fix(codeact): Set wait_for_response=False for non-tool-call messages
openhands-agent
committed
ci(review): Pass --llm-num-retries to review_pr script
openhands-agent
committed
Revert "fix(codeact): Set wait_for_response=False for non-tool-call messages"
openhands-agent
committed
fix: Use fake_user_response_fn to handle AWAITING_USER_INPUT
openhands-agent
committed
fix(ci): Add missing line continuation in review_pr call
openhands-agent
committed
feat: Add --llm-temperature argument and set to 2.0 in CI
openhands-agent
committed
feat: Implement conditional temperature retry for code review
openhands-agent
committed
Revert "feat: Implement conditional temperature retry for code review"
openhands-agent
committed
Revert "Revert "feat: Implement conditional temperature retry for code review""
openhands-agent
committed
feat: Increase temperature to 2.0 on LLMNoResponseError retry
openhands-agent
committed
fix: Refine AWAITING_USER_INPUT handler message
openhands-agent
committed
docs(prompt): Remove diff-based line number calculation instructions
openhands-agent
committed
docs(prompt): Explicitly forbid using diff for line numbers
openhands-agent
committed
Fix: Remove unicode_escape decoding for agent final_thought JSON parsing
openhands-agent
committed
Feat: Update code review prompt to avoid duplicates and ensure actionable feedback
openhands-agent
committed
Fix: Parse finish message instead of thought for review JSON
openhands-agent
committed
Revert "Fix: Parse finish message instead of thought for review JSON"
openhands-agent
committed
Fix: Add logging and strip() for parsing final_thought JSON
openhands-agent
committed

feat(code-reviewer): Implement async calls and tests#13hamiltop wants to merge 108 commits intorelease/stable-with-patchesremind101/OpenHands:release/stable-with-patchesfrom feat/code-reviewer-implremind101/OpenHands:feat/code-reviewer-implCopy head branch name to clipboard

Commits

Commits on Apr 25, 2025

Commits on Apr 26, 2025

Commits on Apr 27, 2025

feat(code-reviewer): Implement async calls and tests#13
hamiltop wants to merge 108 commits intorelease/stable-with-patchesfrom
feat/code-reviewer-impl