Commit Graph
10 Commits
Author SHA1 Message Date
can1357 abf8c1efc9 refactor: unslop common utilities 2026-02-22 11:01:11 +01:00
can1357 94ad716db2 wip: fixture metadata improvements 2026-02-20 12:01:30 +01:00
can1357 21c71cf755 chore(react-edit-benchmark): updated benchmark reports 2026-02-20 12:00:40 +01:00
can1357 2bdcd618ad feat: introduced hashline API redesign with structured operations and LINE#ID format
- Redesigned hashline edit API with new operation names (set, set_range, insert) and structured body parameter accepting string arrays for multiline edits.
- Changed hashline reference format from LINE:HASH to LINE#ID throughout tools and documentation for improved clarity.
- Enhanced insert operation to support optional before/after anchors enabling flexible insertion positioning and boundary echo stripping.
- Made hashline autocorrect heuristics conditional on PI_HL_AUTOCORRECT environment variable for controlled behavior.
- Added benchmark reports for claude-haiku-4-5 and GPT-5.2-Codex models demonstrating hashline edit variant performance.
2026-02-19 15:02:59 +01:00
can1357 a9fb9c9791 feat(react-edit-benchmark): resolved hashline regression with strict validation
- Added provider failure detection and exponential backoff retry logic to handle authentication and authorization errors in benchmark tasks.
- Implemented HashlineMismatchError behavior in coding-agent to fail on stale hash references instead of silently relocating edits.
- Simplified hashline validation by removing automatic line relocation logic and hash tracking infrastructure.
- Added benchmark report for claude-sonnet-4-6 model showing 85% task success rate with detailed failure analysis and performance metrics.
2026-02-19 12:52:13 +01:00
can1357 a5d26bb5f5 feat(react-edit-benchmark): migrated benchmark to directory-based fixtures with RPC resource management
- Added `--no-rules` CLI flag to coding-agent to disable rules discovery and loading.
- Added `rules` option to CreateAgentSessionOptions to allow custom rules configuration.
- Added `sessionDir` option to RpcClientOptions and implemented Symbol.dispose() for resource cleanup.
- Removed tarball-based task loading; migrated to directory-based fixtures with required inputDir and expectedDir properties.
- Refactored runner to use RpcClient resource management with `using` statement and simplified fixture handling.
- Consolidated type definitions and removed tarball.ts module in favor of streamlined task interface.
2026-02-18 05:14:19 +01:00
can1357 3c65b3b1f5 feat(react-edit-benchmark): implemented AST-based mutation system
- Replaced regex-based code mutation with Babel AST.
- Introduced a wider range of precise, structural bug types.
- Integrated code formatting into generated benchmark fixtures.
- Updated dependencies with Babel and `regexp-tree`.
2026-02-18 03:44:14 +01:00
can1357 aab245bc63 feat(edit): better hashline & benchmark prompts, re-ran codex 2026-02-15 15:38:06 +01:00
can1357 0b492d9fbb fix(react-edit-benchmark): fixed incorrect path resolution for CLI module using Bun.fileURLToPath()
- Fixed incorrect path resolution for CLI module by using Bun.fileURLToPath() to convert import.meta.resolve() result to a proper file path.
- Reordered constant declarations to ensure TMP_DIR is initialized before CLI_PATH which depends on it.
- Updated scripts/repro-stuck.ts to use consistent CLI path resolution with Bun.fileURLToPath().
2026-02-12 15:10:17 +01:00
can1357 2dc1efad24 feat(coding-agent): exported utility functions and reorganized package structure for better module access
- Exported renderPromptTemplate function from config/prompt-templates module for programmatic prompt template rendering.
- Exported computeLineHash function from patch/hashline module for external patch utility access.
- Added ./cli export path in package.json for direct CLI module access.
2026-02-12 09:40:44 +01:00