## Summary
Fixes an issue where `thinking` content blocks in Claude Code responses
were not being parsed, resulting in empty `AssistantMessage` content
arrays.
## Changes
- Added `ThinkingBlock` dataclass to handle thinking content with
`thinking` and `signature` fields
- Updated client parsing logic in `_internal/client.py` to recognize and
create `ThinkingBlock` instances
- Added comprehensive test coverage for thinking block functionality
## Before
```python
# Claude Code response with thinking block resulted in:
AssistantMessage(content=[]) # Empty content!
```
## After
```python
# Now correctly parses to:
AssistantMessage(content=[
ThinkingBlock(thinking="...", signature="...")
])
```
Fixes#27
---------
Co-authored-by: Dickson Tsai <dickson@anthropic.com>
`claude`'s `--mcp-config` takes either a JSON file or string, so support
this in `ClaudeCodeOptions`
```
--mcp-config <file or string> Load MCP servers from a JSON file or string
```
---------
Co-authored-by: claude[bot] <209825114+claude[bot]@users.noreply.github.com>
Co-authored-by: Sam Fu <shunfu@users.noreply.github.com>
## Summary
- Adds `extra_args` field to `ClaudeCodeOptions` to support passing
arbitrary CLI flags
- Enables forward compatibility with future CLI flags without requiring
SDK updates
- Supports both valued flags (`--flag value`) and boolean flags
(`--flag`)
## Changes
- Add `extra_args: dict[str, str | None]` field to `ClaudeCodeOptions`
- Implement logic in `SubprocessCLITransport` to handle extra args:
- `None` values create boolean flags (e.g., `{"verbose": None}` →
`--verbose`)
- String values create flags with arguments (e.g., `{"output": "json"}`
→ `--output json`)
- Add comprehensive tests for the new functionality
## Test plan
- [x] Added unit tests for settings file path handling
- [x] Added unit tests for settings JSON object handling
- [x] Added unit tests for extra_args with both valued and boolean flags
- [x] All tests pass (`python -m pytest tests/`)
- [x] Type checking passes (`python -m mypy src/`)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
The test was checking source code for specific strings rather than
testing behavior. The existing subprocess tests already verify the
functionality works correctly.
Remove anyio.create_task_group() from receive_messages() to fix the
RuntimeError "Attempted to exit cancel scope in a different task than
it was entered in" when using the SDK with FastAPI's SSE streaming.
The issue occurred because FastAPI can move async generators between
different asyncio tasks during the streaming lifecycle, which conflicts
with anyio's cancel scope tracking.
Changes:
- Remove task group usage from receive_messages()
- Read stderr sequentially after stdout completes
- Add test to ensure no task groups are used
- Fix existing test expectation for buffer overflow
This is a minimal fix that maintains compatibility while solving the
core issue. The trade-off is that stderr is read after stdout instead
of concurrently, but this is unlikely to cause issues in practice.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Removed redundant comments that simply restate what the code is doing.
Kept only comments that provide valuable context or explain complex behavior.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Adds buffering to accumulate incomplete JSON messages that are split
across multiple stream reads due to asyncio's 64KB default buffer limit.
- Implement 1MB buffer to accumulate partial JSON
- Clear buffer on successful parse or size limit exceeded
- Add comprehensive tests for split JSON scenarios
Fixes CLIJSONDecodeError when reading large minified JSON files.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fixed quotes to use double quotes consistently
- Adjusted line breaks per ruff formatter rules
- No functional changes
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fixed formatting in test_subprocess_buffering.py
- No functional changes, only code style improvements
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Convert async test to use anyio.run() pattern consistent with other tests
- Remove unused CLIJSONDecodeError import
- Add tests for JSON with embedded newlines
- Add tests for multiple newlines between objects
- All tests passing
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Remove cost_usd field from test expectations
- Update to use total_cost_usd instead of total_cost
- All 30 tests now pass
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
