Conversation
Introduce OpenAIProviderManager plus JSON-backed metadata to hydrate /models payloads for OpenAI-like providers such as Synthetic. Hook ModelInfoManager, Model, and CLI completions/listings into that registry, expose configuration data in aider/resources/openai_providers.json, and ensure LiteLLM is initialized with the custom handler so cecli can call these endpoints reliably.
Some OpenAI-compatible providers emit costs like '/bin/bash.00000055', which our float parser treated as invalid and left the UI without per-token pricing (e.g., synthetic MiniMax-M2). Strip currency symbols/commas before parsing and add a regression test that proves static model caches with dollar-prefixed pricing still populate ModelInfoManager.
Synthetic/other OpenAI-like providers returned both reasoning_content and content, but our consolidation skipped storing the final content whenever reasoning existed, so cecli printed only the THINKING section. Always capture the message.content (including list-style OpenAI blocks) and add a regression test that feeds a recorded MiniMax completion via a heredoc JSON snippet to assert both THINKING and ANSWER text render.
Reinstalled the dev toolchain, ran the documented pre-commit hooks, and applied the resulting isort/black fixes across provider modules plus removed an unused import/variable in tests so the branch now passes the project’s formatting gate.
Co-authored-by: aider-ce (synthetic/hf:deepseek-ai/DeepSeek-V3.2)
Added test_main_smoke.py with two smoke tests: - test_main_async_executes: Tests async entry point - test_main_executes: Tests sync entry point These tests expose a bug where Path object is passed to generate_search_path_list() which expects a string, causing AttributeError when trying to call lstrip() on Path object.
Convert 76 async test methods to synchronous by: - Removing 'async def' -> 'def' for test methods - Removing 'await' from main() calls (main() handles asyncio.run internally) 3 tests remain async (marked with TODO comments) as they call async helper functions (setup_git, check_gitignore, coder.allowed_to_edit) and will need further refactoring. This fixes the issue where async tests were never executing because unittest.TestCase doesn't support async test methods.
The --yes flag is ambiguous as it matches both --yes-always and --yes-always-commands. Update all test cases to use the explicit --yes-always flag.
…ibute Add autospec=True to 6 InputOutput mock tests to handle async methods correctly. Configure 'pretty' attribute since autospec requires instance attributes to be set manually. Fixes 6 tests, bringing suite to 63 passed / 17 failed.
The create_env_file method was marked as async but only performs synchronous file operations. This caused RuntimeWarning about unawaited coroutines in tests that call it. Fixes 2 tests, bringing suite to 65 passed / 15 failed.
Add mock_autosave_future() helper that returns AsyncMock()() to create an awaitable coroutine for _autosave_future attribute. Apply fix to all tests that patch Coder.create to avoid "object AsyncMock can't be used in 'await' expression" errors. For tests calling main() multiple times, create fresh coroutine before each call since coroutines can only be awaited once. Fixes 9 tests. Remaining 5 failures are unrelated to mock issues.
Configure mock_io.confirm_ask as AsyncMock to handle async confirm_ask calls in check_gitignore. Fixes TypeError: object MagicMock can't be used in 'await' expression.
Configure confirm_ask as AsyncMock on InputOutput mock return_value.
Commands.cmd_lint no longer exists, similar to cmd_add. Mark as xfail for future refactoring.
Update test_return_coder and test_main_exit_with_git_command_not_found to expect exit code 0 (success) instead of None. This aligns with the graceful_exit() wrapper introduced in commit 7f813af which changed main() to always return integer exit codes for proper shell integration. Fixes final 2 test failures.
Windows uses USERPROFILE environment variable for Path.home() instead of HOME. Use platform.system() to conditionally set the appropriate environment variable for the platform.
- Convert test_setup_git from async def to def - Convert test_check_gitignore from async def to def - Convert test_argv_file_respects_git from async def to def - Wrap async function calls with asyncio.run() - Add asyncio import - Remove @pytest.mark.skip decorators All three tests now execute and pass (79 passed, 0 skipped, 2 xfailed)
…chitecture - Remove @pytest.mark.xfail decorator - Replace coder.commands.cmd_add() with asyncio.run(coder.commands.do_run()) - Add try/except SwitchCoder blocks (AddCommand raises this to signal refresh) - Remove patch.object for confirm_ask (--yes-always handles this) - Add SwitchCoder import The add command was refactored from method-based to class-based in commit 82ba897. The test now uses the new Commands.do_run() API to execute the AddCommand.execute() method through the CommandRegistry. Test verifies gitignore behavior: - Default: gitignored files rejected - --add-gitignore-files: gitignored files accepted - --no-add-gitignore-files: gitignored files rejected
- Update LintCommand to read CLI files from system_args instead of kwargs - Add glob pattern expansion support for --lint flag - Move expand_glob_patterns from main.py to utils.py to avoid circular imports - Add tests for explicit files and glob patterns with --lint This fixes the --lint flag after the command system refactoring (82ba897) that moved commands to class-based architecture but didn't update callers. Tests: - test_lint_option: verifies --lint with no files (dirty file fallback) - test_lint_option_with_explicit_files: verifies --lint file1.py file2.py - test_lint_option_with_glob_pattern: verifies --lint *.py All 83 tests pass.
CLI files were being resolved to absolute paths twice: 1. At line 26: str(Path(f).resolve()) 2. At line 43: coder.abs_root_path(fname) Now all file paths (CLI, in-chat, dirty) are resolved uniformly at line 43.
The check 'if not fnames and coder.repo' is redundant because we already checked 'if not coder.repo' at line 27 and returned early. At line 35, coder.repo is guaranteed to be truthy.
The 'root' parameter was defined but never used in the function body. Verified that all 3 call sites (2 in main.py, 1 in lint.py) call it without this parameter.
- Remove unused Path import from lint.py - Remove unused glob import from main.py - Remove unused pytest import from test_main.py - Apply black formatting All pre-commit hooks now pass.
add an example docker setup for local nvidia
Fix testing related to the main module
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Addresses:
#330
Includes: