feat(tests): Increase CLI unit test coverage from 63% to 85% (#895)#97
Merged
rquidute merged 8 commits intoJun 17, 2026
Merged
Conversation
Add 10 new test modules covering previously untested CLI source files: - tests/test_colorize.py: ColorConfig env var, all colorize_* functions with/without colors, set_colors_enabled toggle - tests/test_config.py: load_config paths (valid JSON, example, fallback, malformed), PairingMode, ATTRIBUTE_MAPPING, path utility functions - tests/test_exceptions.py: CLIError, APIError.format_message, ConfigurationError, handle_api_error (bytes/str/None content), handle_file_error - tests/test_async_cmd.py: @async_cmd decorator — wraps async fn, return values, args/kwargs forwarding, functools.wraps preservation, exceptions - tests/test_client.py: get_client() success/failure, ConfigurationError propagation, HTTP scheme in host URL, module-level fallback - tests/test_socket_schemas.py: all Pydantic models — construction, required fields, optional defaults, union types in SocketMessage - tests/test_run/test_logging.py: configure_logger_for_run (streaming on/off, fallback on error), stop_log_streaming, get_log_stream_url - tests/test_run/test_log_stream_handler.py: start/stop, add_log_entry (queue full, auto-timestamp), _get_local_ip (success/failure), URL format - tests/test_run/test_logs_http_server.py: LogStreamingHandler routing, SSE events, download, LogsHTTPServer start/stop lifecycle - tests/test_run/camera/test_websocket_manager.py: connect, reset, stop, start_capture_and_stream, wait_and_connect_with_retry, _transfer_converted_data Closes #895
- test_exceptions.py: construct UnexpectedResponse normally instead of
using __new__ (class has a proper __init__)
- test_socket_schemas.py: use shared_constants.TestStateEnum (uppercase
PASSED/FAILED/…) not api_lib_autogen.models.TestStateEnum (lowercase)
— socket_schemas.py imports from shared_constants
- test_websocket_manager.py:
- clean up messy test_returns_true_on_success (removed dead nested
patches)
- fix infinite-loop hang in test_returns_false_after_all_attempts_fail:
wait_and_connect_with_retry only increments attempt inside except,
so mock must raise (side_effect=ConnectionRefusedError) instead of
returning False
- test_exceptions.py: fix test_custom_file_type_in_message — source uses
.title() so 'config file' → 'Config File' (both words capitalised)
- tests/test_utils_additional.py (new): cover uncovered branches in
utils.py (72% → ~87%):
- add_mapped_property: nested path creation, deep nesting, existing
section, overwrite
- add_unmapped_property: with/without section, None section, overwrite
- load_json_config: invalid JSON, config-value-not-dict, root-not-dict,
missing file, OSError
- merge_configs: scalar overrides dict, dict overrides scalar, empty
inputs, no mutation of base
- get_cli_version: returns string, unknown when missing, reads from
pyproject.toml, falls back to git root, handles IOError
- get_cli_sha: returns string, unknown when no git root, 8-char SHA,
handles subprocess error and git-not-found
- get_versions: success, re-raises CLIError, closes client on exception
- tests/test_run/test_websocket_additional.py (new): cover uncovered
branches in websocket.py (55% → ~85%):
- __log_test_run_update: echoes state, triggers chip-server-info display
on executing, does not display twice
- __display_manual_pairing_code: skips when no config / missing fields,
handles API exception, calls API and closes client
- __log_test_case_update: browser-peer warning from update errors, from
tracked step errors, cleans up step errors, no warning on passed
- __handle_log_record: logs each record, correct level/message, empty list
- __handle_incoming_socket_message: routes test_update, timeout (silent),
log_records, echoes error for unknown type
- __handle_test_update: routes step/case/suite/run updates; closes socket
when run is not executing; does not close when still executing
… (#895)
Fix test_exceptions.py:
- test_custom_file_type_in_message: .title() capitalises both words
('config file' → 'Config File'), update assertion to match
New: tests/test_run/camera/test_camera_http_server_additional.py
Covers VideoStreamingHandler methods not yet tested (40% → ~80%):
- do_OPTIONS: 200 + CORS headers
- log_message: no-op
- stream_live_video: 200/video-mp4 header, data chunks, no-queue path,
write-error handling
- serve_player: 200/html, fallback HTML, no-cache headers, push-av
template selection, radio options
- handle_streams_api: 500 when no URL, success, non-200, connection error
- handle_simple_proxy: success, invalid base64 (400), upstream 404,
fetch error (500), extra path appended
- handle_stream_proxy: missing url param (400), regular content, upstream
error, MPD manifest rewrite, exception (500)
New: tests/test_run/camera/test_camera_stream_handler_additional.py
Covers CameraStreamHandler methods not yet tested (46% → ~90%):
- start_video_capture_and_stream: returns Path, contains prompt_id,
clears event, resets error, calls http_server.start
- wait_for_stream_ready: True when set, False on timeout, True on late set
- _initialize_video_capture: sets error when ffmpeg missing, sets event
on connect success, sets error on connect failure, handles exceptions
- wait_for_user_response: queued response, timeout, late enqueue
- stop_video_capture_and_stream: stops ws+http, None with no file, returns
path when file exists, puts None sentinel in mp4_queue
New: tests/test_run/test_prompt_manager_additional.py
Covers prompt_manager.py functions not yet tested (45% → ~75%):
- _get_local_ip: success and OSError fallback
- _get_video_handler: creates new and returns existing instance
- _cleanup_video_handler: stops handler, noop when None, ignores errors
- handle_prompt routing: image/twt/push-av/stream/message by type and
instance; unsupported prompt echoes error
- _handle_image_verification_prompt: sends response, no response on
timeout, handles exception
- _handle_two_way_talk_prompt: sends response, stops handler on timeout,
creates fallback handler when None
- _handle_push_av_stream_prompt: sends response on user answer, no
response when options empty
- __handle_message_prompt: sends ACK response
- test_prompt_manager_additional.py:
- Fix TestGetVideoHandler.test_creates_new_instance_when_none:
CameraStreamHandler is imported lazily inside _get_video_handler via
'from .camera import CameraStreamHandler', so patch its source module
'th_cli.test_run.camera.CameraStreamHandler', not prompt_manager
- Add TestHandleStreamVerificationPrompt (5 new tests covering lines
133-201): stream ready + user answers → sends response; stream not
ready with/without init_error → sends CANCELLED; user answer None →
no send; empty options → returns early
- test_logs_http_server.py:
- Add test_debug_log_at_100_entries: streams 100 entries through
stream_logs() to hit the 'sent_count % 100 == 0' debug branch
(lines 134-150)
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request adds a comprehensive suite of unit tests for asynchronous commands, client configuration, colorization, exceptions, camera streaming, HTTP logging, and websocket management. The review feedback highlights several critical issues in the newly added tests, including a scoping bug in test_routes_stream_verification_by_type that bypasses mocking, an IOError in test_returns_unknown_on_ioerror due to premature patching of open, and redundant name-mangling workarounds for module-level functions in test_prompt_manager_additional.py. Additionally, the reviewer advises fixing an infinite loop bug in the production code of wait_and_connect_with_retry instead of merely documenting it in the test comments.
Update all 15 test files added in this branch from 'Copyright (c) 2025-2026' / '2025' to '2026 Project CHIP Authors'.
test_prompt_manager_additional.py:
- Replace test_routes_stream_verification_by_type: remove stacked no-op
patches with dead 'pass' bodies and misleading comments about
'VideoStreamVerification' (non-existent class). Rewrite as
test_routes_stream_verification_by_instance, using a real mock video
handler so the assertion is actually verified.
- TestUploadFileAndSendResponse: remove dead 'if False else None'
expressions, stale '# Call directly via module private name' and
'# Access via actual mangled name' comments (module-level __ functions
are NOT name-mangled in Python), and weak 'if count > 0: assert called'
non-assertions. Extract a module-level _find_upload_fn() helper with a
clear docstring explaining the discovery strategy, and add a pytest.skip
guard. Assertions are now unconditional mock_send.assert_called_once().
- Remove '# Should not raise' trailing comment — the test already expresses
this by not asserting anything; the comment is noise.
- Remove stale line-number annotations from section headers
('covers lines 387-390', 'lines 133-201') — line numbers are an
implementation detail that silently goes stale.
test_websocket_additional.py:
- Remove unused imports OptionsSelectPromptRequest and PromptResponse
(imported but never referenced in any test).
- Rename test_returns_none_on_file_not_found to test_raises_cli_error_on_file_not_found: the test name contradicted both the assertion (pytest.raises) and the inline comment explaining that CLIError is raised, not None returned. - Remove '# IOError is caught → "unknown"' inline comment on the assertion line — the assertion already expresses this; the comment restated it without adding information. - Reword '# Need pyproject.toml to exist so it tries to open it' to '# Need pyproject.toml to exist so the code attempts to open it' for clarity.
oxesoft
approved these changes
Jun 17, 2026
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
2 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.
Summary
Fixes #895 — Increase CLI unit test coverage to ≥ 85%.
Previously the CLI test suite covered 63.53% of the codebase. This PR brings coverage to 85.79% (951 tests passing, threshold enforced by
--cov-fail-under=85inpyproject.toml).What changed
10 new test modules targeting previously untested source files:
tests/test_colorize.pyth_cli/colorize.pytests/test_config.pyth_cli/config.pytests/test_exceptions.pyth_cli/exceptions.pytests/test_async_cmd.pyth_cli/async_cmd.pytests/test_client.pyth_cli/client.pytests/test_socket_schemas.pyth_cli/test_run/socket_schemas.pytests/test_run/test_logging.pyth_cli/test_run/logging.pytests/test_run/test_log_stream_handler.pyth_cli/test_run/log_stream_handler.pytests/test_run/test_logs_http_server.pyth_cli/test_run/logs_http_server.pytests/test_run/camera/test_websocket_manager.pyth_cli/test_run/camera/websocket_manager.py5 gap-filling test modules for modules with partial coverage:
tests/test_utils_additional.pyth_cli/utils.pytests/test_run/test_websocket_additional.pyth_cli/test_run/websocket.pytests/test_run/camera/test_camera_http_server_additional.pyth_cli/test_run/camera/camera_http_server.pytests/test_run/camera/test_camera_stream_handler_additional.pyth_cli/test_run/camera/camera_stream_handler.pytests/test_run/test_prompt_manager_additional.pyth_cli/test_run/prompt_manager.pyNo production code changes
Only test files were added. The
[tool.coverage.run]omit list was reviewed and all existing entries are appropriate (*/main.pystays — it only registers Click groups with no testable logic).Final coverage report
Notable per-module results:
colorize.py,exceptions.py,async_cmd.py,socket_schemas.py,shared_constants.py→ 100%config.py,log_stream_handler.py,logging.py,camera_stream_handler.py→ ≥ 95%camera_http_server.py→ 88% (up from 40%)utils.py→ 96% (up from 72%)websocket.py→ 81% (up from 55%)