logging: add logging mechanism, hooks to log uncaught crashes, and an example use of the logging tools - #1
Merged
Merged
Conversation
…e top level check
hongquanli
added a commit
that referenced
this pull request
Jan 28, 2026
Address code review issues #1 and #2: 1. Add port index validation (0-15) to Microcontroller methods: - set_port_intensity, turn_on_port, turn_off_port, set_port_illumination - Raises ValueError for out-of-range ports, TypeError for non-integers 2. Add wait_till_operation_is_completed() to IlluminationController.set_port_intensity() - Makes behavior consistent with turn_on_port() and turn_off_port() - Prevents race conditions when setting intensity before turning on Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
hongquanli
added a commit
that referenced
this pull request
Mar 24, 2026
- Use SCREW_PITCH_W_MM and config bounds for step_size calculation instead of hardcoded 1.0/8 (comment #1) - Add assert not t.is_alive() after join(timeout=...) to fail deterministically on deadlock (comment #2) - Assert call_order sequence is non-interleaving, accounting for the offset move within _home_wheel_unlocked (comment #3) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Alpaca233
added a commit
that referenced
this pull request
Apr 27, 2026
…hanges Bug: clicking Start Acquisition a second time made the plate view show wells at a different grid position than before. Root cause: the channel layer's canvas was allocated by _create_channel_layer on the first tile of run #1 and reused as-is for run #2, but a new run can produce a different well_slot_shape (max-extent over scanned wells) or scan a different subset of wells — both shift where new tiles compute their slot offsets, while the canvas geometry from run #1 has not moved. Fix: setPlateLayout now wipes existing channel canvases to fresh zero-filled plate-sized arrays whenever the new slot dimensions don't match the current canvas, OR the set of wells in the new run differs from the previous one. For re-runs of the *same* wells with the *same* extent, the canvas is preserved (existing tiles stay visible until overwritten by new ones). Plumbing changes: - PlateViewInit gains a `well_ids: List[str]` field. - multi_point_worker._emit_plate_layout populates it from scan_region_fov_coords_mm. - gui_hcs.MultiPointController.plate_view_init Qt signal carries the new list. - UnifiedMosaicWidget.setPlateLayout accepts well_ids, compares against the cached _plate_well_ids frozenset, and zeroes affected layers when the comparison shows coverage changed (or dims changed). - Plate boundary layer is dropped so it gets redrawn at the new slot scale. Tests: 1132 passed, 5 skipped, 1 xfailed. Smoke test clean. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
hongquanli
added a commit
that referenced
this pull request
May 18, 2026
…tbeat-skip, etc. Addresses six issues from the copilot-pull-request-reviewer bot review: #1 (cephla.py): Call acknowledge_aborted_command() after catching CommandAborted (and after the inner resend failure if it was also a CommandAborted), so the next send_command doesn't log the spurious "Last command aborted and not cleared before new command sent!" warning. The inner ack is gated on isinstance(e2, CommandAborted) to avoid the "ack with nothing to ack" path on TimeoutError. #2 (cephla.py): Drop the redundant `target_usteps = ...` recompute after _home_wheel. config and target_pos haven't changed and _target_pos_to_usteps doesn't depend on current_pos. #3 (cephla.py): Fix _home_wheel docstring — wheel is driven to config.min_index (typically slot 1), not "slot 0". #5 (firmware/serial_communication.cpp): Skip the `mcu_cmd_execution_status = COMPLETED_WITHOUT_ERRORS` reset when processing a HEARTBEAT. The keepalive has no result to report, and resetting would clobber a pending CMD_EXECUTION_ERROR from the previous command if the broadcast hasn't fired yet. Eliminates the narrow race where heartbeat traffic interleaves a failure broadcast. #6 (firmware/stage_commands.cpp): Split mark_move_failed() into two helpers — mark_move_failed() (for paths that already set mcu_cmd_execution_in_progress = true) and report_move_error() (for early-return paths that didn't). The !enabled branch in dispatch_filterwheel_move now uses report_move_error() so it doesn't spuriously unwind in_progress for an unrelated motion in flight on another axis. Invariant: only the function that claimed in_progress gets to clear it. #7 (test_filter_wheel.py): Add `getattr(mc, move_rel_attr).assert_not_called()` to both parametrized CommandAborted/TimeoutError tests, so the absolute-MOVETO recovery path is enforced — fall-back to relative MOVE would now be caught.
Alpaca233
added a commit
that referenced
this pull request
May 26, 2026
Picks up the remaining cheap, reasonable fixes from PR #535 review: #1 cv2.imwrite return value: cv2 returns False (without raising) on missing dir, missing codec, or unsupported dtype. We now check the return value, log a throttled WARNING per failure, increment a _silent_write_failures counter (surfaced in the diagnostics summary), and skip counter increment + CSV row write so no row references a file that isn't on disk. #4 channel-name path sanitization: get_image_filepath previously only replaced spaces; now also strips '/', '\\', ':', '*', '?', '"', '<', '>', '|'. Strict improvement for both the recording widget and the multipoint pipeline. #7 frames.csv open guarded: a PermissionError on the CSV file no longer leaves the widget in a half-started state. On failure, experiment_ID is cleared and the exception is re-raised so the toggle_recording slot doesn't run streamHandler.start_recording(). #8 drop the dead image_format constructor parameter: process_queue hasn't read it since the cv2.imwrite refactor — it derives the extension from control._def.Acquisition.IMAGE_FORMAT via get_image_filepath. No in-tree callers passed it. #9 OSError auto-stop emits only once per recording: a saturated queue hitting a persistent OSError (e.g., full disk) no longer fires 10 redundant stop_recording signals. _stop_requested_from_writer is reset in start_new_experiment; subsequent errors during shutdown log as WARNING instead of re-emitting. #18 channel_provider returning None at start_new_experiment now logs a distinct WARNING. Before, a user who launched the GUI and clicked Record without first selecting a live channel got every frame tagged 'live' with no warning (because the provider lambda itself was non-None, just its return value was). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Alpaca233
added a commit
that referenced
this pull request
May 30, 2026
Three fixes from the post-commit review of 3ece218: 1. Move set_exposure_time(self.get_exposure_time()) INSIDE the _pause_streaming with-block. It chains into _update_internal_settings → _calculate_strobe_info which itself writes PRECISE_FRAMERATE (camera_toupcam.py:130). With the call after the with-block, that write happened on a *live* stream — exactly the path the diff's own comment says is unreliable. Now strobe-info, ExpoTime, and the PRECISE_FRAMERATE write all run while paused. 2. Drop the now-redundant CONTINUOUS-specific MAX→PRECISE_FRAMERATE put. _calculate_strobe_info already does the same write, and with #1 it now does it inside the pause window. 3. Reset self._trigger_sent = False after the pause. stop_streaming() cancels any in-flight frame callback that would have cleared the flag, so a mode switch initiated shortly after send_trigger() would leave _trigger_sent=True until ~1.5×exposure+4s timeout — silently blocking the next trigger. Also broadened the diagnostic log's except from HRESULTException to Exception so a non-numeric get_Option return (TypeError on the format spec) doesn't propagate out and break the mode switch at the last step. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
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
1 participant
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.
This adds logging tools in the squid.logging module (along with the squid package), and two helpers there for getting loggers and setting the squid logger logging level. This allows us to have a dedicated hierarchy of loggers that we can configure while leaving all the loggers setup by other libraries alone.
It also adds an example usage in the control._def.py module. We can start converting individual files after this.
It also lays the groundwork for better crash reports by allowing us to hook up handlers for uncaught exceptions. By default these just print to console, but we want to be able to hook up a FileHandler to our logger and have the uncaught exceptions go there as well. Actually creating the FileHandler and attaching it to the root squid logger is to come.
NOTE: This was originally hongquanli#114 but has been moved here since this is going to be the main development repository moving forward.
Tested by:
python -m pytest.