fix(wsen-hids): Restore sensor config after power cycle. by nedseb · Pull Request #240 · steamicc/micropython-steami-lib

nedseb · 2026-03-25T11:08:54Z

Summary

Fix one-shot measurement timeout after a power_off() / power_on() cycle.

Root cause

power_on() only set the PD bit in CTRL_1 but did not restore the sensor configuration (averaging, BDU, calibration). After power_off(), these settings are lost, so the sensor was electrically powered but not properly configured for measurements.

Fix

Save init parameters (avg_t, avg_h, bdu) as instance attributes
power_on() now restores averaging, BDU, and calibration after setting the PD bit

Tests added

Power on restores averaging config (mock)
Power on restores BDU setting (mock)
Read one-shot works after power cycle (mock)

Hardware validation

--- Test: read_one_shot after power cycle ---
  Cycle 1: humidity=51.85 %RH, temperature=27.46 C
  Cycle 2: humidity=37.19 %RH, temperature=27.48 C
  Cycle 3: humidity=35.99 %RH, temperature=27.39 C
  Cycle 4: humidity=36.30 %RH, temperature=27.45 C
  Cycle 5: humidity=36.54 %RH, temperature=27.37 C

PASS: 5 power cycles completed without timeout

Closes #238

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 3f42ec46dd

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-03-25T11:12:25Z

+        self.set_average(self._avg_t, self._avg_h)
+        if self._bdu:
+            self.enable_bdu(True)


Preserve runtime config before restoring in power_on

power_on() now reapplies self._avg_t, self._avg_h, and self._bdu, but those fields are only initialized in __init__ and never updated by set_average()/enable_bdu(). After a caller changes averaging or BDU at runtime, a power_off()/power_on() cycle silently reverts to constructor-time settings (for example, enable_bdu(False) can be undone if the instance was created with enable_bdu=True), so the driver mutates user-selected configuration unexpectedly.

Useful? React with 👍 / 👎.

Copilot

Pull request overview

This PR fixes WSEN-HIDS one-shot measurement timeouts after a power_off() / power_on() cycle by reapplying sensor configuration that is lost across power-down.

Changes:

Persist initial averaging/BDU settings on the driver instance and reapply them in power_on().
Re-read calibration data on power_on() to restore conversion parameters after a power cycle.
Add mock scenario tests intended to validate config restoration and post-cycle one-shot reads.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
`lib/wsen-hids/wsen_hids/device.py`	Store init config and restore averaging/BDU/calibration during `power_on()` to recover correct sensor state after power cycling.
`tests/scenarios/wsen_hids.yaml`	Add mock scenario tests for averaging/BDU restoration and one-shot reads after a power cycle.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-25T11:12:53Z

        self._write_reg(REG_CTRL_1, ctrl1)
+        self.set_average(self._avg_t, self._avg_h)
+        if self._bdu:
+            self.enable_bdu(True)
+        self._read_calibration()


power_on() now restores averaging/BDU using _avg_t/_avg_h/_bdu, but those attributes are only set once in __init__. If a user calls set_average(...) or enable_bdu(False/True) after initialization, a later power_on() will restore the initial settings (and can even re-enable BDU after it was disabled). Consider updating _avg_t/_avg_h inside set_average() and _bdu inside enable_bdu(), and have power_on() apply enable_bdu(self._bdu) (both enable and disable) to reflect the current configuration.

Copilot · 2026-03-25T11:12:54Z

+      dev.power_on()
+      av_conf = i2c.readfrom_mem(dev.address, 0x10, 1)[0]
+      result = av_conf != 0


This test doesn't currently validate the new behavior: with FakeI2C, power_off() only clears the PD bit and does not reset AV_CONF, so av_conf will remain whatever was written during __init__ even if power_on() didn’t restore it. To make this meaningful, either (a) simulate a power-cycle reset by explicitly clobbering AV_CONF (e.g., write 0x00) after power_off() before calling power_on(), or (b) assert that power_on() performed a write to REG_AV_CONF via i2c.clear_write_log() / i2c.get_write_log(). Also, asserting av_conf == 0x1B (per the scenario’s mock_registers comment) is stronger than != 0.

Suggested change

dev.power_on()

av_conf = i2c.readfrom_mem(dev.address, 0x10, 1)[0]

result = av_conf != 0

# Simulate power-cycle reset of AV_CONF

i2c.writeto_mem(dev.address, 0x10, bytes([0x00]))

dev.power_on()

av_conf = i2c.readfrom_mem(dev.address, 0x10, 1)[0]

# Expect AVG_T=16, AVG_H=16 -> 0x1B as per mock_registers

result = av_conf == 0x1B

Copilot · 2026-03-25T11:12:54Z

+  - name: "Power on restores BDU setting"
+    action: script
+    script: |
+      dev.power_off()
+      dev.power_on()
+      ctrl1 = i2c.readfrom_mem(dev.address, 0x20, 1)[0]
+      result = (ctrl1 & 0x04) != 0
+    expect_true: true
+    mode: [mock]


Similar to the averaging test, this will pass even if power_on() never restores BDU because the mock starts with CTRL_1=0x04 and power_off() doesn’t clear the BDU bit in FakeI2C. To ensure the test actually covers the fix, simulate register loss after power_off() (e.g., overwrite CTRL_1 to 0x00) before power_on(), or assert that power_on() wrote the BDU bit back by inspecting i2c.get_write_log().

Copilot · 2026-03-25T11:12:54Z

+  - name: "Read one-shot works after power cycle"
+    action: script
+    script: |
+      dev.power_off()
+      dev.power_on()
+      h, t = dev.read_one_shot()
+      result = isinstance(h, float) and isinstance(t, float)
+    expect_true: true
+    mode: [mock]


As written, this likely passes even on the pre-fix implementation because the mock STATUS register is hardcoded to “data ready” and power_off() doesn’t invalidate calibration/config in FakeI2C. To make it exercise the power-cycle regression, consider simulating lost state (e.g., clear dev._calibration and/or clobber key config registers after power_off()) and then assert power_on() repopulates calibration/config before read_one_shot() succeeds (or assert the expected configuration writes occurred using i2c.get_write_log()).

nedseb · 2026-03-25T11:17:50Z

All 5 review comments addressed in 8066562:

Runtime config sync — set_average() and enable_bdu() now update _avg_t/_avg_h/_bdu so power_on() restores the current config, not just the constructor defaults.
Mock tests simulate register loss — All 3 mock tests now clobber registers (write 0x00) after power_off() before calling power_on(), ensuring the tests actually exercise the restore logic.
New test for runtime changes — Added "Power on preserves runtime config changes" which disables BDU at runtime, power-cycles, and verifies BDU stays disabled.

## [0.0.3](v0.0.2...v0.0.3) (2026-03-25) ### Bug Fixes * **wsen-hids:** Restore sensor config after power cycle. ([#240](#240)) ([56e5d39](56e5d39))

semantic-release-updater · 2026-03-25T11:47:24Z

🎉 This PR is included in version 0.0.3 🎉

The release is available on GitHub release

Your semantic-release bot 📦🚀

fix(wsen-hids): Restore sensor config after power cycle.

3f42ec4

Copilot AI review requested due to automatic review settings March 25, 2026 11:08

github-actions Bot assigned nedseb Mar 25, 2026

Copilot started reviewing on behalf of nedseb March 25, 2026 11:09 View session

chatgpt-codex-connector Bot reviewed Mar 25, 2026

View reviewed changes

test(wsen-hids): Add hardware test for power cycle read_one_shot.

5fc7208

Copilot AI reviewed Mar 25, 2026

View reviewed changes

steamicc deleted a comment from chatgpt-codex-connector Bot Mar 25, 2026

fix(wsen-hids): Sync runtime config and improve mock tests.

8066562

test(wsen-hids): Cover runtime averaging restore after power cycle.

feaf8bb

nedseb added the bug Something isn't working label Mar 25, 2026

nedseb merged commit 56e5d39 into main Mar 25, 2026
3 checks passed

nedseb deleted the fix/wsen-hids-power-cycle branch March 25, 2026 11:47

semantic-release-updater Bot pushed a commit that referenced this pull request Mar 25, 2026

chore(release): v0.0.3. [skip ci]

5c605a6

## [0.0.3](v0.0.2...v0.0.3) (2026-03-25) ### Bug Fixes * **wsen-hids:** Restore sensor config after power cycle. ([#240](#240)) ([56e5d39](56e5d39))

semantic-release-updater Bot added the released label Mar 25, 2026

nedseb mentioned this pull request Mar 25, 2026

examples: Replace test scripts with practical examples for wsen-hids. #239

Merged

9 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(wsen-hids): Restore sensor config after power cycle.#240

fix(wsen-hids): Restore sensor config after power cycle.#240
nedseb merged 4 commits intomainfrom
fix/wsen-hids-power-cycle

nedseb commented Mar 25, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot Mar 25, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 25, 2026

Uh oh!

Copilot AI Mar 25, 2026

Uh oh!

Copilot AI Mar 25, 2026

Uh oh!

Copilot AI Mar 25, 2026

Uh oh!

nedseb commented Mar 25, 2026

Uh oh!

Uh oh!

semantic-release-updater Bot commented Mar 25, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

nedseb commented Mar 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Root cause

Fix

Tests added

Hardware validation

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Mar 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 25, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 25, 2026

Choose a reason for hiding this comment

Uh oh!

nedseb commented Mar 25, 2026

Uh oh!

Uh oh!

semantic-release-updater Bot commented Mar 25, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

nedseb commented Mar 25, 2026 •

edited

Loading