Repository navigation
Conversation
…energy (lbbrhzn#2115) * Switch to session-energy mode instead of publishing negative session energy Some chargers report a lifetime register in StartTransaction but session-relative energy in the MeterValues that follow. The `meter_start == 0` detection never fires for them, so the else-branch derives `session = EAIR - meter_start` and publishes a large negative value to a total_increasing sensor. Nothing is logged when it happens. A transaction-bound EAIR sample below meter_start cannot be a lifetime register reading, so treat it as proof the charger reports session energy: set _charger_reports_session_energy and use the sample as-is. Chargers that do report a lifetime register are unaffected, since `derived` is never negative for them. Not scoped at the candidate pre-selector deliberately: rejecting below-baseline EAIR globally would break the existing _charger_reports_session_energy path, where a session legitimately starts at 0 and counts up. Adds a regression test covering a lifetime meter_start followed by session-relative MeterValues. It fails on main with "session energy went negative: -4076.334" and passes with this change. The existing test_eair_monotonic_increments_single_connector does not catch this: its fixture is monotonically increasing, so the derived value is never negative and the branch is never reached. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> * Move the mode switch ahead of skip_eair Addresses CodeRabbit review on lbbrhzn#2115. skip_eair is evaluated before the session-handling block, so on the first below-baseline sample _charger_reports_session_energy was still False and the session-relative value was written to the lifetime Energy.Active.Import.Register metric before the mode switch ran. Detect the below-baseline sample earlier, immediately before skip_eair is computed. The existing session-energy branch then handles the sample, so the derivation branch needs no change at all: this commit is a pure addition and modifies no existing line. The regression test now also asserts the lifetime register is not overwritten with the session value. Note the check is made only after the sample is processed: _metrics is a defaultdict, so reading a key beforehand creates it and alters the behaviour under test. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> --------- Co-authored-by: kwilson9 <190990796+kwilson9@users.noreply.github.com> Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
…bbrhzn#2071) * build(deps): update setuptools requirement from >=83.0.0 to >=84.0.0 Updates the requirements on [setuptools](https://github.com/pypa/setuptools) to permit the latest version. - [Release notes](https://github.com/pypa/setuptools/releases) - [Changelog](https://github.com/pypa/setuptools/blob/main/NEWS.rst) - [Commits](pypa/setuptools@v83.0.0...v84.0.0) --- updated-dependencies: - dependency-name: setuptools dependency-version: 84.0.0 dependency-type: direct:production ... Signed-off-by: dependabot[bot] <support@github.com> * update hacs version --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: drc38 <20024196+drc38@users.noreply.github.com>
* Allow removing a charge point device via the UI Implement async_remove_config_entry_device so that deleting a charger's device in Settings -> Devices removes the charge point from the entry's CONF_CPIDS. The update listener reloads the integration, so the device is not re-created on setup; a charger reconnecting later goes through the normal discovery flow instead of being silently re-added. The central system device (and anything matching no configured charge point) is refused, so it can never be removed by accident. Adds tests for the happy path (sibling charge point survives, no re-creation), the central-system refusal, and removing the last charge point (chargerless entry still loads/unloads), plus a user-guide section. Refs lbbrhzn#1885 * Reject central-system identifier collisions in device removal When CONF_CSID equals a charge point's cp_id or cpid, setup registers both devices under the same identifier and Home Assistant merges them into a single device. The removal hook would then match it as a charge point and silently drop the entry from CONF_CPIDS. Refuse to remove any device carrying the central system identifier, and add a regression test for the collision. * test: cover unknown device refusal in async_remove_config_entry_device * test: snapshot full config_entry.data on removal refusal * test: migrate device registry usage and fix deepcopy of config_entry.data * test: add entity rebuild assertion and ruff format async_get_devices * test: probe the surviving sibling's entity for the reload, not the removed one * style: apply ruff format to async_get_devices assertions
…bbrhzn#2129) Time.Session has no device class, so its fallback unit is None, and the unit getter replaced a known unit with that fallback whenever the live metric had no unit yet: before the charger connected, while it was offline, and on every restart or reload before a transaction ran. Home Assistant saw a measurement sensor's unit change from min to blank and raised a units_changed repair each time. Both protocol handlers compute the session time in minutes, so the sensor now reports minutes unconditionally. The debugging guide explains the one-time statistics repair for history recorded with a blank unit. Closes lbbrhzn#1938 Co-authored-by: KingHavok <shane@shanewilliams.com.au>
Bumps [pytest-homeassistant-custom-component](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component) from 0.13.361 to 0.13.363. - [Release notes](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/releases) - [Changelog](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/blob/master/CHANGELOG.md) - [Commits](MatthewFlamm/pytest-homeassistant-custom-component@0.13.361...0.13.363) --- updated-dependencies: - dependency-name: pytest-homeassistant-custom-component dependency-version: 0.13.363 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
… stops (lbbrhzn#2128) The integration allocates 1.6 transaction ids itself as int(time.time()), and three things leaned on that: two connectors starting in the same second shared an id, so the stop for one was applied to whichever came first in the active map; a StopTransaction matching nothing was applied to connector 1, ending a session that may still have been running there; and the id doubled as the session's start epoch, so an id adopted from a charger's MeterValues after a restart made the timer count from an arbitrary reference. Ids are now max(now, last + 1) per charge point, with the floor raised by every id seen, restored or persisted. The session start is kept separately and persisted per connector in a small HA Store with the last allocated id, loaded before the message loop; an estimated start is marked on the session time sensor. An unknown stop is never guessed: the live connectors are held with their timers frozen until the charger says something about them, a pending stop is applied only to the one connector shown to have ended, and evidence that a candidate ended by another route discards the stop instead. Remote stops are refused for held connectors and the Charge Control switch is unavailable for them. Closing values speak only for the transaction whose id they carry, are never adopted as a live session, and no longer reset the timer. Closes lbbrhzn#2123 Co-authored-by: KingHavok <shane@shanewilliams.com.au>
Bumps [ruff](https://github.com/astral-sh/ruff) from 0.16.5 to 0.16.6. - [Release notes](https://github.com/astral-sh/ruff/releases) - [Changelog](https://github.com/astral-sh/ruff/blob/main/CHANGELOG.md) - [Commits](astral-sh/ruff@0.16.5...0.16.6) --- updated-dependencies: - dependency-name: ruff dependency-version: 0.16.6 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Create one flat Maximum Current entity per charger and remove obsolete connector entities, preserving existing flat identities and restore data. Route the slider through a station-only API: OCPP 1.6 requires ChargePointMaxProfile; OCPP 2.0.1 keeps its current station behavior. The set_charge_rate action retains its existing fallback chain. Document the breaking entity migration and removal of the 1.6 slider fallback, including single-connector chargers. Add regression coverage for registry cleanup, restoration, protocol errors and service behavior. Co-authored-by: KingHavok <shane@shanewilliams.com.au>
Bumps [pytest-homeassistant-custom-component](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component) from 0.13.363 to 0.13.364. - [Release notes](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/releases) - [Changelog](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/blob/master/CHANGELOG.md) - [Commits](MatthewFlamm/pytest-homeassistant-custom-component@0.13.363...0.13.364) --- updated-dependencies: - dependency-name: pytest-homeassistant-custom-component dependency-version: 0.13.364 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
…brhzn#2141) TransactionEvent (lbbrhzn#2135): keep a per-transaction seqNo guard so stale or replayed events cannot resurrect an ended transaction; resolve a missing connectorId only when the EVSE topology makes it unambiguous; apply offline events only to the transaction currently displayed; hold events that arrive before the inventory settles and replay them once the EVSE map is known, falling back to dynamic allocation when it never is; treat an online event for another transaction as current evidence, except an Ended or an Updated stamped no later than the displayed transaction's latest event, which is replayed history. Rate unit (lbbrhzn#2134): read SmartChargingCtrlr/RateUnit from the base report and convert managed limits for W-only or A-only stations. Foundations for a transaction-scoped control: the electrical helpers move to the base charge point, reconnect demotes held events while BootNotification resets transaction ordering, and both handlers report transaction start and end to an optional hook. Co-authored-by: KingHavok <shane@shanewilliams.com.au>
Bumps [pytest-homeassistant-custom-component](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component) from 0.13.364 to 0.13.365. - [Release notes](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/releases) - [Changelog](https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/blob/master/CHANGELOG.md) - [Commits](MatthewFlamm/pytest-homeassistant-custom-component@0.13.364...0.13.365) --- updated-dependencies: - dependency-name: pytest-homeassistant-custom-component dependency-version: 0.13.365 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
…without Authorize (lbbrhzn#2147) * ocpp16: charger level id_tag sensor not updating on StartTransaction without Authorize When an OCPP 1.6 charger performs local authorization it skips the Authorize message and sends StartTransaction directly. The on_start_transaction handler now also sets charger-level (connector 0) id_tag and not only the connector-level id_tag * Only set charger IdTag from StartTransaction on single-connector chargers. A multi-connector station tag must not be overwritten by whichever plug started last.
Replace fenced terminology in timeout messages, the initial-start comment, and lifecycle documentation. Wording only; no control-flow changes. Addresses review feedback on lbbrhzn#2122.
Integrate lbbrhzn/ocpp main at 0a3b753: - lbbrhzn#2146: update pytest-homeassistant-custom-component to 0.13.365. - lbbrhzn#2147: update the charger-level ID-tag sensor for locally authorized starts. Preserve PR lbbrhzn#2122 reconnect ownership, bounded retirement, regression tests, and blocked terminology without rewriting the existing commits.
This branch was successfully deployed
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
6 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.
Validation only — do not merge
This draft runs the existing GitHub Actions workflows against the synchronized candidate for lbbrhzn#2122. Neither the fork's main branch nor upstream main has been changed by this validation PR.
Candidate and source identity
e44aaf4c5347afe442358c0f77e39997e4edf099.0a3b753454716c9e46f1ee3d4adee4d49c1769db(build(deps): bump pytest-homeassistant-custom-component from 0.13.364 to 0.13.365 lbbrhzn/ocpp#2146 and ocpp16: charger level id_tag sensor not updating on StartTransaction without Authorize lbbrhzn/ocpp#2147).matiebird.c2c68088c3b1a5723b90b5b4064334655bbd4b1b.a37979bba0a31c6b496834cc4c09ab9a9fd98e8d, and upstream's proposed merge,fc3de3f388e78a0b6fb1b3a172e286a179b17771, have that same tree. The fork therefore tests the exact candidate source, not a divergent version.Results at last inspection
Fork workflow: https://github.com/matiebird/ocpp/actions/runs/35397799877
The existing 95% coverage requirement and slow reconnect tests remain enabled. No workflow files, test thresholds, repository settings, permissions, or secrets were changed to obtain these results.
Upstream checks are separate
The new upstream Linting run (35397728191) and Pull Request Docs Check run (35397728328) report
action_required. The upstream Linting run has no jobs. Fork-side results do not replace the upstream approval requirement or establish that upstream CI has passed.This draft remains open only as the validation record while the test run is unfinished. It must not be merged into the fork's main branch.