Repository navigation
docs(debug-journal): fold the 2026-10-31..11-08 queue slice (14 of 15 candidates); consolidate the largest areas - #6058
Merged
Conversation
… candidates); consolidate the largest areas Co-Authored-By: Claude Code <noreply@anthropic.com>
springfall2008
marked this pull request as ready for review
November 9, 2026 08:20
springfall2008
enabled auto-merge (squash)
November 9, 2026 08:20
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Automated journal-update run (2026-11-09), folding the 2026-10-31–11-08 queue slice. I read the journal end to end, verified every candidate against current
main(9947c0d), re-checked entries against the merges since the last fold, ranrun_pre_commit(hooks + quick suite green, 5 slow skipped), and left the file at 161.5KB (was 156.4KB). 99 candidates remain queued. Per-candidate and per-correction detail below.Candidates — 14 folded, 1 dropped, 99 left queued.
Folded:
HoldChrgrequires the min SoC over the window at/above the limit, so such windows fall back to generic "Planned Charge". Verified the output.py rule and the prediction.py reserve-pinning on main plus the reporter's own plan/log.battery_rate_maxreading a literal 0 W float defeats the ≤0 W fallback (it reads the same broken register), and a restart mid-incident clears the saved curve so aZeroDivisionErrorstarting long after the trigger means look for a restart between; a crashed run never saves a curve, so repeats recurse. Verified the autodetect gate on main; fix PR fix(inverter): treat a 0W live rate limit as unset, so Predbat's own 0W write cannot latch the rate at 0 (#5219) #5626 still open.nulls, sodict.get(key, 0)defaults are inert; PR fix(solax): treat a null battery chargeDischargePower as 0 in the plant sum #5894 (merged) + the sibling guards. The candidate's not-yet-fixed siblings are no longer live on main (thebatterySOC/batteryRemainings/batteryTemperatureguards changed shape on main); the candidate's "check the siblings" grep recipe is what survived, withfloat(ratedPower)noted as the one shape still unguarded (device-list null, suspect).*_todayarrays' minutes-ago convention.io_confirmed_creditbullet under Car charging and Intelligent dispatch: PV10's dispatch-gone hedge is skipped for confirmed minutes in both engines (verified prediction.py + the kernel mirror, parity 19); a dump with non-emptyio_confirmed_creditscores PV10 kinder than the fix(octopus): model a vanishing IOG dispatch for octopus_intelligent_slot users #5304-era math.rate_low_thresholdmanual cap,combine_charge_slots,set_export_freeze_only), the sale-cap arithmetic, and that additive event boosts can never move a min-rate scan. The maintainer's bug reclassification is noted in the PR body only (per-issue narrative stays out of the journal).best_charge_limit/status_detail) and a new Replays bullet: the dump's rate tables sit back-to-back, grep each^rate_key's line bounds before extracting.*_todaysensors and the empty-slot glyph is== 0after dp2, so ~20 W-equivalent counter ticks render as 0.01–0.02.run_execute_test()inherits fleet count and reserve semantics from the previous scenario._publish_raw()deliberately (display-mode block),run()ordering carries screen-off revocation, and gateway test fakes of_publish_rawmust return True + seed_last_oversize_warn(verified against test_gateway.py fakes).Dropped:
car_force_holdable/car_force_hold_untiland the hold-writer/iteration race on PR feat: cancel a car slot for export, part 2 - planner, valuation and kernel #5905's feature branch (feat/charger-force-holdworkstream). Neither dict nor those writers exist anywhere on main (grepped the whole package). Folding it would have put symbols from an unmerged branch into the journal; the finding lives on the PR thread.Corrections to existing entries (each invalidated by a merge since the last fold):
adjust_battery_target_multi()divides by the fleet rate too" is overtaken: PR Redistribute fleet target split when an inverter has no room, split exports by export rate (#4856 F2) #5906 (3f07cf1, 2026-11-07) replaced that divide withsplit_change_by_rate(), which self-sums and breaks out at 0. Entry now credits the fix and keepsfind_charge_curve()as the live unguarded divide. (Also carries the GH#5890 facts above.)ratedChargePower/ratedDischargePowerper direction, auto-bindsinverter_limit_charge/_discharge, and leaves the MQTTinverterMax*ActivePowerWunread — rewritten from "plans sit under real capability", which fix(sigenergy): use the modules' discharge rating for the discharge ceiling #6054 largely resolved for the battery ceiling.export_limitto GivTCP's user-wins set; entry extended, with the hand-typed-template line caveat from givtcp.py'sGIVTCP_AUTO_CONFIG_USER_WINS_KEYS._settable_reserve()and its constants (80/97/100/3), so that entry needed no change.Consolidation (file was past the ~150KB threshold at 156.4KB):
-1codes); Solis (demo handover,charge_discharge_time_windows); Octopus (IOG: half-hour after Ohme dispatch ends is repriced cheap, triggering Dyness grid charge (v9.6.3) #5884 tail, Ohme immediate Intelligent Octopus dynamic slot visible in HA but not ingested by Predbat #5773 detail); Plan (GH#5991/charging is not following plan #5987 fixed story, GH#5358 and GH#5969 pointers); Startup (GH#5885, GH#5437); inverter service dedup; HA write/verify false-alarms; GitHub owner-gate; gateway EV-planning tail; and the merged PR feat: cancel a car slot for export, part 2 - planner, valuation and kernel #5905-review test bullet. Net after all edits: +5.1KB of verified content while consolidating ~3.2KB — the file grows because 14 verified facts are worth more to the next triage run than the byte count; every condensed bullet retains all its facts, and the largest areas (Solis, Octopus, Plan-charging) now carry more per byte than before.🤖 Generated with Claude Code