ci9-gateproof
33
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 glenn schrooyenandClaude Opus 5
|
a6af332048 |
SAFETY-48: what the lead cap actually does to a frozen integrator, and what the peak figure depends on
checks / unit (pull_request) Successful in 7m55s
Four review findings, none of them in executable code. The control law, the tests and the instruments' arithmetic are untouched; every number below was re-measured against the committed 16-day capture before it was written down. R1 - app/control.py carried a justification the decision record and MUTATION.md habit 22 had already retracted, and it was false about the code it sits in. It said an unconditional placement "would also pull the integrator toward a command the SATURATION freeze is pinning", priced at 118 of 256 250 ticks. The 118 was measured against a different pair; on the pair the sentence names the two forms are bit-identical - 0 of 256 246 ticks, re-measured. And the narrow form that shipped DOES move a saturation-frozen integrator: it runs before the freeze block and is not gated on `frozen`. Of 79 160 frozen ticks the cap changes `i_w` on 1095, and on 679 of those it moves it further out on the side it was already on - putting back magnitude the freeze had just let fall, which the freeze's own rule forbids. Worst case 2424 W, at prev_w -3400 W, grid +4697 W, actual -4144 W, sat_count 2 -> 3: the freeze leaves the integrator at -176 W and the cap drags it to -2600 W. It never overrides a same-cycle refusal (the freeze clipped on 75 512 ticks, the slew bound on none of them), and that is stated too. It is also where SAFETY-13's settle values went: gating the cap on `not frozen` puts test_control.py part 5 back at -1190 W exactly, while part 4(b) does not come back - that one moved for the winding reason. R3 - the record read its own sensitivity table backwards. `fit_plant()` picks lag 7.5 s on battery-power RMS, which separates it from 2.5 s by 0.3 %, while the tool's own printed fidelity measure (export against the recorded 29.43 kWh) prefers 2.5 s by the whole 35 % gap. Swept properly: at every lag <= 5.0 s release/1.0's worst quarter is ALREADY 3434 W and this card changes it by exactly zero. The 3739 -> 3434 W headline exists at one plant. New section 3a carries the table; the export reduction now leads every summary and the peak is presented as lag-dependent. The form-A rejection is strengthened rather than weakened: form A is worse on the billed quarter at every lag where release/1.0 produces a peak at all, and at every top-N cut of the quarters. R2 - DOCS.md stated modelled watts to the owner as measurement. It now leads with the export reduction, says the figures are simulated, and says plainly that no capacity-tariff improvement is promised. R4 - tools/slew_sim.py --slew narrated "the OVER-ABSORPTION does not move, and at the slowest slews it is slightly WORSE" while printing a monotonically falling column. That sentence is TRUE on release/1.0 (reproduced: 958/1107/1133/1058/987/1029/1150/1143/1081) and false under this branch's law (777/734/663/563/553/507/435/358/321), so SAFETY-39 is not retracted: its record gets a forward reference, and the script now reads SLEW_LEAD_STEPS out of the law it imported and narrates from the column it just printed. Both branches of that narration were driven. R5 - three counts in the L table were decoration and are re-driven in the CI image on a `cp -a`, purged, `:ro`, one file at a time: L5 is 5+5 with the function names exchanged and 5+4 with the branch bodies exchanged (two mutations wearing one name, now listed separately), L6 is 6+3. The separable- state count is grid-dependent and three sweeps disagree, so all three are quoted with their grids and the load-bearing claim is WHERE they separate. R7 - "a large import answered too weakly" is not what the aggregate shows. Form A cuts import by 5.45 kWh and export by 5.18 kWh, and paired per tick delivers 8.56 kWh less offtake energy on the ticks release/1.0 was importing. The harm is a tail redistribution inside quarters that already net to offtake. The conclusion stands; the sentence is narrowed to it. Verified: the full suite in the CI image, purged tree, `:ro` mount, one file at a time, exactly the command in .gitea/workflows/checks.yml - 30/30, exit 0. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
03ef61fa3b |
SAFETY-48: the slew step ceiling reaches the integrator, and what it bounds is the LEAD
`compute()` freezes the integrator with an asymmetric rule and the comment beside it states the requirement outright - "the requirement is that the INTEGRATOR stop accumulating, not only the command" - but `frozen` is computed from ONE signal, `abs(prev_w - actual_w) > saturation_w`, and slew-limiting never enters it. The integrator moves `gain * error` a cycle and the command moves at most `slew_w`, so the integrator outruns the command whenever `|error| > slew_w / gain`: 667 W at the live shape, against a measured median grid error of 1114 W at the onset of a recorded episode. SAFETY-39's retune made it MORE frequent, not less - a smaller step is a lower threshold. ⚠️ THE FORM THE CARD ASKED FOR RAISES THE BILLED QUANTITY, WHICH IS WHY IT IS NOT WHAT SHIPPED. Reusing the asymmetric freeze on a carried slew flag was built from release/1.0's own source and measured over 16.00 days / 256 252 meter instants of the live install's recorder: export 39.61 -> 34.43 kWh, and the worst quarter-hour average OFFTAKE 3739 -> 3935 W with the mean quarter +2.73 W, in five of six plant and tuning variants. A frozen integrator stops at `prev + gain*error`, which undershoots by (1 - gain); for an export that is free and for an import it is offtake, for as long as the ramp lasts. The card said "do not invent a second rule"; the measurement says to, and both sentences are recorded. WHAT SHIPS BOUNDS THE LEAD AT ONE SLEW STEP, inside the branch where the slew actually bound. Same 16 days, same ticks, same disturbance, same plant, same seed: export 39.61 -> 34.07 kWh, wasted throughput 4.67 -> 3.19 kWh, battery sign reversals 2992 -> 2770, median recovery 232 -> 199 s, p99 849 -> 850 s, max 901 -> 898 s, unrecovered 53 -> 31, worst |i_w - prev_w| 9200 -> 800 W - and the worst 15-minute average offtake 3739 -> 3434 W, i.e. the billed quantity IMPROVES. `i_w = slewed` (a cap of zero) is better again at 32.02 kWh and is refused: it is the velocity form this module's docstring says `i_w=None` collapses to. TWO COMMITTED NUMBERS MOVED, DELIBERATELY, WITH THE REASON BESIDE THEM. SAFETY-13's released command settles at -800 W and -1200 W where release/1.0 settled at -1190 W and -2390 W, because there is less wound integrator to settle onto; AC 4's property - it stops at the frozen integrator, not the rail - is what is asserted and the second is now graded against `i_w` rather than a literal. And SAFETY-39's "the slew does not own the over-absorption" is now FALSE - INVERTED, on the independent first-order plant test_control.py writes itself: 1005/1109/1285/1330 W rising with a slower slew becomes 729/550/424/326 falling. Two plant models that share nothing agree on the sign. Evidence: `tools/replay_recorded_loop.py` plus the read-only capture it made (scenarios/loop-2026-09-02_2026-09-17). The "before" law is the shipped file with the block removed by a named substitution - never a transcription - and a missing anchor is a REFUSAL carrying "NOTHING HAS BEEN MEASURED". The instrument prints its own fidelity first: it reproduces 39.6 kWh of export where the install exported 29.4, i.e. it is ~35 % pessimistic about the loop's tracking, so only the delta is claimed. 30/30 suites green, including test_control.py's exhaustive i_w=None equivalence sweep against the transcribed release/1.0 law, untouched: the cap moves `i_w` and not `target`, and that sweep compares (target_w, sat_count, frozen, reason). Six law mutations, one instrument battery. L4 - the cap applied unconditionally, -92 B - was a SURVIVOR of all thirty files: it is bit-identical on the shipped configuration (0 of 256 246 ticks) and separable on 2050 of 257 400 swept states, all at `integrator_max_w` above `max_w`, which part 4 now drives. L5 and L6 are ZERO BYTES. AND TWO COMMENTS IN THIS CARD WERE INVENTED AND ARE CORRECTED IN PLACE. The one-sided spelling was justified by a two-sided form that would "drag a trailing integrator" - it does not, and the two agree on 205 920 of 205 920 swept states; and a "118 of 256 252 ticks" figure had been measured against a different pair. MUTATION.md gains habits 21 and 22 for both shapes. NOT DEPLOYED, NOT VERSIONED. The live install still runs release/1.0's law. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
b548525d10 |
TEL-10: the loop's integrator stops being invisible
`self.i_w` has been a separate quantity from the command since SAFETY-04 and has left the process nowhere at all - no entity, no MQTT attribute, no log line. The investigation behind SAFETY-48 could only RECONSTRUCT it from the grid trace plus a replay; the reconstruction reproduced about a third of the engagements it was checked against and was discarded, so the card it was gathered for had to be argued from a measurement nobody could stand behind. It could not have been made to work either: `i_w` depends on every cycle before it, including the ones where the arbiter overrode the loop and `main.py` re-seeded `self.i_w = self.target`, and that re-seed is invisible in the grid trace. THREE SITE ROWS, APPENDED AFTER release/1.0's EIGHT. `loop_i` is `Decision.i_w` - the accumulator after the cycle, before the output clamp, the slew limit and quantisation. `loop_lead` is `decision.i_w - prev_w` against the command `compute()` was HANDED, which is not a subtraction of the other two published rows: `Setpoint` is the arbiter's output and is this cycle's, so on any cycle the arbiter overrode the loop the two differ - and that is exactly the cycle on which the integrator was re-seeded. `loop_reason` is `compute()`'s own word; the arbiter's `explain()` prints the winning claim's NAME and has never carried the claim's reason, so `slew-limited` appears in neither the `Why` row nor `Controller status`. Measured on the same cycle: arbiter `'loop -> -5000 W'`, loop `'i-clamped'`. NOTHING IN THE CONTROL PATH MOVED. `app/control.py` is not in this diff. `compute()`, its returned `Decision` and every constant it reads are identical to release/1.0, and all nine of release/1.0's retained discovery messages are byte-identical and still FIRST in the announcement. Evidence: 30/30 suites green in the CI image. Nine mutations, ONE survivor, reported as one: `decision.i_w - self.target` for the lead is +2 B and green because `cycle()` has not reassigned `self.target` at that line - the name is kept for the coupling, not for a behaviour it buys, and M2b (+8 B) is the mutation that does change the value and is red. AND TWO OF THIS CARD'S OWN CHECKS COULD NOT REACH THE FAILURE THEY NAMED. "the published integrator reaches its bound" was GREEN under the card's named mutation (publish `prev_w`, -3 B), because the command reaches the bound too - twelve cycles later. It asserts the LEAD TIME now and is red on 5. "a stopped controller publishes not-running" was GREEN under deleting that branch's assignments (-45 B, -67 B), because the fixture STARTED stopped and `__init__` writes the same values; it stops the controller part-way through a run that has already wound the integrator to -5000 W. Both are MUTATION.md habit 4, found by the battery rather than by review. No CHANGELOG section: `test_version_gate.py` 11b refuses a newest `## ` heading that is not config.yaml's `version:`, and opening one would mean taking a packaging decision this card does not get to take. The operator account is in DOCS.md and the release note belongs to the next PKG ticket. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
f6f3f227c1 |
INSTR-01: restore M12 on the leg I generalised past, and measure the clause that had no row
checks / unit (pull_request) Successful in 7m6s
Review found two things this PR asserted about itself that were false. Both are mine and both are worse than the defect the card was filed for. ⚠️ "NO FIXTURE CAN RESTORE IT, MEASURED BOTH WAYS" WAS NOT TRUE, AND TWO MEASUREMENTS ARE NOT AN IMPOSSIBILITY PROOF. I deleted M12 - SAFETY-44's working end-to-end control for the off-roster narrowing - on the argument that the provenance clause made it unrestorable. The structural claim behind it, that `run()` builds both snapshots from one state file so "already latched" and "made by this transition" are incompatible, IS TRUE OF SAME-STEP AND FALSE OF STAGED: `snapshot_over()` starts the after-leg over the file the BEFORE-leg wrote, so a transition-made destination is available there. Both arrangements I measured were same-step shaped. Review found the third. RESTORED, and it discriminates with all six clauses in: staged leg, legacy+shed, shipped _m12_new = VANISHED unit/shed/sign_tripped ...C5 narrowing removed _m12_new = [] ⚠️ suppressed ⚠️ AND MY FIRST RESTORATION DIDN'T DISCRIMINATE EITHER, for a reason worth recording: `data.pop(k, False)` RETURNS the value, so shed lost the KEY and kept the LATCH - `_grade_store_migration()` saw `on_roster_latched`, never re-armed garage, and there was no transition-made destination. The finding was reported whether or not C5 existed and the control looked restored. `and False` is the difference between "the key is gone" and "the LATCH is gone", and only the second reaches the re-arm this arrangement needs. ⚠️ WHAT ACTUALLY BROKE WAS M12's exit-0 PRECONDITION, NOT M12. A legacy-shaped fixture is never clean. That precondition is the one SAFETY-44's own commit called "belt-and-braces, load-bearing only if the exit-0 requirement is ever relaxed", saying what closed the gap is `_m12_new` BEING A SET DIFFERENCE - and a stable dirty baseline is fine for a set difference. So I deleted a control on the strength of a precondition the same commit had already called not load-bearing. That is removing coverage rather than failing to add it, which is the worse of the two directions. ⭐ AND THE BARE-KEY CLAUSE HAD NO ROW AT ALL. Review deleted `k.rsplit("/", 1)[-1] == bare` and the whole 28-file suite stayed GREEN. 5e mutated five clauses; the comprehension carries more, and this one is load-bearing: a vanished SAFETY-02 `sign_tripped` was suppressed by a SAFETY-42 `delivery_tripped` on a roster unit - a DIFFERENT FAILSAFE's key. The rule's own comment says "THE TEST IS NOT 'the name appears somewhere'. It is: the SAME bare key", and nothing measured that sentence. `unit/*/delivery_tripped` is machine-read, so it is reachable on any real two-latch store. R7 added; C2 now flips exactly R7. All six clause-removals now flip exactly their own row: C1 machine_read -> ['R6-machine-read'] C5 off-roster -> ['R2-narrowing'] C2 same bare key -> ['R7-bare-key'] C6 injectivity -> ['R5-injectivity'] C3 same value -> ['R4-value'] C7 provenance -> ['R3-provenance'] THREE DANGLING CITATIONS TO A CONTROL THIS PR HAD DELETED, one of them in the shipped tool eight lines below the paragraph condemning exactly that ("a citation is the worst possible carrier, because it reads as having been checked"). The tool no longer names a check; it states the property - every clause has a row, and each removal flips its own. A STALE SAFETY-44 BLOCK saying both fixes were "REVERTED OUT OF THIS CARD, because THIS SUITE CANNOT MEASURE EITHER ONE" - describing a tree this PR replaced. Rewritten as a map of where each clause is now measured. A comment that outlives its subject is the same defect as a citation that outlives its check. AND A SELF-CONTRADICTION about the driver, in the file and in the last commit message: "THE DRIVER IS WRITTEN INTO THE TREE UNDER TEST" eight lines above "THE DRIVER GOES IN A TEMPDIR, NOT IN THE TREE". The code does the tempdir; the first sentence stated the read-only-CI bug the same commit says it fixed, as if it were the design. It now says the driver IMPORTS the tree and is WRITTEN to a tempdir. ONE HOLE REMAINS OPEN AND IS STATED AS UNMEASURED: attribution. `sorted()` decides which vanished key claims a destination, so a deleted key can sort first and be named as the one that migrated while the key that really moved is reported VANISHED. Review measured that `sorted(` -> `list(` leaves the suite green - every 5e case has at most one candidate, so no row can see ordering. Still exit 1, so a wrong sentence rather than a hidden loss. Recorded rather than left looking guarded. 28/28 files pass. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
91b0d5e4dc |
INSTR-01: grade the suppression rule at the level it is written at, one case per clause
checks / unit (pull_request) Successful in 7m4s
SAFETY-44 produced THREE inert controls for this one rule, each inert differently, and split the two remaining fixes out rather than ship them unmeasured. This is that card, and the first thing it did was explain why the end-to-end route could never have worked. ⚠️ `moved_to` IS ONE COMPREHENSION CARRYING FIVE CONDITIONS WITH FIVE DIFFERENT REASONS TO EXIST, AND `run()` BUILDS BOTH SNAPSHOTS FROM ONE STATE FILE. So "the destination was already latched" and "the destination was made by this transition" are the same file answering two incompatible questions: the arrangement that reaches one forecloses the other. That is why rebuilding a fixture to reach clause N kept moving the others, and it is not a fixture problem. ⭐ SO THE RULE IS GRADED WHERE IT IS WRITTEN - `diff()` on constructed Snapshots, ONE CASE PER CLAUSE. Six rows, and on the SHIPPED tree before any fix they already separated the two known defects: R1 rename (legitimate) migrated correct R2 narrowing vanished correct R3 provenance MIGRATED ⚠️ the defect R4 same value vanished correct R5 injectivity BOTH migrated ⚠️ the defect R6 machine_read vanished correct The two one-line fixes SAFETY-44 reverted are back, and R3/R5 flip while R1/R2/R4/R6 do not - so the legitimate rename is still suppressed and this is not an over-correction. ⭐ AND EACH CLAUSE REMOVAL FLIPS **EXACTLY ITS OWN ROW**, which is the acceptance criterion this card was filed with: C1 machine_read removed -> ['R6-machine-read'] C3 same value removed -> ['R4-value'] C5 off-roster removed -> ['R2-narrowing'] C6 injectivity removed -> ['R5-injectivity'] C7 provenance removed -> ['R3-provenance'] A removal that flips nothing is an unmeasured clause; a removal that flips someone else's row means the cases are not separable after all. Neither happened. ⚠️ AND M12 IS DELETED, WHICH IS THE MOST USEFUL THING THIS CARD FOUND. M12 was SAFETY-44's end-to-end control for the narrowing and it was a REAL control - red when the narrowing was removed. Adding the provenance clause disarmed it, and NO FIXTURE CAN RESTORE IT, measured both ways: with the sibling latched in its own right, C7 rejects the destination before C5 is consulted, so the mutation is reported whether or not C5 exists; making the destination transition-made means a flat latch, which makes the shipped run dirty and breaks M12's own precondition. The second half of the card's acceptance criterion - "M12 still turns red with the new clauses in" - is therefore UNSATISFIABLE, and the honest answer is to remove the control rather than keep one that has become inert. R2 measures C5 strictly better. Part 5 keeps the end-to-end coverage; discrimination lives in 5e. `state-both-latched.json` goes with it. Nothing referenced it once M12 was gone, and the first draft of that paragraph said it was "kept, because 5e's R2/R3 are derived from the arrangements it represents" - a nice sentence and not a reference. An unused fixture a comment calls load-bearing is an inert check one directory over. TWO THINGS THE DRIVER GETS RIGHT ON PURPOSE: - it is written to a TEMPDIR, not beside the tool. `checks / unit` mounts the checkout READ-ONLY, so writing into the tree works on a dev box and is `OSError: Read-only file system` in CI - green for whoever wrote it and red on every PR. Caught by running it in the check image before committing. - the TABLE stays in the test file and the driver goes into the tree under test, so a mutation to the tool cannot also move the answer it is graded against. 28/28 files pass. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
775f3a2e82 |
SAFETY-44: keep attribution in-tree, add the third clause, and correct a claim I made for my own precondition
checks / unit (pull_request) Successful in 6m56s
Review APPROVED |
||
|
glenn schrooyenandClaude Opus 5
|
f109153216 |
SAFETY-44: ship only what this suite can measure, and split the two clauses it cannot
checks / unit (pull_request) Successful in 6m58s
⚠️ REVIEW FOUND THE INJECTIVITY CHECK INERT IN THE SAME WAY M12's FIRST DRAFT
WAS, three paragraphs below the comment diagnosing it. Removing the whole
mechanism left 28/28 GREEN: part 5's two destinations are `sign_reason` and
`sign_tripped`, DIFFERENT bare keys, which can never contend. The
`len(_dests) >= 2` precondition was true and established nothing - "not
asserting over an empty list" was correct, "this run can exercise injectivity"
was not.
Review also measured a second, narrower silent loss: `moved_to` accepts a
destination regardless of whether the TRANSITION put the value there, so a
roster unit already latched in its own right is a free suppression slot. A build
deleting `old-a` while renaming `old-b` -> `shed` reported exit 0, NO FINDINGS,
with `old-a` "migrated" into a key that already held True.
⚠️ I WROTE BOTH ONE-LINE FIXES AND HAVE REVERTED BOTH, because THIS SUITE
CANNOT MEASURE EITHER. Three attempts, three different ways of proving nothing:
1. assert injectivity over part 5's real destinations
-> passes with the entire mechanism deleted (non-contending bare keys)
2. build a dedicated fixture for the two-claimant case
-> discriminates NOTHING: both snapshots read one state file, so nothing
vanishes and there are no migrated lines with the clause or without it
3. rebuild M12's fixture so the destination is transition-made, to reach the
provenance case
-> ⚠️ DISARMED M12 FOR THE NARROWING. All three clause-removals came back
GREEN at once, and the control I had measured RED last round was gone.
⭐ (3) IS WHY THIS IS A SPLIT AND NOT A JUDGEMENT CALL. A later fix to the same
function silently disarmed the control for an earlier one, and I saw it only
because I re-drove the removal rather than trusting the suite's exit code.
Shipping two more unmeasured clauses into the rehearsal instrument, on a card
whose entire subject is checks that cannot reach what they name, would be the
fourth inert artifact in one PR. They go on their own card, with review's
measurements as the starting evidence and the fixture built BEFORE the fix -
which is the ordering that worked for M12 and failed every time I inverted it.
⚠️ AND THE SPLIT IS SAFE, WHICH IS WHY IT IS AVAILABLE. Both remaining holes are
in the REHEARSAL TOOL, not the controller. The round-2 narrowing guarantees a
suppressed key is always OFF the roster, so no roster unit loses protection;
what is left is rollback-view/history mis-reporting on a tool an operator runs
before commissioning. Nothing here can disarm a latch on a live site, and
inverter B goes in on ~2026-09-12. Review offered this split explicitly.
M12's PRECONDITION IS STRENGTHENED, because the exit code could not carry it.
It now requires the shipped tree to be clean AND to not already report the
finding the mutation exists to produce. The first draft passed on exactly that
gap: the fixture's own litter made the run red and nothing separated the two.
Re-driven after the revert, on the round-3 fixture:
shipped exit 0
`vanished_unit not in after.units` gone ⚠️ exit 1
28/28 files pass. The controller half is unchanged since
|
||
|
glenn schrooyenandClaude Opus 5
|
f6d7bfcb04 |
SAFETY-44: give the moved-not-lost branch a control that reaches it, and make the move injective
checks / unit (pull_request) Successful in 7m38s
⚠️ THE CONTROL I WROTE LAST ROUND WAS INERT, AND REVIEW MEASURED IT. M12 ran its
mutant over `state-legacy.json` and the findings diff against the UNMUTATED tree
was EMPTY. It "passed" because that fixture's STAGED leg is already exit 1 on
pre-existing litter, and `"migrated unit/shed/" not in out` was trivially true
because a pre-DEV-02 FLAT store has no shed latch to lose - after migration only
the FIRST roster entry is ever latched.
The decisive measurement is the one it could not make: reverting the
`vanished_unit not in after.units` narrowing - restoring exactly the rule that
hides a real latch loss - left 28/28 GREEN.
⚠️ SO IT WAS HABIT 4 WEARING THE CLOTHES OF ITS OWN FIX, AND WORSE THAN THE
"NO CONTROL" IT REPLACED, because it reads as checked. That is this commit's
own argument about the M3 mis-citation, turned on the thing I wrote to answer it.
⭐ THE FIXTURE CAME FIRST THIS TIME. `state-both-latched.json` is a store at
SCHEMA_VERSION with `unit/garage/*` AND `unit/shed/*` both latched by ONE site
event, so the reasons match too - the arrangement in which same-value
suppression is most dangerous, and the one no existing fixture could produce.
M12's mutant then makes ONE roster unit's latch key VANISH while its sibling's
survives at the same value:
shipped tree, both-latched fixture exit 0, no findings
one unit's latch key deleted exit 1, ADDS
VANISHED STORE KEY unit/shed/sign_tripped
narrowing reverted ⚠️ RED (was 28/28 green)
And it asserts the PRECONDITION - the fixture is clean on the shipped tree -
plus that the mutation ADDS findings rather than coinciding with ones the
fixture already had. An empty diff there is exactly what made the last draft
inert, and nothing said so.
⭐ AND THE MOVE IS NOW INJECTIVE, which is the narrower hole review found still
open. `moved_to` is "any machine-read key with the same bare name and the same
value", and `DEFAULTS["sign_tripped"] is False` means `True` is the ONLY
non-default value a latch can hold - so N vanished keys could all claim the SAME
destination, by construction rather than by coincidence. Measured on a rename
carrying `inverter-a` -> `garage` correctly while DROPPING `inverter-b` -> `shed`:
migrated unit/inverter-a/sign_tripped = True -> unit/garage/sign_tripped
migrated unit/inverter-b/sign_tripped = True -> unit/garage/sign_tripped
Zero store-key findings, on a run where `shed` came up UNLATCHED. Two keys
cannot both have become one key. `diff()` claims each destination once.
⚠️ AND THE CHECK FOR IT IS ANCHORED TO A RUN THAT ACTUALLY PRODUCES MIGRATED
LINES. A dedicated mutant was written first and asserted over an EMPTY list -
vacuous, passing, indistinguishable from working, the same defect twice in one
sitting. These fixtures cannot produce two claimants (`migrate_unit_keys()`
moves FLAT keys only, so one flat latch yields one migration), so rather than
build a fiction to grade, injectivity is asserted over part 5's real migrations,
with a PRECONDITION that there are at least two of them. THE TWO-CLAIMANT
ARRANGEMENT IS A STATED RESIDUAL: bounded in `diff()` by construction, not
measured end-to-end.
Finding 4, and my correction to it was also wrong. The comment cited "line 315"
and "two hundred lines below"; the fix I wrote said "~54 lines down", which was
review's distance to the WRONG line, carried to a new referent without
re-measuring - inside the sentence apologising for a wrong number. Recomputed at
|
||
|
glenn schrooyenandClaude Opus 5
|
d1b7e3f113 |
SAFETY-44: decide per key, because one shared flag let an unrelated latch veto the repair
checks / unit (pull_request) Successful in 6m58s
⚠️ REVIEW DEFEATED THIS CARD'S OWN FIX WITH ONE ADDED KEY, and the project's own instrument agreed on the unmodified tree. The guard classified each key and then wrote ONE site-wide `namespaced_latched` for BOTH "on the roster" and "off-roster but not UNIT_A", so a single latched key belonging to a unit the operator had removed from `inverters:` vetoed the repair of the misdirected `inverter-a` latch beside it. Measured, roster ['garage','shed'], store stamped so the migration fix cannot reach it: sign_tripped + unit/inverter-a/sign_tripped -> [('garage', True), ...] ...AND unit/old-shed/sign_tripped -> ⚠️ [('garage', False), ...] A site DRIVING on a state file that reads `sign_tripped: true` - the sentence in this card's own first paragraph - and `tools/rehearse_roster.py` printed `LATCH CAME UP DISARMED` on it, exit 1, against shipped code. It matters more than a normal hole because this guard is, as main.py says, the ONLY repair path for an install already stamped by 0.9.0. ⚠️ AND IT IS INSTANCE #17 OF HABIT 4, IN THE EXTENSION MUTATION.md ADDED TWO DAYS AGO. The two fixtures written for this card each hold ONE arrangement - `_staged_store` has only `inverter-a`, `_other_store` only `inverter-b` - so the flag was never written twice with opposite intents in a single store. A per-key check whose keys have never disagreed WITHIN ONE STORE is a check over whichever key came last. Three variables now, one per case, and a fixture holding BOTH. Four cases driven: stranded-only repairs, +1 unrelated repairs, +2 repairs, off-roster-only does NOT (which is DEV-02's control, still green). ⭐ AND THE MOVED-NOT-LOST RULE HID A GENUINE LATCH LOSS. The match was bare key plus value over the whole of `machine_read`, with no identity - and `DEFAULTS["sign_tripped"] is False`, so the ONLY non-default value a latch can hold is `True`. Any surviving latched roster unit therefore suppressed any other unit's vanished latch. Review measured a build that drops `unit/shed/sign_*` while `unit/garage/sign_*` survives, both latched by one site event so the reasons matched too: none in LEG SAME-STEP: nothing vanished, nothing stranded ... NO FINDINGS in the rehearsed transitions exit 0 A roster unit's persisted SAFETY-02 latch deleted, transition reported clean. The legitimate case is a RENAME - the key left `inverter-a`, which is not on this roster, and arrived under `garage`, which is - so the vanished key's own unit must be OFF the roster. A key whose unit is still on the roster has not been renamed, it has been lost, and no amount of the same value elsewhere makes that untrue. ⚠️ AND THE BOUND I CITED FOR IT DID NOT EXIST. Both the comment and the last commit message said "what M3 (delete the flat rollback view) measures". M3 is not that mutant - it drops `sign_reason` from `LEGACY_UNIT_KEYS` - and the one I meant, M9, vanishes only FLAT keys, so neither can satisfy the `unit/` gate the rule sits behind. NEITHER EXERCISED THE BRANCH; it had no negative control at all, in a card whose subject is checks that cannot reach the failure they name. A citation is the worst carrier for that, because it reads as having been checked. M12 is written for it and driven: a roster unit that stops restoring its latch is exit 1, and the suppression does not swallow it. `misdirected` IS UNREACHABLE FROM THE CALL SITE, and the claim that it was a benefit is withdrawn rather than restated. `unit` is `roster_names(opts)[0]` and `build_roster()` names its first entry from the same list, so `unit` is a member of `roster` by construction; review measured it - dropping the `roster` argument inside `migrate_unit_keys()` survives 28/28. The argument stays for the reason `_verify_moves()` gives for its own `roster=None` branch, it is what a NEW caller gets - and a guard nothing exercises is a guard nobody knows works, so it is now exercised at the function's own level: a wrong `unit` comes back misdirected, the marker is not stamped, and the same call with a roster name stamps it. Dead `if _roster_names else UNIT_A` removed; `roster_names()` never returns []. THE OPERATOR MESSAGE IS ASSERTED. Review deleted it whole (`if stranded:` -> `if False:`) and 28/28 stayed green - `stranded` had no other consumer, so the list existed only to print a line nothing checked, and that line is the entire operator-facing half of the repair. Now: exactly one SAFETY-44 line at ERROR, naming the stranded key and the roster it was measured against. A wrong line citation removed: the call-site comment said "line 315's comment" and "two hundred lines below", and both were wrong (315 is the CFG-03 defaults note; the comment meant is ~54 lines down). Cite the thing, not its coordinates. ⚠️ AND WHAT 5b/5c ACTUALLY PIN IS RECORDED, because review is owed the distinction: `mutant_tree()` asserts its target exists, so EDITING the guarded line makes the file red at the ANCHOR before any behaviour is measured - which looks identical from the exit code and proves much less. It happened twice while writing this commit. The behavioural coverage lives in test_maintenance.py section 13 and test_inverter.py, which build stores rather than strings. Fixture strings no longer say "measured" - they are synthetic, and the repo's other fixture carries an explicit SYNTHETIC header. 28/28 files pass. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
5e382f52f6 |
SAFETY-44: the latch migration takes the roster, and the guard stops being defeated by the key the defect wrote
checks / unit (pull_request) Successful in 6m56s
⚠️ A DISARMED SAFETY-02 GUARD ON AN INSTALL WHOSE STATE FILE SAYS IT IS PROTECTED. `Controller.__init__` called `migrate_unit_keys(store, UNIT_A)` - a hardcoded constant - while its sibling `migrate_maint_keys(store, unit, roster)` takes the roster and refuses a write no machine reads. On a store that ran pre-DEV-02 flat and declares `inverters: [garage, shed]`, the latch moved to `unit/inverter-a/sign_tripped`, `_verify_moves()` read it back under the name it had been HANDED, was satisfied, and stamped the marker. Not one inverter came up latched, and the next start skipped the migration too. That is the MAINT-05 defect verbatim, still live in the function whose own docstring describes it, and `_grade_store_migration()` did not catch it: its "is anything namespaced latched" test was satisfied BY THE MISDIRECTED VALUE ITSELF. ⚠️ THE REASON IT WAS LEFT WAS A HALF-TRUTH. "There is no roster yet - the `Inverter`s restore their latch in their constructors." The KEYS must be written before the `Inverter`s read them; the NAMES do not come from the `Inverter`s. `roster_names(opts)` reads them off the options before anything is constructed - which is exactly what it already existed to do for `MqttPublisher` - and returns `[UNIT_A]` on the flat shape, so today's upgrade path is byte-identical. TWO EDITS, AND THEY DO DIFFERENT JOBS. Measured per leg by reverting each one: SAME-STEP leg STAGED leg both edits (shipped) clean latch armed, litter reported guard reverted latch armed ⚠️ DISARMED call site reverted latch armed, litter latch armed, litter So the GUARD carries the safety and the CALL SITE carries the hygiene and the withheld marker. The table is in test_rehearse.py part 5 because a reader who assumed the roster-aware migration re-arms the latch would delete the guard as redundant in a tidy-up and take the STAGED column with it - and STAGED is what "upgrade now, declare the roster later" produces, which is the path an operator actually takes. Only the guard reaches an install already stamped by 0.9.0. ⚠️ AND "OFF THE ROSTER" ALONE WAS TOO BLUNT, WHICH test_inverter.py CAUGHT. The first draft treated every off-roster latch as litter. In DEV-02's restart case the store carries `unit/inverter-b/sign_tripped = True` from B's own trip while the roster being graded holds only A - and the blunt rule re-armed a HEALTHY A off the flat key B's trip had mirrored. DEV-02's defect rebuilt inside its own guard, which is what the paragraph I was overruling had been right to fear. Narrowed to `UNIT_A`, which is the only name the broken migration ever wrote - `roster_names()` falls back to a POSITIONAL `inverter-N`, so `inverter-a` is not a name this code hands to anyone else. A latch stranded under any other name is a real unit out of today's config and still counts as protection. Residual, stated in the code rather than left to be found: a site whose own first unit is NAMED `inverter-a`, latched, then taken out of the roster gets re-armed on `inverters[0]` - a healthy unit held at 0 W until an operator clears it. The conservative direction, against a site DRIVING on a file that says it is latched. ⭐ AND FIXING THE CODE EXPOSED A FALSE POSITIVE IN THE HARNESS. With the latch now landing under `garage`, `rehearse_roster.py` reported VANISHED STORE KEY for `unit/inverter-a/sign_tripped` three lines above the same value sitting under `unit/garage/sign_tripped`. A key that moved to a namespace a machine reads has not vanished - and an alarm that fires on the harmless case is one an operator learns to dismiss before the real one arrives, which is CONTROL-C's own argument. The test is not "the name appears somewhere": same bare key, same value, under a key in `machine_read`, the set built by driving a real start. A latch that genuinely went missing has no such key and is still a finding, which M3 (delete the flat rollback view) still measures red. Test parts replaced deliberately, as both of them asked to be: - test_rehearse.py part 5 asserted the defect "as REQUIREMENTS so that a fix turns them red". It now asserts the fix, with 5b and 5c as negative controls - one per edit - and 5b doubles as the leg-splitter's positive control, since reverting the guard is the build where the two legs DISAGREE. - test_maintenance.py section 13 was headed "RECORDED, NOT FIXED" and said the fixing card "should expect them to go red and should replace them". Replaced, plus two cases it could not have had: the already-stamped store, and the `inverter-b` negative control for the narrowing above. 28/28 files pass. Not run: nothing - this is host-side only, no rig, no hardware. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
22ca341459 |
ARCH-02: assert the invariant, because a fixture cannot enumerate arrangements
checks / unit (pull_request) Successful in 6m55s
Review approved and then attacked part 5b, the control that stands between B1's
defect and a green suite. It got through in three edits:
A1 delete `END LEG` 5b CATCHES it 2 FAIL
A2 ...and reorder the legs 5b BLIND 1 FAIL
A3 ...and reword the closing summary 0 FAIL, "all checks passed"
with a splitter that closed
ZERO legs
⚠️ TWO OF THOSE THREE ARE ORDINARY REFACTORS - reordering report legs, rewording
a summary line - and neither reads as touching a safety check. 5b caught the
unclosed splitter only because the clean leg happened to be printed LAST, and
the backstop that saved A2 was a hardcoded transcription of the tool's own
sentence, so rewording that sentence blinded it too.
FIXED BY ASSERTING THE STRUCTURE RATHER THAN ADDING A FIXTURE: a leg may not
open while one is open, and every leg that opens must be explicitly closed.
Driven - deleting `END LEG` now fails on the FIRST edit, whatever the order or
the wording:
AssertionError: leg 'SAME-STEP' opened while 'STAGED' was still open - the
tool stopped closing its legs, and a splitter that never closes swallows the
closing summary into the last leg.
All three attacks include that edit, so one assertion kills the class. Clean
tree unchanged: 28/28, all checks passed.
⭐ AND THE RULE IN MUTATION.md IS EXTENDED, because review's finding is the
general case of it. It said "a per-thing check that has never seen the things
DISAGREE is a total". Part 5b DOES make them disagree, every run, and was still
blind. So:
...AND ONE THAT HAS ONLY SEEN THEM DISAGREE IN ONE ARRANGEMENT IS A TOTAL
OVER THAT ARRANGEMENT. When a fixture cannot enumerate the arrangements,
assert the invariant they share.
Also recorded, NOT fixed, with the reason: the reference carries `entity_id
unique_id device` and grades all three, but never the discovery TOPIC. Review
measured `homeassistant/... -> ha/...` exiting 0 with no findings while every
announced entity dies. `discovery_payloads()` already returns the topic and
`announced_rows` discards it, so a fourth column closes it - but the prefix is
HA's fixed contract rather than a name this project chooses, so it is not what a
tidy-up rename reaches. Lower LIKELIHOOD than the N1 gap it resembles, not lower
consequence, and it stops being a note the day the topic becomes configurable.
Two stale figures corrected: the tool claimed five migration mutations where
there are nine, and pointed at a DOCS.md heading that does not exist (the real
one is "Before you commission inverter B", verified present at DOCS.md:1577).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
Claude Opus 5
|
638c213944 |
ARCH-02 round 2: the pin could not tell one red leg from two, and the tool went
checks / unit (pull_request) Successful in 6m57s
quiet when a flag was forgotten Both blocking findings were reproduced before anything was changed, and both are the class this card exists to guard against. B1. THE PIN WAS A COUNT, AND A COUNT CANNOT SEPARATE TWO ROUTES. The rehearsal reaches the disarmed latch by two independent paths and prints every finding TWICE per red leg - once in the leg's block, once in the closing summary - so ONE red leg satisfied `count(...) >= 2`. Measured: shipped code 4; with `migrate_unit_keys()` handed the roster's own name (SAME-STEP fixed) 2, PASSES; with a guard-side fix (both fixed) 0. The check named "NOT THE SAME-STEP PATH ALONE" reported ok on the build whose only remaining failure IS the staged path. The tool now prints `LEG <id>` / `END LEG <id>` and the pin requires the finding in EACH leg by name. ⚠️ AND THE FIRST SPLITTER CARRIED THE SAME DEFECT ONE LEVEL DOWN - it opened on the marker and never closed, so the last leg absorbed the closing summary and read red whenever ANY leg was red. Caught by building the one fixture that separates them. Part 5b is now that fixture, run every time: a tree in which SAME-STEP is genuinely clean and STAGED is not. A per-thing check that has never seen the things disagree is a total with extra steps. ⚠️ AND IT CORRECTS A CLAIM THIS CARD MADE. `main.py:250` is NOT the whole fix. Measured on that exact change: LEG SAME-STEP clean, LEG STAGED still DISARMED - the staged path stamps `schema_version` on the flat start and the roster start's migration returns early before the name it is handed matters. DOCS.md now says so, with the measurement. B2. A FORGOTTEN FLAG WAS A CLEAN BILL OF HEALTH. Omitting `--live-entities` printed "nothing orphaned" and exited 0, with `grep -ci` for the words returning 0 - while an UNREADABLE file exited 2. The asymmetry was backwards. Omitting it now refuses (exit 2); `--no-live-entities` is the deliberate opt-out and carries a standing banner in the header AND on the exit line, because a report is skimmed from the bottom. N1. THE REFERENCE DID NOT CARRY WHAT HA KEYS ON. It held entity_ids only, so renaming `mqtt.BASE` (every unique_id moves) and changing `DEVICE["identifiers"]` (HA sees a new device) both came back exit 0 - real orphanings with every composed id byte-identical. The reference is now three columns (`entity_id unique_id device`), is loaded on EVERY run including a degraded one, and M7/M8 pin both. `object_id` is deliberately not a column and its rename is required to SURVIVE, which is what bounds the four mqtt.py mutations. Also found while fixing the above, and fixed with its own paired positive: the VANISHED check did not carry the default/non-default rule the STRANDED check does, so a pair of migrated DEFAULTS moving namespace read as two findings and CONTROL-C went red on a CORRECT build. M9 drops non-default keys and must still be a finding. N2/N3. DOCS.md withdraws the unmeasured "not recoverable by advancing again" - its own reasoning cut the other way and nothing here tested a re-advance; the measured part (eight rows stop updating, history gaps, automations stop firing) stands. And the byte-identical-input property is `test_rehearse.py`'s assertion, not the tool's. FRAMING. SAFETY-44 is ALREADY RECORDED on the merge base - `test_maintenance.py` section 13, with the "guard is defeated by the very key the defect wrote" sentence, in a file this card does not touch. DOCS.md now says this is independent re-confirmation through a second instrument, reachable from an owner-supplied state file rather than only from a fixture, and not a discovery. app/ is still untouched. 59 checks in test_rehearse.py, 9 mutations red, 2 required survivors, 3 clean controls. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
Claude Opus 5
|
cb8f800180 |
ARCH-02: the harness was red on every fresh install, and a third control says so
A store the add-on has never written has no latch and no history. The DEV-02 migration still runs on it - `schema_version` is absent, so `from_version` is 0 - and it writes `sign_tripped = False` and `sign_reason = ""` under `inverter-a` before `build_roster()` exists. On a declared roster nothing reads either of them, so the first draft called both STRANDED and exited 1. Every fresh install that grows a second inverter would have been reported as a failed transition. ⚠️ THAT IS A DEFECT IN THE ALARM, NOT A CASE TO SUPPRESS, and store.py already carries the distinction it needed: `MigrationReport.lost` reports a dropped `sign_tripped=True` and NOT a dropped `sign_tripped=False`, because "absent" and "never tripped" are the same statement - and because an alarm that fires on the harmless case is one an operator learns to dismiss before the real one arrives. The stranded check now makes the same split, on the VALUE and not on the key name: a key stranded at its default is printed under its own heading, visibly, and is not a finding; the same key stranded at a non-default still is. SAFETY-44 turns on `sign_tripped = True`, so it is untouched, and a check pins both halves against each other so the softening cannot widen into a hole. CONTROL-C is the fresh install, and it is a third clean control rather than a line in a comment: the shipped code is right about it and the harness must say nothing. It is the control that found this. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
Claude Opus 5
|
e06906d65f |
ARCH-02: rehearse the flat-to-roster switch, and it comes up disarmed
Nothing had ever run the real `Controller` through the transition inverter B's commissioning performs on 2026-09-12. `tools/rehearse_roster.py` does, over a COPY of a state file given as an input path, and reports the difference: every store key that moved, appeared, vanished or was STRANDED under a name no machine reads; every entity id before and after with orphans named; and whether the maintenance history MAINT-05 moved once survived being moved again. ⭐ WHAT IT FOUND, AND IT IS SAFETY-44's ANSWER. A store that ran pre-DEV-02 flat and declares a multi-inverter roster comes up with its SAFETY-02 latch DISARMED unless the first roster entry happens to be called `inverter-a`. `Controller.__init__` hands `migrate_unit_keys()` the hardcoded `UNIT_A`, so the latch lands in `unit/inverter-a/sign_tripped`; a roster of ["garage","shed"] reads `unit/garage/sign_tripped`, which is absent. And `_grade_store_migration()` - the guard written for exactly this - sees the migration's OWN misdirected key satisfy its `namespaced_latched` test and returns without re-arming. Measured, not argued: the file says `sign_tripped: true`, no inverter is latched, nothing is held at 0 W. The same state file onto a roster whose first entry IS `inverter-a` comes up latched, so the finding is attributable to the name. The staged upgrade reaches it too: the flat start stamps `schema_version`, and the roster start's migration then takes its early return. ⚠️ NO RUNTIME CODE IS CHANGED HERE. The fix to `migrate_unit_keys()` belongs to SAFETY-44 reopened as a defect, shown red before it; this commit is the instrument that shows it red. THE CHECK CAN FAIL, AND WAS SEEN TO. `test_rehearse.py` establishes two CLEAN CONTROLS first - two transitions the shipped code gets right, on which the harness must say nothing - and then breaks the shipped migration and the entity-id composition six ways in a private copy of the tree, requiring each to turn its control from green to red WITH THE RIGHT FINDING NAMED. One mutation is required to SURVIVE, with the reason written down. The synthetic state carries maintenance history and a legacy-shaped latched `sign_tripped` with no `schema_version` beside it, because state built by the happy path cannot detect a migration that only mishandles legacy shapes. ⚠️ THE FIRST DRAFT OF THE HARNESS HAD HABIT 4 IN IT and its own battery caught it - MUTATION.md's thirteenth instance, third in an instrument. The entity diff composed both sides with the same build, so renaming the MQTT device moved every id equally and it reported "no orphans" about a change that orphans everything. `--live-entities` is the fix: a reference that does not move with the code. THE RETREAT is supported and does not corrupt state - the flat rollback view keeps the latch and the first pack's history live - but it orphans eight HA entities, and the ROUND TRIP is not supported: a maintenance cycle completed while rolled back is discarded on re-advance. Stated in DOCS.md as a decision. Fetching /data/state.json, the options blob and the entity list are OWNER STEPS and are written down. ⚠️ Nothing here was run against real state and nothing here claims to have been. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
477911afa1 |
UX-04: the 59 was computed and then graded against "> 15"
checks / unit (pull_request) Successful in 6m22s
Review approved and found four things worth fixing. Three are cosmetic; the fourth is the same defect this card is about, one level up. ⚠️ THE FLOOR WAS PRINTED AS 59 AND ASSERTED AS `> 15`. So a manifest change dropping the required count to 20 would have passed while the sentence beside it still said 59 - a number displayed but not graded, which is exactly what let UX-01's AC 3 survive review in the first place, and exactly the shape `test_translations.py` was already fixed for. Now DERIVED and graded in two separate assertions, because they are two different claims: 1. `_EXPECT_REQ = len(schema) - len(_OPTIONAL)`, where optional is Supervisor's own rule - a `?` suffix, plus `inverters` because `UiOptions` marks a list optional - asserted EQUAL to Supervisor's required count. The printed number is computed, so it cannot go stale. 2. ⚠️ AND THE WITHDRAWAL'S PREMISE, SEPARATELY. The equality alone is not enough and I checked why: both sides read the same `?` convention, so marking a key optional moves BOTH and it stays green - measured, by making `slew_w` optional and watching it pass. That grades Supervisor agreeing with us, NOT the thing AC 3 was withdrawn for. The second assertion grades the premise: the floor is still far above the ~15 AC 3 asked for, and if it ever is not, the withdrawal needs re-arguing rather than re-asserting. Driven, both directions, against the real 2026.08.0 image: clean PASS "floor is DERIVED: 71 schema keys minus 12 optional ones is 59, and Supervisor's own required flag agrees" premise threshold 15 -> 40 FAIL "the floor of 59 is still far above the ~40 AC 3 asked for" Also: a dangling reference to `DECISIONS/UX-04-hiding-and-the-deepmerge.md`, which was never the filename shipped, and `DOCS.md` still saying section F "prints" what it now grades. Not changed, and review agreed both are disclosed rather than defects: the pair-keyed ledger cannot see a SECOND site for a pair it already holds (the site count can grow while the pair set cannot), and the probe's assertions run by hand rather than in CI because they need the Supervisor image - the same status as before this card, not a regression. Suite: 27 files, all pass. Probe: all cases passed on ghcr.io/home-assistant/amd64-hassio-supervisor:2026.08.0, --network none. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
Claude Opus 5
|
21f22b393a |
UX-04: nothing was hidden, the defaults pin was never built, and CFG-04 was closed by UX-02
checks / unit (pull_request) Successful in 6m20s
UX-01's three unmet acceptance criteria, closed out. It shipped in 0.7.0 and is LIVE, so all three describe the running install. No behaviour, no schema and no default moves here: config.yaml, translations/ and every module under app/ are byte-identical to |
||
|
glenn schrooyenandClaude Opus 5
|
1ba35f9c99 |
PKG-13: one safe.directory flag, not two, and the tighter of the pair
checks / unit (pull_request) Successful in 6m19s
Review found `-c safe.directory={where}` and `-c safe.directory=*` set on the
same invocation, the specific value redundant once the wildcard is present.
Dropped the WILDCARD rather than the specific path, which is the opposite of
the suggestion and the tighter end of it: `_git()` always runs `-C where`, so
the only repository it ever reads is the one it was pointed at, and naming that
one path grants exactly the trust the tool needs. `*` granted trust to every
repository on the machine for the lifetime of each subprocess - harmless for
read-only `log`/`show` with no tty, as review measured, but broader than the
job.
Verified after the change: the documented container invocation still runs BOTH
halves and reports `no runtime change since the last bump` / `CLEARED for
deploy`, so the fix this branch exists for is intact.
Suite: 26 files, all pass.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
glenn schrooyenandClaude Opus 5
|
0b653c35a6 |
PKG-13: the no-op half of the pre-deploy check has never run as documented
checks / unit (pull_request) Successful in 6m19s
⚠️ THE INVOCATION ON PKG-09's CARD AND IN THE TOOL'S OWN DOCSTRING RUNS ONLY HALF THE CHECK. The container runs as uid 0 against a checkout owned by the operator, and git refuses that: fatal: detected dubious ownership in repository at '/w/goodwe-addon' `_git` returned non-zero, `last_version_change()` returned None, and the silent-no-op check - THE REASON PKG-09 EXISTS - reported SKIPPED. Measured, on the documented command against release/1.0: ⚠️ SKIPPED: could not ask git what changed since the last bump; the no-op check is INCONCLUSIVE, not clean NOT CLEARED and with `--user $(id -u):$(id -g)`, unchanged otherwise: no runtime change since the last bump - nothing to deploy CLEARED for deploy. ⚠️ THE TWO HALVES WERE MUTUALLY EXCLUSIVE ACROSS THE TWO DOCUMENTED ENVIRONMENTS. In the container the validator ran and git skipped; on a bare host git ran and the validator skipped on aiohttp. NEITHER DOCUMENTED COMMAND RAN BOTH. The pre-0.9.0 clearance was sound only because it happened to be driven as two separate invocations. ⚠️ AND IT WAS INVISIBLE IN THE ONLY CASE ANYONE HAD RUN. `main()` consults the git result only when `shipped == deployed`; every real run so far compared 0.7.0 against 0.9.0 and printed its verdict without ever needing git. FIXED IN `_git()` WITH `safe.directory`, NOT BY DOCUMENTING `--user`. The tool should be correct however it is invoked rather than requiring every future caller to remember a flag; the read is read-only, on a repository the operator is about to deploy from. ⚠️ WHY THE TESTS COULD NOT SEE IT, WHICH IS THE FINDING BEHIND THE FINDING. `test_version_gate.py` invokes the gate with `sys.executable` against fixtures it creates as its own uid, so no fixture could ever meet an ownership mismatch. MUTATION.md habit 4 in a fourth shape, inside the file written to prevent it - after the pickaxe bug, after the one-bump history, after the filter-only survivor. Two fixtures added: - a directory that is NOT a git repository, so the lookup genuinely fails: the gate must exit 2 saying INCONCLUSIVE and must NOT say CLEARED. - ⚠️ THE EXIT CODES PINNED BY VALUE, NOT BY TRUTHINESS. `returncode != 0` is satisfied by collapsing NOT-CLEARED into REFUSED, which review found the earlier draft could not tell apart. 1 says "this deploy is wrong"; 2 says "nobody checked". Driven: collapsing 2 into 1 now goes red on exactly that check. Suite: 26 files, all pass. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
04b3330def |
PKG-09: refuse a deploy that would be a no-op, or that the add-on would reject
checks / unit (pull_request) Successful in 6m16s
Two ways a deploy of this add-on has gone wrong, and the Supervisor reports both as success. THE SILENT NO-OP. The Supervisor keys the built image by `version:`, so `ha apps update` against an unchanged version installs nothing and says it worked. PKG-06 recorded this on 2026-09-05 with six merges behind an unbumped version; PKG-07 recorded it again on 2026-09-06 with six. Both times a human noticed. Nothing else was going to. THE REFUSED CONFIG. A config the shipped validator rejects is SystemExit(78) - the add-on does not start. ⚠️ AND THE INGRESS PANEL GOES WITH IT, because the add-on serving it is not running, so recovery means the Supervisor UI. ⚠️ IT CAUGHT A LIVE ONE ON ITS OWN PULL REQUEST. Run against release/1.0 while this branch was in review: version 0.8.0 on both sides with UX-02's and OPT-09's runtime commits already merged behind it. The class this exists to catch had recurred, the same day, unnoticed - which is the argument for PKG-10, a non-required post-merge check, since this script only helps someone who remembers to run it. ⚠️ THE FIRST VERSION WAS BROKEN ON EVERY REAL REPOSITORY AND ITS TESTS COULD NOT SEE IT. The "when did the version last change" lookup was `git log -S`, the pickaxe, which finds commits that change the COUNT of a literal string. A bump edits the line IN PLACE, so the count never moves and it returned only the commit that first INTRODUCED the line. Measured here: -S -> |
||
|
glenn schrooyenandClaude Opus 5
|
fdaf3fb54f |
UX-02: the validator's refusals, next to the field, before the save
checks / unit (pull_request) Successful in 6m17s
The Configuration tab can only render types and ranges. `validate.py`'s most
useful refusals are CROSS-FIELD - two inverters naming one `setpoint_entity`, a
`maintenance_charge_w` above `max_w`, a flat `setpoint_entity` left set beside a
declared roster - and the Supervisor's UI has no way to express a rule spanning
two keys. Today the operator meets those as exit 78 in a log, after saving and
restarting. This puts them on the screen beside the field they name, before
anything is written.
WHAT IT ADDS
app/configui.py the field model, built from `GET /addons/self/info`'s own
`schema` + `translations`. No option key, no label, no
rule of its own. UX-01's `Group · ` label prefix becomes a
real section heading, in whichever language resolved.
app/selfconfig.py the Supervisor conversation and the write gate.
app/config_page.py the page. Its longest function decides which HTML input a
type deserves; it contains no rule.
app/web.py five routes. Still the only file a reviewer has to read to
answer "does anything here reach the inverter".
THE FOUR THINGS THAT COULD HURT SOMEBODY, AND HOW EACH IS HELD
1. A blob the SHIPPED validator refuses is never written - not warned about,
not written. An invalid blob plus a restart is exit 78, and the panel is
gone with the add-on that serves it.
2. A blob that moved underneath is refused, not clobbered. Read hands out a
sha256 of the canonical blob, save hands it back, and the stored blob is
re-read IMMEDIATELY BEFORE the write. This is SAFETY-28's discipline
applied to configuration - "a blob reaches the Supervisor only if it was
derived from the blob stored right now" - with the honest weakening stated
in the decision record: a second writer is DETECTED, not made
inexpressible.
3. Every message is the validator's own, byte for byte. test_selfconfig.py
part 6 inserts ONE new rule into a COPY of validate.py, hashes every file
under app/ to show validate.py is the only one that differs, and serves the
real /config/validate route out of that copy in a subprocess. The new
sentence arrives, filed under the key it names.
4. Nothing on the surface commands an inverter or clears a latch. Saving does
not restart: applying costs a control interruption - the wind-down commands
0 W on every inverter - so that is a separate, confirmed button that says
so first.
MEASURED, NOT CITED (tools/supervisor_selfapi_probe.py, run against
Supervisor 2026.08.0's own code; it contacts nothing)
- the three self-paths are permitted, and `/addons/self/info` is admitted by
the `default` role too while `options` and `restart` rest on the bypass
ALONE. The probe's first draft claimed the role admitted none of the three
and went red.
- `/addons/self/options/validate` is NOT reachable from inside an add-on, so
the only validator this panel can call is ours - which is the one with the
cross-field rules.
- a POST that omits a key does not preserve it: the key reverts to
config.yaml's default. Hence whole-blob read-modify-write.
- the Configuration tab CAN be removed, and doing so drops 0-of-71 options
through and takes the add-on down. The card said it could not be; it stays
by decision, which is a better reason.
DOES NOT TOUCH config.yaml, main.py or any control path. UX-01 is a dependency,
not something this supersedes: its translations are where every label,
description and section heading on this page comes from, and the Configuration
tab is still the recovery surface. DECISIONS/UX-02-two-config-surfaces.md
records which surface is authoritative (neither: the stored blob is) and what
the second surface costs.
Also corrects a stale "70 options" in three prose spots to the 71 the probe
counts - config.yaml grew a key in SAFETY-42 and the sentences did not follow.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
---
⚠️ REBASED ONTO release/1.0 @
|
||
 adminandClaude Opus 5
|
5d7e275df4 |
SAFETY-39: slew_w is a ratio to the rail, and gain is what is still wrong
checks / unit (pull_request) Successful in 5m16s
`slew_w: 1000` was derived on 2026-08-22 against a clamp of 1500 W, where "capping how far the loop can jump ahead of the inverter" was true because the rail itself bounded the excursion. The live install went to max_w 5000 and the constant did not move, so the loop crosses its whole output range in 10 s against a plant NOTES.md §5.10 measured at ~18 s. Measured on the owner's system 2026-09-05 11:43-11:46: a 3.4 kW export drove the setpoint 380 -> 5000 W in five 2 s ticks, the battery to -5154 W past a 5000 VA rating, and the grid from 3.4 kW of EXPORT to 2.8 kW of IMPORT. Nothing was broken; the rail moved. REPRODUCED FIRST, and one half of it is not reproducible at all. On plantsim.Plant and on an independent in-repo model: the setpoint stepping a full 1000 W per tick to the rail (48 of 51 export steps), and the meter crossing from export into import. The -5154 W figure is NOT reproducible in either model, because both clamp the plant at its rating - so the card's AC "battery no longer exceeds max_w in simulation" is unfalsifiable and was rewritten rather than met. THE DELIVERABLE IS THE RELATIONSHIP. SLEW_RAIL_FRAC = LOOP_TICK_S / PLANT_SETTLE_S = 2 s / 18 s = 1/9 slew_w = 0 now means FOLLOW THE RAIL - SLEW_RAIL_FRAC * max_w - exactly as integrator_max_w = 0 already means follow max_w. It is the shipped default: 222 W at max_w 2000, 556 W at 5000, 1067 W at a ~9600 W two-inverter site. validate.py rule 10 refuses an explicit value faster than that, graded against sum(unit_caps(opts)) - the roster sum, i.e. Controller.site_max_w - so a second inverter moves the bound with no operator action. VERIFIED BY BREAKING IT: restoring slew_w 1000 at max_w 5000 is refused, and the refusal names the rail, the crossing time, the measurement and what to do. Equality, one watt of slack, the live interim 400 and slew_w 0 at every rating all still start. GAIN IS THE CONSTANT THAT IS STILL WRONG, and it is NOT changed here. Over 51 export steps the over-absorption is FLAT in slew (958 W at 1000, 1081 W at 250) and monotone in gain (987 W at 0.6, 876 at 0.3, 435 at 0.2): slew_w bounds the OUTPUT, while the integrator moves gain*error per cycle and reaches a 5000 W rail in under three ticks whatever the slew is. Two control-law constants moving in one release on a running house makes the next measurement uninterpretable. ⚠️ N=2: the site rail becomes sum(max_w) ~ 9600 W (inverter B FROM DATASHEET AND SERIAL, not measured, and sim/plantsim.py's INVERTER_B is A copied), and the enforced ratio produces 1067 W there automatically. But a growing rail LOOSENS this bound, and the >= 900 W symptom goes stale at N=2 the same way the constant did - both stated in the record rather than left to be discovered. Nothing reached the inverter. The live install still runs the interim slew_w 400 applied by hand on 2026-09-05, which rule 10 accepts unchanged. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
Claude Opus 5
|
e2efc765af |
SAFETY-38 round 4: the enumeration covered one sink, and it was not the one this codebase uses
checks / unit (pull_request) Successful in 5m12s
The scanner's two censuses between them scored ONE syntactic sink and called it
the family. Every one of census B's 16 members was a sentence that is APPENDED,
ASSIGNED to a `reason`-shaped name, or passed as a LOG ARGUMENT. Not one was
RETURNED - and census A's return bucket only ever held a callee formatting its
own PARAMETERS, so a method formatting its own ATTRIBUTES and handing the string
back was invisible to both. That is the whole describe() / explain() /
reason-builder family, which is where this repository actually keeps its
operator sentences. Worse than the gap: it was in NONE of the five blind spots
the tool printed about itself.
THE SCANNER, WIDENED - and it rediscovered both members without being told:
* CENSUS C, the return sink, resolved through the returning method's CALL
SITES so a returned KEY (store.unit_key, mqtt.unit_state_key - MAINT-05's
card) is not mistaken for a returned SENTENCE. 53 returned pairs, 39
reaching an operator, 16 of those sharing a format spec.
* A sentence handed to a constructor as `reason=` or placed under a "text"
dict key is a MESSAGE. That one line of the sink walker was hiding SIX more
caller-built members, census B 16 -> 22.
* Every census B and C row now prints whether the two interpolations share a
FORMAT SPEC - the difference between a transposition that is silent and one
that merely looks broken.
* The blind-spot list is now TWO lists: what it scans (printed from the
constants, so they cannot drift) and WHAT IT DOES NOT SCAN AT ALL, largest
hole first - `%s`-style lazy logging, MEASURED at 31 calls across 8 modules
and each one listed, not asserted.
CLOSED, both zero-byte survivors of a green 21-file suite:
* clamp.py:253 ClampResult.describe(), requested_w <-> applied_w. No
test_*.py mentioned describe() at all. It is the SS15 action-log row
(main.py:1366) AND the ClampWatchdog notification reason (main.py:2452) -
the alarm SAFETY-40 and SAFETY-41 are about. Under the swap the guard
reports a setpoint clamped UPWARD, to +9999 W by a bound whose rating the
same sentence gives as 2000 W. Pinned on the ORDERED PAIR on the contiguous
literal at BOTH consumers, operands read off ctl.target and
hass.writes[-1] - what the loop asked and what the device was sent - not
off the ClampResult the swap lives in. 4 checks red.
* peakguard.py:493 PeakDecision.explain(), projected_w <-> ceiling_w. "peak
guard idle" was in no test_*.py. The row then reads "projects 5013 W under
a 2300 W ceiling" - 2713 W OVER - on a row whose own words say idle and
whose own flag says ok. Pinned on the decision and on the shipped status
row. 3 checks red.
* test_arbiter.py's scope-violation check, whose CLAIM was wider than its
ASSERTION. Measured this round rather than left at source level: 0 bytes,
21/21 green before, red now. 2 checks.
DISCLOSED, NOT CLOSED - 11 measured zero-byte survivors, each with its file
fingerprint and its 21/21 verdict, in the decision doc SS7e-SS7g. Three in
allocator.py's UnitPlan.explain(), three across arbiter.py's two explain()s,
two in tariff.py, two in arbitrage.py's trade reasons, one in main.py's
SoC-floor notification. An enumerated list with honest gaps beats a fifth round
finding a fifth instance.
CORRECTIONS THIS ROUND OWES:
* The idle sentence does NOT reach main.py:3299/3302 - those are the
floor-limited branch, where explain() returns self.reason before the idle
literal is built. It reaches main.py:3313 and, on the release edge, 1951.
1951 is disclosed unpinned.
* SS4's "Every one of the ten is red" sat under a twelve-row table: it is the
ten site_standing entries.
* SS6b said "weaker evidence than A1-A3"; A4 was measured, so SS7d's
"A1-A4 and B1" is the right list and SS6b now says it.
* AND THE PROBE THAT GRADES THESE MUTATIONS WAS THE DEFECT, ONCE. Its first
cut numbered the operands BY ORDER OF APPEARANCE, so after a transposition
the other leaf took the same number and all thirteen mutant renders came
back byte-identical - which reads exactly like "the edit is inert". Keying
the value to the NAME fixed it. Habit 4, in the measuring instrument:
MUTATION.md habit 16 is the new entry, and the count there is corrected to
eleven rather than inflated.
14 mutations measured: fresh full-repo copy each, mounted :ro, bytecode purged
and the purge COUNT printed, the edit re-read from disk, length and sha256
printed on both sides, and every one required to change the EMITTED SENTENCE
and not merely the bytes. 12 survived the pre-round suite, 1 was already
caught, 3 are caught now. Full 21-file suite green on the final tree.
NOT ONE BYTE OF app/ MOVES. This card changes coverage only.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
adminandClaude Opus 5
|
357cb4f07f |
SAFETY-38 round 3: pin the banner the SHIPPED install actually prints
checks / unit (pull_request) Successful in 5m12s
Round 2 pinned the operand order of `SignWatchdog.evaluate()` at main.py:1620 -
the SITE detector. Part 9c asserts that detector is never armed at a roster of
one, so on every install this add-on has, the PER-UNIT banner is the only
sign-disagreement sentence that can ever be printed, and its call site is a
different line. main.py:1560 was left unpinned.
1. THE SAME ZERO-BYTE DEFECT ONE CALL SITE OVER.
`evaluate(stood, against, ...)` -> `evaluate(against, stood, ...)`:
app/main.py 234 542 bytes before and after, delta 0, fingerprint
61aa413f... -> e57e6026..., bytecode purged, tree mounted :ro, line re-read
from disk. On the round-2 tree the FULL 21-FILE SUITE PASSES.
Nothing behavioural could ever have caught it. `watchdog.disagrees()` is
SYMMETRIC in its first two arguments - `commanded_w * against_w < 0` with an
abs() gate on both - so the mutant takes identical actions on every input.
Not "no fixture happens to separate them": no fixture CAN.
What moves is the sentence. Measured by diffing the suite's own printed
banners between the round-3 tree and the same tree carrying only this edit:
9 emitted banner lines invert, in 3 distinct sentences.
shipped commanded -2000 W against a meter reading +2000 W
swapped commanded +2000 W against a meter reading -2000 W
shipped commanded +1000 W against its own measured power -1000 W
swapped commanded -1000 W against its own measured power +1000 W
shipped commanded -2000 W against a meter reading +4000 W (the N=1 arm)
swapped commanded +4000 W against a meter reading -2000 W
Persisted to `sign_reason` and re-read after the restart the latch exists to
survive; also the action log, `_LOG.error` and `status_payload()`'s headline.
CLOSED by the ORDERED PAIR on both per-unit banners, both operands read off
the OBSERVED CYCLE - `inv.last_write` and `inv.batt` off the objects the
plant moved, never off the arguments the detector was handed, which would
compare the swap against itself:
- part 3, the N>=2 trip: "commanded {stood:+.0f} W against its own
measured power {meas:+.0f} W", plus the persisted namespaced copy;
- part 7, the N=1 controller-driven trip: "commanded {stood:+.0f} W
against a meter reading {grid:+.0f} W", plus the namespaced key AND the
flat rollback mirror;
- a PRECONDITION on each that the two operands are different numbers of
opposite sign - without it a transposition produces the same string and
the check is decoration.
Measured after the fix: RED on four named checks in test_watchdog.py, GREEN
in the other twenty files. M11 (the site swap) re-measured on this tree:
still red. Round 1's and round 2's kills are intact.
2. A FALSE STATEMENT ROUND 2 INTRODUCED, AND IT WAS WORSE THAN REPORTED.
Section 4 said "three of these left the whole suite green - M1, M2 and M3 -
and the remaining seven were caught by SAFETY-30's own N>=2 checks". Round
1's vaguer wording was harmless; naming three made it specific and wrong.
Re-measured mutation by mutation on clean release/1.0 @
|
||
|
Claude Opus 5
|
fef6ef20a4 |
UX-01: translate the add-on config, and order it by task
checks / unit (pull_request) Successful in 5m1s
The add-on had NO translations/ directory, so Home Assistant fell back to rendering the raw option KEYS. An operator opening Configuration read `unit_sign_disagree_w`, `saturation_cycles`, `integrator_max_w` - seventy of them, in file order, with no description and no way to tell a constant derived from 1890 closed-loop test runs from their own inverter's rating. ⚠️ PRESENTATION ONLY. Not one default, not one schema range and not one line of behaviour moved. The 175 non-comment lines of config.yaml are byte-identical to release/1.0's, reordered; the only comment lines that changed are the group banners and five sentences whose "above"/"below" the reorder made false. * translations/en.yaml and translations/nl.yaml: a name and a description for all 70 options and all 7 per-entry keys of the `inverters:` repeatable section. The Dutch is written for a Flemish homeowner - afname/injectie, kwartierpiek, capaciteitstarief - not transliterated from the English. * ⭐ Every derived constant says WHERE IT CAME FROM and whether to touch it. `unit_sign_disagree_s` cites its 1890 closed-loop runs (SAFETY-30); `circulation_w` its 924 runs and the measured 400 W knee (SAFETY-31); `meter_freeze_s` its 2399 s capture (SAFETY-18); `bms_soc_floor` register 0x0560. The two numbers that are the operator's - max_w and peak_cap_w - say "THIS IS YOUR NUMBER". `cycle_cost_threshold_eur`, which is neither, says it is an open product question (OPT-08). * config.yaml reordered into task order in BOTH halves, comments carried with their keys: hardware -> meter -> ratings and limits -> tariff -> maintenance -> safety -> advanced. ⚠️ test_inverter.py:1788's trap is PRESERVED: the nested per-entry `max_w:` still precedes the flat one under `schema:`, and `inverters: []` still ships in the `options:` half. Moving the nested block below the flat one is a 1-byte edit; it was run and goes red in test_inverter.py and test_translations.py. * tools/supervisor_shape_probe.py section F MEASURES the result against ghcr.io/home-assistant/amd64-hassio-supervisor:2026.08.0 - Supervisor's own _read_app_translations(), its own UiOptions, and the frontend's label resolution transcribed from the shipped bundle - and prints every rendered row in both languages. `all cases passed`, exit 0. * ⚠️ ONE MEASURED NEGATIVE RESULT, NOT GLOSSED: the per-entry keys inside the `inverters:` section still render RAW on frontend 20250925.1, because a list-of-objects goes to ha-selector-object and the converter sets no `label` and no `translation_key`. The `fields:` translations are shipped and validated - that is Supervisor's own shape for this - but this card does not claim they reach the screen. The `inverters` label and description do. * test_translations.py: a translation naming a non-existent option, or an option with no translation, fails CI - both directions, both languages, top level and per entry. Mutation-tested BOTH WAYS: eleven in-process mutations each required to NAME the key they broke, plus a negative control; six more run on disk through the purge-and-re-read protocol. Two of the eleven are the ones that cost the WHOLE FILE rather than one label - an unrecognised top-level block, and a `network:` value that is not a string. Supervisor drops a file that fails SCHEMA_APP_TRANSLATIONS with only a log warning, so those silently return the entire tab to raw keys; neither is reachable through the `configuration:` coverage, which is why they were hunted for and closed. * tools/translations_lint.py is the hand parser CI needs (no PyYAML in the add-on image) and the probe grades it, key by key and string by string, against Supervisor's own loader. It is strict and RAISES rather than skipping; nine malformed shapes prove it. Full 20-file suite green (19 existing + test_translations.py). Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
adminandClaude Opus 5
|
19decf433f |
SAFETY-32 round 5: stop closing sites, build an instrument that finds its own scope
checks / unit (pull_request) Successful in 4m59s
Rounds 2, 3 and 4 each closed exactly the survivor they were handed and each time
the reviewer found the same shape one step over: the reduction, the call site, the
sibling argument, and now the sibling MODULE. Round 4's enumeration stopped at the
`Controller` class boundary; round 5's four blockers were reductions in validate.py,
over a list DERIVED from the roster three calls away from it. Closing four more sites
by hand would have produced a fifth rejection, so this round widens the instrument
instead and lets it enumerate.
⚠️ THE BLOCKERS ARE NOT EQUIVALENT MUTANTS - a safety refusal disappears, verified by
calling the shipped validator both ways on the real tree:
roster max_w 5000 (inverter-a) beside 2000 (inverter-b), sign_disagree_w 3000
clean -> 1 problem, "...max_w (inverter-b) is 2000 W, so the
SAFETY-02 sign-disagreement watchdog can never fire"
min(caps[:1]) +4 B -> 0 problems. The site STARTS.
caps_ix[:1] +4 B -> 0 problems.
the SAME roster with the small cap FIRST (2000, 5000)
clean -> 1 problem, naming inverter-a
min(caps[-1:]) +5 B -> 0 problems.
⚠️ AND THE ONE-ROSTER FIXTURE COULD NOT HAVE SEEN THAT. On [5000, 2000] the `[-1:]`
mutant still refuses, because last IS min there. A fixture of N members cannot separate
two reductions that agree on N members - which is why the behavioural half below is a
SET and not a roster.
THE INSTRUMENT (test_peakguard.py part 12j, widened past `Controller`)
Every `.py` under app/ is read off disk and parsed - NO MODULE IS NAMED IN THE CHECK -
and the roster is found by transitive closure from ONE seed, the `inverters` option key:
through returns, through lists ACCUMULATED in a loop over a roster (which is exactly
what `caps`, `roster` and `problems` are, and what round 4's rule could not express),
through attributes and properties, and THROUGH CALL ARGUMENTS, which is how it walks
into peakguard, arbiter, allocator, circulation, maintenance and store without being
told they exist. A slice of a roster stays a roster so a mutation cannot cut the taint;
an index is one unit and does not.
183 roster reductions across 9 modules, out of 23 files parsed:
main 96, allocator 24, validate 19, arbiter 14, circulation 10, maintenance 10,
peakguard 4, inverter 3, store 3. 14 modules reduce over no roster at all.
⚠️ THE HAND-WRITTEN MODULE LIST WOULD HAVE BEEN WRONG IN BOTH DIRECTIONS. It named
watchdog.py, which reduces over no roster at all (SAFETY-30's sign watchdogs are
strictly per-unit), and omitted maintenance.py and store.py, which both do. A vacuous
claim about one module and no claim at all about two others - which is the argument for
computing the scope, made with data.
The pin rediscovers ALL FOUR blockers without being told where to look, and finds a
FIFTH beside them (validate.py:783, rule 5's own message) that no review round named.
Its counterfactual runs at two depths: every enumerated site's expression mutated with
each of the four cuts this card has been defeated by (183 x 4 = 732 spellings, all
reported), and `[:1]` written into app/validate.py at each blocker position with the
WHOLE analysis re-run over the mutated source.
THE SWEEP, AND THE NUMBER THAT MUST NOT BE QUOTED ALONE
All 183 sites mutated to `[:1]`, +4 B each, against the full 19-file suite in
goodwe-addon-ci (:ro, -B, fresh rsync per mutant, 0 .pyc before and after - a read-only
mount cannot write one): 183/183 RED. ⚠️ That is TAUTOLOGICAL on its own, because the
pin makes every slice red. So the 51 whose first red was test_peakguard.py were re-run
with the pin NEUTRALISED:
145 of 183 a behavioural fixture separates the site from a slice of itself
38 of 183 ONLY the structural claim. Behaviourally UNGRADED.
All 38 are listed by site with what a `[:1]` there would cost, in the DECISIONS record
and in part 12j itself: 15 are rows and counts on the operator's page, 3 are
`len(targets)` arithmetic beside a write path that IS graded, 6 are allocator internals
ALLOC-01 reaches but not with enough units to separate. NONE of the 38 is closed
behaviourally by this card and none is counted as covered.
THE BEHAVIOURAL HALF, because a shape pin grades an identity and never a meaning
test_validate.py grows a ROSTER SET for the three rules that reduce over
`unit_caps_indexed()`: widths 2, 3, 4, 5 with the one small cap at EVERY position,
14 rosters x 3 rules = 42 runs of the real validator, each asserting the refusal fires
AND names the unit the number came off. Beside it a COMPUTED family table of twelve
candidates - max, first, last, the mean, [:1], [-1:], [:2], [:3], [1:], [:-1], [::2] -
each compared against what the shipped validator actually answered on the same roster.
Only `min` agrees everywhere; a member no roster kills is printed as a NAMED SURVIVOR.
⚠️ A THIRD RULE WAS READING THE FLAT `max_w`, AND IT IS SAFETY-30's
Found by reading the three branches side by side, NOT by the pin - and said that way
because it is the shape of what the pin cannot see. `unit_sign_disagree_w` was graded
against `_number(opts, "max_w", 2000.0)`, three lines under a branch that had already
stopped doing that. Both directions were measured before the fix:
two 2000 W units + a leftover `max_w: 5000` + unit_sign_disagree_w 3000 -> 0 problems.
The site started, the status page said `armed`, and the per-inverter watchdog could
not trip on EITHER unit. Identical sentence to the blocker, one branch over.
two 5000 W units with NO flat key -> REFUSED against a 2000 W default the operator
never wrote, for a threshold both units can reach.
It takes the same `weakest` the branch above takes and names the same unit. At a roster
of one `unit_caps()` is `[max_w]`, so the shipped single-inverter refusal is the same
comparison, the same boundary and the same sentence.
MUTATION PROOF (each: bytecode purged and counted 0 before and 0 after, the edit
re-read FROM DISK with the new text present and the old absent, byte delta printed)
B1 validate min(caps) -> min(caps[:1]) +4 B RED test_validate.py
B2 validate min(caps) -> min(caps[-1:]) +5 B RED test_validate.py
B3 validate caps_ix -> caps_ix[:1] +4 B RED test_validate.py
B4 validate rule 6 min -> min(caps_ix[-1:]) +5 B RED test_validate.py
B5 validate rule 5 msg -> min(caps_ix[:1]) +4 B RED test_validate.py
B6 validate rule 5b msg -> min(caps_ix[-1:]) +5 B RED test_validate.py
B7 CONTROL: weakest -> max(0.0, weakest) +10 B GREEN, as intended
R3a peakguard max(floors) -> min(floors) 0 B RED test_peakguard.py
R3b peakguard max(floors) -> max(floors[-2:]) +5 B RED test_peakguard.py
R3c main call site packs -> self.packs[:2] +4 B RED test_peakguard.py
R4a main pack_floor_soc -> vals[:2] +4 B RED test_peakguard.py
R4b main pack_floor_soc -> vals[-2:] +5 B RED test_peakguard.py
R4c main pack_floor_soc -> vals[:3] +4 B RED test_peakguard.py
R4d main cycle() any(...) -> self.inverters[:1] +4 B RED test_watchdog.py
B1-B6 are caught by the BEHAVIOURAL set in test_validate.py, before the pin is reached.
B7 is a deliberate no-op control: `max(0.0, weakest)` is `weakest` for weakest > 0, and
it stays green, so the battery is not merely red for everything.
R3a is a ZERO-byte delta, so the shipped function was called both ways as well:
packs floors [15.0, 30.0, 20.0], peak_soc_floor 15
binding_soc_floor(...) = 30.0 the 0-byte mutant would answer 15.0 (+15.0 pp)
-> the site would discharge to 15 % while pack-b's own BMS cuts out at 30 %.
⚠️ PRODUCTION IS DEGENERATE ON THE FLOOR AXIS. SPIKE-04: pack A and pack B BOTH floor
at 10 %, so a differing floor exists ONLY in the fixtures. That is the argument for
keeping them, not for simplifying them - collapsing them deletes the only data in this
repository that can see the reduction at all. Restated in config.yaml's example, in the
probe and in the fixtures.
⚠️ INVERTER B DOES NOT PHYSICALLY EXIST (~2026-09-12). Every second-unit figure here -
4600 W FROM ITS DATASHEET, 15 % illustrative, the whole roster SET - is SIMULATED and
labelled. Only inverter A's 10 % is read off hardware.
REBASED onto
|
||
|
adminandClaude Opus 5
|
f9e7fbf86d |
SAFETY-32: bms_soc_floor per pack, rule 8 per unit
`bms_soc_floor` describes A BATTERY, not a site. At two packs of different models the cut-outs differ, and one number for both graded the shallower pack against the deeper one's BMS - so a site with cut-outs of 10 % and 30 % under `peak_soc_floor: 20` STARTED, and the guard would command a forced discharge that pack B's BMS, not the operator's floor, is the thing that stops. That is rule 8's own failure mode, reached through rule 8. Scoped to the card's third bullet, per its PM PROOFREAD: the aggregate anti-windup (DEV-02) and per-instance `achievable_w` (CFG-04) were verified as already shipped and are untouched. * config.yaml: `bms_soc_floor: int(0,50)` REQUIRED per `inverters:` entry. Nothing is inherited; the flat key stays the single-inverter shape's own cut-out. A per-entry DEFAULT was declined - a default is one number for every pack, which is this card's defect - and the decision is recorded. * inverter.py: `Inverter.bms_soc_floor`; an entry that states none gets None, not a borrowed number and not 0.0 (the most permissive value the key has). * peakguard.py: `Pack`, `pack_floor_pct()`, `pack_permissions()`, `binding_soc_floor()`. `discharge_permitted()` is unchanged and is still the one place the question is answered. * validate.py: `unit_bms_floors_indexed()` carries the ROSTER index; rule 8 grades EVERY unit and NAMES the one it refuses. Graded with the offender SECOND and with the same two numbers in the other order. * main.py: `Controller.packs`; `decide(soc_floor=...)` is the per-pack reduction of the same option; the floor-limited notification names the pack that stopped the site. `pack_floor_soc` - the emptiest pack, and its None-on-missing behaviour - is untouched. * tools/supervisor_shape_probe.py: RE-RUN and repaired. The required key broke §C and left §D printing PASS on a refusal about a different key. §D is now an isolated counterfactual over an accepted merged baseline, one key varied, asserting the key name. `all cases passed` on 2026.08.0. Stated rather than hidden: under any config rule 8 accepts the per-pack FLOOR reduces to `peak_soc_floor` for every pack. The reduction is written for `unit_caps_indexed()`'s reason and graded on both sides - 1775 rule-8-valid pairs answer the option, the pair rule 8 refuses answers the cut-out. `peak_soc_floor` is still one number for the site, so the "deeper pack needlessly withheld" half is filed as an open question, not closed. Full 19-file suite green. Twelve source mutations through tools/mutation_run.sh: eleven red, one argued survivor. The battery also found `_flat_bms_floor()` defaulting in two places, each covering for the other. No measurement of inverter B appears anywhere: it does not exist yet. Every second-pack figure is simulated and labelled. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
a9993aeda2 |
SPIKE-04a: inverter B is 4600 W FROM ITS DATASHEET, and the copies are pinned
checks / unit (pull_request) Successful in 4m51s
The desk half of SPIKE-04. No control behaviour changes; this is documentation and the check that stops it drifting again. Three places stated inverter B's rating - config.yaml's commented roster example, DOCS.md's, and tools/supervisor_shape_probe.py's "a blob an operator would POST" - and all three said 3000 W, a number nobody had measured or read off anything. The owner supplied B's model and serial on 2026-09-04: a GoodWe GW5048D-ES, serial `55048ESU22AW0992`, max charging and discharging power 4600 W. Serial characters 6-8 are `ESU`, platform 105 / AA55, so SPIKE-01 does not gate this site. All three copies now read 4600. ⚠️ 4600 W IS FROM A DATASHEET AND IS NOT A MEASUREMENT, and the one ambiguity is recorded rather than guessed. The datasheet prints it under BATTERY INPUT DATA, so it is probably DC, while max_w is graded against 0x0518 - the inverter's own AC power. If it is DC the AC ceiling is nearer 4324 W. Neither reading is written in as fact; SAFETY-20's min(max_w, dev_max) already bounds it correctly under either, once dev_max is read off the unit, so `max_w: 4600` is what an operator should configure today. goodwe/NOTES.md §1.1b carries the full record with the date and the method on every figure. A NEW CHECK, BECAUSE A DOCUMENTATION FIX WITHOUT ONE DRIFTS AGAIN. test_validate.py now pins all three documented rosters: findable, in agreement, and equal to 4600 - with an explicit len(...) == 3 guard so that a regex which stopped matching cannot leave "they all agree" green on the two sites still read. It also grades the PROVENANCE beside the number (the datasheet, the 0x0518 it is graded against, and the 4324 W the DC reading would make it) WITHIN 2500 characters of the roster rather than file-wide. ⚠️ The first version of that provenance check searched whole files and was ALREADY VACUOUS on two of them: 0x0518 appeared once in config.yaml and once in DOCS.md before this card, for unrelated reasons, so `"0x0518" in src` was green no matter what happened next to B's rating. That is a check whose data cannot reach the failure it names, in the file that exists to complain about them. It is fixed, and the window is what fixes it - deleting the caveat beside the number now goes red even where the same token survives elsewhere in the file. The windowed version immediately found that the DOCS.md caveat did not name 0x0518 at all; DOCS.md now says it. Four mutations, bytecode purged and counted at 0 before and after each, all RED (two of them 0-byte deltas that change a printed answer): M1 probe's operator blob 4600 -> 3000 0 B RED (value + agreement) M2 DOCS.md's DC/AC caveat deleted -157 B RED (provenance) M3 config.yaml alone 4600 -> 4324 0 B RED (agreement) M4 probe entity renamed, regex finds nothing RED (the vacuity guard) Restored, green again. The probe itself was re-run against Supervisor 2026.08.0's own AppOptions validator with the 4600: all cases passed, exit 0. ⚠️ THE TEST FIXTURES KEEP 3000 W ON PURPOSE AND NOW SAY WHY AT THE LINE. test_inverter.py, test_maintenance.py, test_peakguard.py and test_watchdog.py build a second inverter at 3000 W against A's 5000 W. The only property those checks consume is that B's number is unmistakably not A's, and the real pair is a 400 W gap - narrower than slew_w, inside the range where an off-by-one-unit bug and ordinary control arithmetic produce indistinguishable numbers. Collapsing them onto the real values would buy realism those checks do not use and cost them their separating power, which is exactly what test_peakguard.py's own §10g PRECONDITION exists to prevent. Their comments no longer call 3000 W a hypothesis about B's rating; they call it illustrative, and not a placeholder either - the placeholder era ended on 2026-09-04. Also recorded in the changelog for the cards that own them: 94 % battery-to-AC => ~88 % round trip => ~360 W burned continuously to circulate 3 kW, which is the number SAFETY-31 argues its threshold from without having one; < 13 W night self-consumption per unit, which nothing accounts for; and the A/B asymmetry in opposite directions (~33 % more energy, ~8 % less power), which means MAINT-05's 12 h drain timeout was derived on A and should be re-checked against B. ⚠️ Honest survivors: - goodwe/NOTES.md §1.1b, where all of this is actually recorded, is graded by nothing here. Rewriting its 16 occurrences of 4600 to 3000 and turning every FROM DATASHEET label into MEASURED ON THE UNIT leaves all 19 files GREEN. This CI cannot fetch admin/goodwe - private, and GITHUB_TOKEN is minted for this repository only, as .gitea/workflows/checks.yml already states - so the cross-repo check is not buildable here. - No check can see whether 4600 is the RIGHT number. Everything above pins a datasheet reading against itself. Only the bench half of SPIKE-04 can say whether the unit agrees, and the DC-vs-AC ambiguity is the live example. - The constant is still named B_RATING_HYPOTHESIS in four test files, which reads as a hypothesis about B's rating when it is now an illustrative fixture. It is left alone deliberately: renaming ~20 sites across test_peakguard.py and test_inverter.py would collide hard with the open SAFETY-32 pull request, which rewrites 1255 lines of the first. The comment directly above each definition says what the value is and is not. Suite: 19/19 green in goodwe-addon-ci, mounted :ro, bytecode purged and verified 0 before and after. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
22c016fadd |
CFG-04 fix round: the probe now measures the key it is cited for
checks / unit (pull_request) Successful in 4m33s
Review blocked this on ONE finding, and it was in the EVIDENCE, not the code.
The design and the shipped behaviour are unchanged.
THE INSTRUMENT DID NOT MEASURE WHAT FOUR DOCUMENTS SAID IT MEASURED
tools/supervisor_shape_probe.py section B validated the RECONSTRUCTED LIVE
options blob - which is missing many required keys that have nothing to do with
this card. Its refusal therefore named whichever of THOSE the validator reached
first: the committed transcript recorded `meter_freeze_s`, the reviewer's run
printed `p1_phase_export_entities`, and a run today printed `sign_disagree_w`.
The string those documents quoted - `Missing option 'inverters' in root` - was
never produced by the instrument.
And section B asserted a bare `not ok`. Any refusal at all satisfied it, so
deleting `inverters: []` from config.yaml left it printing PASS: the probe could
not detect the one regression it is the committed evidence for.
THE UNDERLYING CLAIM IS TRUE AND IS NOW MEASURED PROPERLY
Section B is an ISOLATED COUNTERFACTUAL now: the full shipped `options:` block
(expect accept), and the SAME block with `inverters` removed and nothing else
varying (expect refusal). One key differs, so the refusal has nowhere else to
come from - and the assertion is on the KEY NAME IN THE REFUSAL STRING, so the
line goes red if the refusal is ever about a different key. The probe also exits
non-zero when any case fails; without that a transcript is only evidence to
someone who reads it.
The old live-blob case survives as section B2, relabelled for what it is: an
incomplete reconstruction whose RAW half is refused for unrelated missing keys,
kept because its MERGED half genuinely demonstrates what App.options is.
VERIFIED BY BREAKING IT, WHICH IS THE WHOLE POINT
`inverters: []` deleted from config.yaml, probe re-run against
ghcr.io/home-assistant/amd64-hassio-supervisor:2026.08.0:
== B. THE LOAD-BEARING ONE: an options blob with no `inverters` key ==
FAIL the shipped `options:` block validates AS IT STANDS - ...
Missing option 'inverters' in root in GoodWe RS485 Controller
...
== 2 CASE(S) FAILED ==
exit 1. Key restored, same image, same command: `== all cases passed ==`, exit
0. The transcript in DECISIONS is that restored run, re-run today, not the old
one.
THE FOUR MIS-CITATIONS
DECISIONS/CFG-04-inverter-roster.md section 4's "section B shows both halves"
sentence and its line-103 transcript; CHANGELOG.md; test_inverter.py's part-9
comment and its part-12 mutation row; and the PR description. All now attribute
the string to section B and say plainly what section B2 does not show. The
part-12 row additionally records that the probe was blind to that mutation until
review, because a mutation table that lists a defeat nothing catches is the
failure this repository keeps finding.
MUTATION.md gains it as the sixth instance of "the check's data cannot reach the
failure it names" - the first found OUTSIDE the suite, in a committed evidence
artifact, because nothing in that document had ever been pointed at tools/. Three
habits with it: assert on the CONTENT of a refusal rather than that one happened;
vary ONE thing, from an ACCEPTED baseline; and treat a committed transcript as a
check subject to the mutation battery.
TWO NON-BLOCKING FINDINGS, ALSO FIXED
validate.py rules 5 and 6 NAMED THE WRONG INVERTER. `unit_caps()` SKIPS an
entry with no usable `max_w`, so caps position i stops being roster position i
the moment one entry is malformed - and rule 5 fed `caps.index(weakest)` to
`_unit_label()`, which indexes the roster, while rule 6 paired `caps[0]` with a
hard-wired `_unit_label(opts, 0)`. Right number, another inverter's name: an
operator sent to the wrong device by the message that exists to stop that. New
`unit_caps_indexed()` carries the roster index with the cap. It is latent -
rule 9(d) refuses such a config in the same pass - but that was a property of
today's rule ordering, not of the code, and validate_startup_config() is also
sim/inject.py's entry point. Part 10 gains a check that asserts the fixture
really does desynchronise the two lists, then that both messages name the right
unit, then the counterfactual: the OLD expressions, on that same config, still
evaluate to the wrong names.
DOCS.md called a downgrade "safe". True of the SCHEMA - the older schema drops
the key with a warning - and incomplete about the ADD-ON: rule 9(e) requires the
flat `setpoint_entity` be cleared when a roster is declared, so a config that
downgrades cleanly is by construction one whose flat actuator is blank, and the
older version has nothing to write to. Confirmed by running release/1.0's own
validate_startup_config() on that blob: one refusal, naming setpoint_entity.
Recoverable and loud, but an outage until a field is refilled. The note says so
now, and tells the operator to write the entity down before downgrading.
19/19 green in the goodwe-addon-ci image.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
adminandClaude Opus 5
|
6b85acb0f1 |
CFG-04: an operator can declare a second inverter, and the shape was measured
checks / unit (pull_request) Successful in 4m30s
DEV-02 made every per-device value per-instance - rating, SoC, watchdogs, heartbeat, store keys - so the code stopped assuming one inverter. The options schema did not. `build_roster()` returned a list of exactly one and said so in its own docstring, and there was no `inverters:` key for an operator to write, so the roster could only ever be a roster of one. THE SHAPE, AND WHY IT WAS MEASURED RATHER THAN CHOSEN config.yaml had no list-of-objects anywhere, and `static_profile` carries a whole tariff structure as a JSON string in a `str?` precisely because a nested schema was doubted. That is a question about somebody else's software, so it was answered by running it: `tools/supervisor_shape_probe.py` (committed) runs Supervisor 2026.08.0's OWN `AppOptions` validator and `SCHEMA_APP_CONFIG`, out of the release image, against this repository's config.yaml. 2026.08.0 is the version the live install reports. The full transcript and the argument are in DECISIONS/CFG-04-inverter-roster.md. It accepts the list, enforces per-entry types, refuses a missing `name` / `setpoint_entity` / `max_w` BY NAME, renders the option in the UI as a repeatable section, and drops the key with a warning under an older schema - so a downgrade is not the forward trap in reverse. ⚠️ AND THE `inverters: []` DEFAULT IS LOAD-BEARING, WHICH IS THE MEASUREMENT THAT MATTERED MOST. Supervisor answers `Missing option 'inverters' in root` for an options blob that has no such key - which every existing install's has, since the key did not exist when they were written. `App.options` is the shipped `options:` block deep-merged with the operator's stored one, so the empty default is the difference between an upgrade nobody notices and an add-on that cannot start with a schema pinned to the version that would fix it. That is docs/deploy-rollback.md §4's trap in the FORWARD direction, found before it was sprung. The merger was checked too: lists override rather than concatenate, so a non-empty default could not have grown a roster behind the operator's back. NOTHING IS INHERITED, WHICH IS DEV-02 RESTATED AS A CONFIG RULE `name`, `setpoint_entity` and `max_w` are required of every entry. An entry that omits `soc_entity` has no SoC sensor; it does not borrow the flat one. A three-line fallback would have rebuilt the sibling-grading defect DEV-02 spent a card removing, in YAML. Defaults still live in exactly one place (CFG-03): there is no per-entry default anywhere, because there is no optional per-entry number. A flat `setpoint_entity` left set beside a declared roster is refused rather than ignored - it names an actuator nothing will write. TWO PERMISSIVE READERS IN SHIPPED CODE, AND THREE MORE IN THE CHECKS validate.py rules 5 and 6 compared their thresholds against the FLAT `max_w`. At a roster of two 2000 W units with `max_w: 5000` left over from a single-inverter install, `sign_disagree_w: 3000` passed rule 5 while SAFETY-02's watchdog could fire on NEITHER unit - rule 5's own failure mode, reached through rule 5. They read `unit_caps()` now, which at one inverter is `[max_w]` exactly. main.py's `run_control()` re-ran the device-rating check behind `if self.o.get("setpoint_entity")` - the flat key, which a roster leaves blank - so on a two-inverter site no inverter's `max_w` was graded against the device it is written to, and the skip was silent. `validate_unit_ranges()` grades each unit against its own ESPHome and runs unconditionally. test_peakguard.py's `read_option()` and `_schema_bound()` and test_inverter.py's `read_schema_max()` matched a STRIPPED line and took the first hit, so the nested per-inverter `max_w:` began outranking the flat one - silently, and correctly, because the two bounds are the same int(100,5000) today. A reader that is right by coincidence is this project's named failure mode. All three are anchored at exactly two spaces now; the file's two halves disambiguate `peak_cap_w` instead of first-match, and an ambiguous key raises rather than picking one. test_tariff.py's copy of the idiom went with them. THE AUDIT IS A CHECK, NOT A PROMISE test_inverter.py part 11 finds every config.yaml-reading function in the suite - transitively, because two of the readers CFG-04 had to change never name the file - EXECUTES each against a synthetic two-inverter config.yaml whose per-inverter numbers differ from its flat ones, and fails if any answers with the per-inverter value. The table is not maintained by hand: the signature decides, so the eleventh reader is named on its first run. config.yaml declares `inverters:` BEFORE `max_w:` on purpose and part 11 asserts that ordering, so the trap stays baited. SAFETY-28's `actuator_writers()` needed no change - it was already a structural, case-folding census - but until now there was no configuration in which two writers on one actuator could be WRITTEN DOWN. There is now, and it is refused. VERIFIED BY BREAKING IT Ten mutations run through tools/mutation_run.sh and observed red, listed in test_inverter.py part 12 with the check that bites each. Two survived the first run and the CHECKS were changed, not the table: the inheritance check had blank flat keys, so "inherited" and "not inherited" were the same empty string; and the `run_control()` check asserted a SOURCE LINE was present, which re-indenting it does not remove. Both are recorded in MUTATION.md's "the check's data cannot reach the failure it names" section, which now carries two more habits: assert on behaviour rather than on source text, and give a check whose failure mode is a hang a deadline. 19/19 green in the goodwe-addon-ci image. The one-inverter path is byte- identical: the whole existing suite runs unmodified against it. Version bumped to 0.6.0 - the Supervisor keys the built image by `version:`, so a schema change without a bump is a schema nobody reads. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU |
||
|
glenn schrooyenandClaude Opus 5
|
e306c74cb8 |
DEV-02: per-instance inverter objects, so nothing is graded against a sibling
checks / unit (pull_request) Successful in 4m15s
Every piece of per-device state in the Controller was a scalar, and several of
them are safety inputs. app/inverter.py introduces an `Inverter` that owns its
entities, its rating, its device range, its SignWatchdog, its last_write and
its last_beat, under store keys namespaced by unit name. The Controller holds a
list of them and keeps only what is genuinely site-level: the P1 reading, the
site integrator, and the tariff quarter.
WHAT THE SCALARS WERE SILENTLY ASSERTING
dev_min/dev_max one pair, read from one setpoint_entity. A second
inverter's rating was NEVER READ AT ALL, so its clamp, its
freeze rail and the peak guard's achievable_w were the
first one's numbers.
soc one reading into the maintenance FSM, SAFETY-24's pack
floor and arbitrage's bounds. An average lets the emptier
pack be driven below its own floor.
sign_wd ONE SignWatchdog, whose own docstring says "one instance
per actuator", latched under FLAT store keys - so a trip
caused by B latched A across restarts and one operator
press cleared both.
last_beat one deadline and one serialised hass.call(), so a hung
write to A ate B's 30 s layer-1 budget.
actual_w BECAME SITE LEVEL WHILE THE READINGS BECAME PER-INSTANCE, and those
are opposite directions on purpose. compute()'s anti-windup test is
`abs(prev_w - actual_w) > saturation_w` with prev_w the SITE target, so feeding
it one inverter's power compares a site number against a unit number, trips the
saturation freeze permanently and leaves a standing grid error. Measured on
admin/goodwe's two-master rig (tools/alloc_sim.py --splits, 2 kW load, 40
cycles), worst |delta| between splits at the meter:
gain one unit's power the site sum
0.2 510 W 0 W
0.4 250 W 0 W
0.6 50 W 0 W
That was ALLOC-01's disclosed finding; test_alloc.py part 8 kept its stimulus
and now grades the other row.
THE STORE MIGRATION IS GRADED, NOT MERELY PERFORMED. sign_tripped/sign_reason
move to unit/<name>/... once, verified by READING BACK what was written, and
the version marker is set only when nothing that carried state was lost. The
flat keys stay in the file so a rollback still finds its latch. And __init__
compares the pre-migration flat latch against what unit A actually came up
with: if the latch did not survive it is RE-ARMED from the legacy value and
shouted about. A latch that silently fails to migrate is a disarmed safety
guard on an install whose state file says it is protected - which is strictly
worse than never persisting it, because now there is evidence saying it is
fine. test_inverter.py part 2 breaks the migration and watches the guard fire.
HEARTBEATS ARE INDEPENDENT AND THE BUDGET IS DERIVED. Each inverter owns its
deadline, the refreshes go out concurrently, and each write is bounded by
WRITE_BUDGET_S = 4 s. The inequality is graded rather than asserted: 4 +
heartbeat_s's schema maximum of 25 < 30, validate.FIRMWARE_INGEST_VETO_S.
Verified by injecting a transport that accepts and never answers.
live_units() ANSWERS HONESTLY. A unit whose every write has been rejected for a
whole layer-1 window is not commandable; the allocator drops it, the survivors
carry its share, the action log names it, and the dropped unit is commanded
0 W rather than left standing on its last share. A roster entry with NO
setpoint_entity is deliberately not handled there - that is a configuration
that cannot be delivered, and it reaches main.py's refusal branch, which is how
that branch keeps a live trigger after this card removed its old one.
AT N=1 NOTHING MOVES. The removed scalars survive as read/write properties of
the sole inverter, so every existing replay runs unmodified - and they REFUSE
at N>=2 rather than answering for the first inverter, which is the tripwire on
a consumer that was never made per-instance.
EVIDENCE. 19 files green. test_inverter.py is new (82 checks). Eighteen
mutations were applied to the shipped modules on a purged tree via
tools/mutation_run.sh and each was observed RED. Two of them SURVIVED first
time and the checks were changed rather than the table: the serialised dispatch
and the shared last_beat both passed while part 6 gathered its own writes and
computed its own `due` list. send_heartbeats() was extracted from run_control()
so the shipped dispatch is reachable by a check at all.
INVERTER B STILL DOES NOT EXIST. build_roster() returns exactly one inverter
from the flat options and is the one function CFG-04 changes; every two-unit
figure here is architecture, and the second rating in the checks is marked a
hypothesis.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BwRLw7eKYj2byVdwB6kbjU
|
||
|
glenn schrooyenandClaude Opus 5
|
186e13ceca |
ALLOC-01: allocate the site grid target, sum-invariant, never broadcast
checks / unit (pull_request) Successful in 4m6s
ARB-03 left `allocate()` as a documented seam that passed through at N<=1 and
DECLINED at N>=2, with the reasoning recorded at the line: the obvious
placeholder - hand the site target to each inverter - IS the ARCH-01 fault, and
"a placeholder that cannot be told apart from the real thing by any guard we
have is worse than no placeholder". This fills the seam.
WHY A BROADCAST IS A FAULT AND NOT AN INEFFICIENCY. The control primitive is
meter emulation: the master reports a FABRICATED grid reading and the inverter
regulates itself to it, so one number means "the battery moves X W" to this
add-on and "the grid is at X W" to the inverter - an identity that holds only at
N=1. Hand two inverters the same target and each supplies it IN FULL, so the
loop gain at the meter becomes N times the per-unit gain. Reproduced on
admin/goodwe's two-master rig against this tree: -400 W export overshoot at
gain 0.6, and at gain 1.0 - which config.yaml accepts as its documented maximum
- a permanent 4000 W peak-to-peak limit cycle, commands alternating
1000/2000/1000/0 W, with NOT ONE GUARD FIRING. The sign watchdog is evaluated on
every cycle and never trips, because the disagreement flips on alternate ticks
and never holds its 180 s window.
WHAT WAS BUILT. `app/allocator.py`, host-side and strictly above the driver
seam. It consumes intervals and a number and nothing else - no claims, no
priorities, and no measurement, because an allocator that read P1 would be a
SECOND integrator on the site measurement.
* ENVELOPE CLAMP, from ARB-03 stage 1's `UnitBounds`, read PER TICK. `plan()`
is a pure function and the module holds no state, so there is nowhere for a
construction-time envelope to be cached.
* WATER-FILL by weight, with the surplus of a full unit moving to the others.
Weights express PREFERENCE, envelopes express PERMISSION: a zero weight is
honoured while another unit has headroom and overridden when it does not,
because the SUM is the safety property and stranding capacity behind a
comfort choice manufactures a standing error at the meter.
* LARGEST-REMAINDER ROUNDING. The target is rounded ONCE, at the top; every
sum below that line is integer arithmetic, so it cannot drift. Measured: a
per-unit round() drifts -125 W over 50 000 ticks at three inverters.
* THE ALLOCATION RECORD, with `unmet_w`, `dropped` and a per-unit reason.
`ok` is one boolean for SAFETY-28 to gate on, and `unmet_w` is deliberately
NOT fused into it - a saturated inverter must not park the whole site.
THE STABILITY PROPERTY IS STRUCTURAL. With sum(a_i) = 1 the gain at the meter is
sum(a_i * g_i), a convex combination of the per-unit gains, so it lies in
[min g_i, max g_i] FOR ANY SPLIT. That is what makes the weights a comfort
choice rather than a tuning parameter, and it is why a bad split cannot
destabilise the site. Graded directly, and measured end to end: 90/10, 50/50 and
100/0 give a bit-identical aggregate trace at the meter, and a gain sweep to 1.0
at every split has a flat tail at all 84 combinations.
THE TWO REFUSALS main.py NOW MAKES, AND WHY THERE ARE TWO. `allocated=False` no
longer means "no allocator"; it means no safe split EXISTS - a mis-routed claim,
an empty envelope set, or a split that could only be made by commanding one pack
to charge while another discharges. Separately, and this guard is the price of
making allocate() succeed rather than something inherited, main.py refuses to
act on a valid split it cannot DELIVER: there is one `setpoint_entity`, so
writing the site target to the one inverter it can reach would over-command that
inverter by exactly the other's share. Commanding a partial site is a worse
instruction than commanding nothing. DEV-02 gives it the second writer.
scope_violations() IS NOW CONSUMED, AND AS A REFUSAL. ARB-03 left it logging
without changing the resolution. A mis-routed claim means the number about to be
divided may itself be wrong, and dividing a number that may be wrong is worse
than not dividing it - the same answer clause 3 already gives a contradiction.
It is NOT promoted into an envelope: a unit-scoped claim that landed on the site
names no unit, so promoting it would mean guessing one, or applying one
inverter's rating to a sibling that does not have it.
AT N<=1 IT IS A PASS-THROUGH, BYTE-IDENTICAL: the share is the site target, as a
float, unrounded and unclamped. All 17 existing suites replay unedited.
test_peakguard.py part 10h and test_arbiter.py part 7 were RE-AIMED, not
deleted. PR #40's reviewer proved that refusal branch was unreachable and
untested once; removing its trigger without giving it a new one would have
rebuilt the hole. Part 10h now drives three real cycles - a deliverable split, a
dead sibling, and every master stale - and grades all three.
KNOWN AND NOT CLOSED HERE: at a roster of two the loop's `actual_w` is one
inverter's power used as a site measurement, because the add-on binds a single
`inverter_power` sensor. That is the SENSING half of the same N=1 identity this
card removes from the COMMANDING half, and it makes the aggregate trace
split-dependent below gain 1.0. Measured with its counterfactual in test_alloc.py
part 8 and tools/alloc_sim.py --splits. It belongs to DEV-02.
Verified: all 18 suites green in the CI image. 16 mutations run through
tools/mutation_run.sh on a purged tree; 13 red, and the 3 survivors are recorded
in test_alloc.py part 12 with the reason each cannot be caught.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
|
||
|
glenn schrooyenandClaude Opus 5
|
7b4877d56d |
TEST-15: purge the bytecode cache before a mutation run, not after
checks / unit (pull_request) Successful in 3m35s
FCST-03's developer reverted `self.provisional` to `here.provisional`, re-ran, and reported the mutation as a survivor. It was not one. CPython validates a cached `.pyc` on (source size, source mtime truncated to whole seconds), and that edit moved neither number - same length, same second - so the interpreter loaded the bytecode it already had. Disassembly settled it at the time: the source read `here`, the running code loaded `self`. A mutation run had graded code that was not there, with no error and exit status 0. This does not make the results already on the record wrong. It makes them UNVERIFIED, and that distinction is the whole card - so nothing here re-runs a campaign and nothing here quietly assumes the old numbers hold. REPRODUCED, THEN CLOSED. test_bytecode.py builds a real package whose two return values are the same length, imports it in a real subprocess, rewrites it inside the same second (retrying until both writes land in one whole second - milliseconds, no sleep) and reads back which value ran. Part 1 shows the interpreter running SELF off a file that says HERE. AND THE TWO OBVIOUS MITIGATIONS DO NOT WORK. Measured, not read off documentation: `python3 -B` and PYTHONDONTWRITEBYTECODE=1 are write-side only. Neither stops an EXISTING stale .pyc being read, and a mutation run always has one, because the previous run wrote it. Part 2 grades both still running the old code. The advice in test_control.py's coverage audit led with `-B`; it is corrected here - the purge is the half that works, and `-B` earns its place by keeping THIS run from leaving a cache for the NEXT edit to inherit. NO SLEEP ANYWHERE. One-second mtime resolution is the mechanism, not the enemy. Waiting it out would slow every run down and would still only cover the edit somebody remembered to slow down. THE HARNESS. tools/mutation_run.sh purges, prints a sha256/byte fingerprint of every app/*.py it is about to grade - so a mutation report can say WHICH source produced it, which is the reviewer's addition to this card - runs each suite under `python3 -B`, refuses a tree with no test_*.py (CI-01's rule, for CI-01's reason), and fails if a cache appears during the run. It takes the tree as an argument so the infra repo's replay harnesses can call it without being moved. WHERE IT IS SAID (AC 2). MUTATION.md carries the mechanism, the measurement, the reporting rules and the survey, and every one of the 16 suites now points at it from its docstring. That list is DERIVED FROM THE TREE by test_bytecode.py part 5, so a new suite that joins this directory without the pointer is red - the same failure shape CI-01's file glob exists to prevent. CI WAS ALREADY IMMUNE, AND IS DESCRIBED THAT WAY (AC 4). Verified rather than reasoned: with the checkout bind mounted `-v "$WS:/w:ro"` the interpreter writes no __pycache__ into the tree at all, and it fails SILENTLY - no error, exit status 0 - while the same probe on a writable mount produces one. Each job also rm -rf's its workspace and clones afresh, so nothing crosses between jobs on the shared runner. PYTHONDONTWRITEBYTECODE=1 is added as belt and braces only: the `:ro` is one character wide and the next person debugging a write will be tempted by it. Part 6 grades both ON THE DOCKER COMMAND LINE rather than on the file - the first draft searched the whole workflow and stayed green after the flag was deleted, because the comment explaining it still named it. THE SHIPPED ADD-ON IS NOT AFFECTED and the question is closed: goodwe_controller/Dockerfile has set PYTHONDONTWRITEBYTECODE=1 since it was written. There is no deploy-side finding. THE SURVEY (AC 3), in MUTATION.md, scoped BY THE PROPERTY. The defect is one-directional - a mutation that never took leaves the suite green, so stale bytecode manufactures SURVIVORS and cannot manufacture a red - which puts every "all red, no survivors" claim in the window out of scope, along with mutations executed in-process (SAFETY-07's B1/B2/B3, test_peakguard parts 7 and 10, where no source file is edited) and everything run after the purge discipline landed on 2026-08-31 (SAFETY-24, SAFETY-26, OPT-04, FCST-03's re-run). What is left is the recorded SURVIVORS from source-edit runs before that date. Two were re-run here under the harness and both still stand: TARIFF-01's record_block() ignoring its policy argument, and SAFETY-03's abs() on the clamp limit - 15 suites green each, on the tree as it stood. The rest are labelled still-unverified rather than assumed or re-run; each is one harness run to close, and the moment to spend it is when somebody next touches that module. OUT OF SCOPE AND RECORDED: the infra repo's sim/ replay harnesses import the add-on and write bytecode into both trees (goodwe/sim/__pycache__ exists today), so SAFETY-02's headline replay carries the same exposure. mutation_run.sh takes a tree argument for exactly that, but the change belongs in that repo and some of those harnesses want the rig. RED AND GREEN. Every new check was broken and watched fail: the purge commented out of the harness turns part 4 red (and its non-vacuity probe already proves the fixture would mis-grade without it); one docstring pointer removed turns part 5 red naming the file; `:ro` and the `-e` flag removed from the workflow turn part 6 red. Full suite green, 16 files, in the CI image on a read-only mount - the shape the runner uses. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> |
||
|
glenn schrooyenandClaude Opus 5
|
04292c4a08 |
DATA-01: thirty days of the real meter, and what AC 3 actually measures
checks / unit (pull_request) Successful in 3m36s
TARIFF-01 and SAFETY-07 both merged with AC 3 - "monthly peak held under
the soft limit in >= 95 % of intervals over a 30-day simulation replay" -
honestly unmet, because two days of capture existed and it was believed
nothing longer could be produced. That belief was a true statement about
the past read as a claim about the future: HA's recorder still held the
whole month, and the reason nobody found out is that the 2026-08-25
export was a one-off script that left no artefact.
So the exporter is committed and runnable now.
scenarios/p1-net-30d-2026-08/ is 2026-08-01 to 2026-08-30, 485,841
samples, 30/30 days read, 0 failed, 4.13 MB gzipped, in the schema the
two committed days already fixed.
VERIFIED AGAINST THE DAYS THAT EXIST. `--verify` re-fetches 2026-08-20
and 2026-08-23 by this exact route and reproduces both committed files
BYTE FOR BYTE after decompression (865,024 B / 15,600 samples and
894,387 B / 16,351 samples). The gzip containers differ by design: this
exporter writes mtime 0 so its output is reproducible, where the one-off
wrote a wall clock.
A FAILED READ IS NOT AN OBSERVATION. A day whose read fails writes no
file, is recorded "read": "FAILED", and the run exits non-zero; the
entity is proved to exist through /api/states before any empty day is
believed, because a typo would otherwise answer 200 [] thirty times and
produce a consistent dataset of nothing. Nothing is interpolated, so the
month's 13 gaps (1291 s, 99.95 % covered) and its 3 `unavailable`
samples stay visible as holes.
SIZE: COMMITTED, DELIBERATELY, AND MEASURED. A git bundle of the whole
history goes from 1.58 MB (release/1.0 @
|