Remove legacy code for format conversion

README.md

@@ -100,14 +100,9 @@ Live and recovered records share the same Parquet table and watt columns:
 `production_ac_w`, `consumption_w`, `grid_w`, `grid_import_w`, and `grid_export_w`.
 Recovered chart values are stored directly in those fields; `production_w` (PV DC),
 lifetime counters, and boiler state remain unknown. `record_kind` and the original
-chart data distinguish recovered values from live snapshots. The older
+chart data distinguish recovered values from live snapshots. Both sources use the
-`history_interval.energy_wh` metadata is retained for compatibility. Existing archives
+same validation and completeness checks. The relay stores power directly; energy
-are normalized on `run` startup, preserving IDs, recording times and upload acknowledgements.
+totals are calculated by the receiving application.
-To normalize an archive before exporting without starting collection:
-
-```bash
-.venv/bin/python -m solar_relay --config config.toml normalize-history
-```
 
 These common power readings match the sibling `solar` application's existing
 model: its `stats` table stores timestamped watts for Production and Consumption,
@@ -177,7 +172,7 @@ for the local API's archive limitations.
 Optional recovery signs in to Solar.web with a real Fronius account and reads its
 private daily charts after the inverter uploads its backlog, using the same login
 flow and chart series as the sibling `solar` website. No cloud calls occur while
-disabled. A public guest link and Query API subscription/keys are not needed.
+disabled. A public guest link is not needed.
 
 Create a dedicated Solar.web / Fronius account and grant it permission to view this
 installation's metrics. Keep public sharing disabled. The account can have guest
@@ -218,16 +213,14 @@ not prevent other dates from being recovered. Login and network failures stop th
 attempt and retry on the next scheduled check.
 
 Set `collector.timezone` to the installation's Solar.web timezone; local chart dates
-include whole 23/25-hour DST days. The old `history.lookback_days` setting is accepted but
+include whole 23/25-hour DST days. `collector.retention_days` controls the automatic
-ignored; `collector.retention_days` controls the automatic recovery window.
+recovery window.
 
 Recovery pairs the `FromGen` Production series from the consumption view with the
 `ToConsumer` Consumption series from the production view, regardless of the display
 language. Paired power values are stored directly in the common watt readings.
-Grid import/export are their nonnegative differences. As in `solar`, multiplying
+Grid import/export are their nonnegative differences. The relay keeps the original
-each five-minute power value by `5/60` yields estimated Wh. The compatibility
+chart watts without converting them to interval energy.
-`history_interval` field retains those estimates and the conversion method;
-the original chart watts remain available without converting Wh back to W.
 Null or unpaired readings stay missing and are tried again later.
 
 Login uses HTTPS and in-memory cookies. Credentials are submitted only to the Fronius
@@ -238,12 +231,6 @@ disk. An expired session is renewed once. Rejected credentials,
 MFA, CAPTCHA, consent, or a changed sign-in form stop that attempt with a redacted
 error. Complete any required interactive account setup in a browser.
 
-When migrating from Query API recovery, replace `access_key_id_env` and
-`access_key_value_env` with `username` and `password` in `[history]`, or use the
-environment fallback above. Obsolete configuration keys are accepted and ignored
-for migration. Restart the service after updating its configuration or environment.
-Recovery requires the account credentials even if API keys remain set.
-
 ```bash
 .venv/bin/python -m solar_relay --config config.toml recover \
   --from 2026-10-29 --until 2026-10-31
@@ -254,8 +241,8 @@ per invocation and can query dates older than the automatic retention window. Un
 the whole retention window, even when longer than 30 days. Only aligned five-minute
 chart samples in W are accepted;
 longer aggregates and unexpected units are rejected.
-Recovery stores stable UUIDs and common power readings, retaining `history_interval`
+Recovery stores stable UUIDs and common power readings with their chart provenance.
-as provenance and compatibility metadata. A future solar-side importer can put both
+A future solar-side importer can put both
 live and recovered production/consumption in the existing `stats` watt table and
 use its normal five-minute Wh conversion exactly once.
 

config.example.toml

@@ -13,7 +13,7 @@ request_timeout_seconds = 5
 min_request_interval_seconds = 4
 
 [upload]
-url = "" # Set to https://your-solar-host/index.php?a=relay to enable uploads.
+url = "" # Set a compatible HTTPS receiver URL to enable uploads.
 token = "" # Shared bearer token; enter the real value only in private config.toml.
 batch_size = 100
 request_timeout_seconds = 10

solar_relay/cli.py

@@ -40,9 +40,6 @@ def next_poll_at(now, interval):
 
 
 def _run(config, store):
-    normalized = store.normalize_history()
-    if normalized:
-        LOG.info("Normalized %d recovered records to common power readings", normalized)
     stop = threading.Event()
     previous = {}
     uploader = Uploader(config.upload, store) if config.upload.url else None
@@ -106,11 +103,10 @@ def _parser():
     commands.add_parser("upload", help="Upload the pending queue, stopping on the first failure")
     commands.add_parser("status", help="Show local archive/queue status without contacting the inverter")
     commands.add_parser("prune", help="Delete expired whole local days after acknowledgement retention")
-    commands.add_parser("normalize-history", help="Backfill common power readings in previously recovered records")
     for name, help_text in [
         ("export", "Export archived samples as JSON Lines, including already uploaded samples"),
         ("replay", "Requeue archived samples for upload, preserving their IDs"),
-        ("recover", "Retrieve missing energy intervals from the authenticated Solar.web API"),
+        ("recover", "Retrieve missing power samples from authenticated Solar.web charts"),
     ]:
         command = commands.add_parser(name, help=help_text)
         command.add_argument("--from", dest="from_date", help="Inclusive local date or ISO timestamp")
@@ -141,8 +137,6 @@ def main(argv=None):
             print(json.dumps(store.status(), indent=2))
         elif args.command == "prune":
             print(json.dumps({"deleted": store.prune()}))
-        elif args.command == "normalize-history":
-            print(json.dumps({"normalized": store.normalize_history()}))
         elif args.command == "recover":
             if not config.history.enabled:
                 raise ValueError("Enable history and configure Solar.web account credentials before recovery")

solar_relay/config.py

@@ -9,8 +9,6 @@ import tomllib
 from urllib.parse import urlsplit
 from zoneinfo import ZoneInfo
 
-LEGACY_HISTORY_SETTINGS = {"access_key_id_env", "access_key_value_env", "lookback_days"}
-
 
 def utc_text(value: datetime) -> str:
     if value.tzinfo is None:
@@ -131,8 +129,7 @@ def load_config(path: Path) -> Config:
         "inverter": set(InverterConfig.__dataclass_fields__),
         "upload": set(UploadConfig.__dataclass_fields__),
         "http": set(HttpConfig.__dataclass_fields__),
-        # Keep existing private configurations valid while using the full retention window.
+        "history": set(HistoryConfig.__dataclass_fields__),
-        "history": set(HistoryConfig.__dataclass_fields__) | LEGACY_HISTORY_SETTINGS,
     }.items():
         if not isinstance(data.get(section, {}), dict):
             raise ValueError(f"{section} must be a table")
@@ -193,8 +190,7 @@ def load_config(path: Path) -> Config:
         raise ValueError("http.base_path must be empty or a URL path without a trailing slash")
     if isinstance(http.history_days, bool) or not isinstance(http.history_days, int) or not 1 <= http.history_days <= 90:
         raise ValueError("http.history_days must be an integer between 1 and 90")
-    history = HistoryConfig(**{key: value for key, value in data.get("history", {}).items()
+    history = HistoryConfig(**data.get("history", {}))
-                              if key not in LEGACY_HISTORY_SETTINGS})
     if not isinstance(history.enabled, bool):
         raise ValueError("history.enabled must be true or false")
     if not isinstance(history.pv_system_id, str) or (history.pv_system_id and not re.fullmatch(r"[0-9a-fA-F]{8}(?:-[0-9a-fA-F]{4}){3}-[0-9a-fA-F]{12}", history.pv_system_id)):

solar_relay/coverage.py

@@ -24,9 +24,6 @@ def _finite(value):
 
 def sample_coverage(sample):
     observed = milliseconds(sample["observed_at"])
-    if "history_interval" in sample:
-        # Cloud history has interval energy, but no cumulative meter or relay history.
-        return (sample["id"], sample["source_id"], observed, observed, 1, observed, 0, observed, 0)
     readings = sample.get("readings", {})
     result = [sample["id"], sample["source_id"], observed]
     for source, fields in [

solar_relay/history.py

@@ -1,8 +1,4 @@
-"""Optional signed-in Solar.web chart recovery of missing five-minute energy.
+"""Recover missing five-minute power samples from signed-in Solar.web charts."""
-
-GEN24's local Solar API exposes no archive. Use the same private account login,
-chart series and five-minute W-to-Wh conversion as the sibling solar website.
-"""
 
 from datetime import datetime, timedelta, timezone
 import logging
@@ -17,91 +13,33 @@ from .http import TransportError
 from .solarweb import ChartDataError, SolarWebClient
 
 LOG = logging.getLogger(__name__)
-CHANNELS = {
-    "production": "EnergyProductionTotal", "consumption": "EnergyConsumptionTotal",
-    "self_consumption": "EnergySelfConsumptionTotal", "bought": "EnergyPurchased", "sold": "EnergyFeedIn",
-}
 
 
-def normalize_history_sample(sample):
+def history_sample(stamp, generated, consumed, source_id, system_id):
-    """Expose recovered power in the same fields as live samples, including old archives."""
+    """Store paired chart watts in the same readings as a live power snapshot."""
-    if "history_interval" not in sample:
+    if isinstance(stamp, bool) or not isinstance(stamp, int) or stamp % 300_000:
-        return sample
+        raise ValueError("Historical record is not an aligned five-minute sample")
-    interval = sample["history_interval"]
+    if any(isinstance(value, bool) or not isinstance(value, (int, float))
-    if interval["duration_seconds"] != 300:
+           or not math.isfinite(value) or not 0 <= value <= 1e18 for value in (generated, consumed)):
-        raise ValueError("Historical record is not a five-minute interval")
+        raise ValueError("Historical chart power is missing or invalid")
-    energy = interval["energy_wh"]
-    if any(isinstance(energy[field], bool) or not isinstance(energy[field], (int, float))
-           or not math.isfinite(energy[field]) or not 0 <= energy[field] <= 1e18 for field in CHANNELS):
-        raise ValueError("Historical energy is missing or invalid")
-    power = {field: energy[field] * 12 for field in CHANNELS}
-    chart = sample.get("raw", {}).get("solarweb_history", {}).get("chart_power_w")
-    if chart is not None:
-        # Retain the chart's original watts instead of a W -> Wh -> W round trip.
-        generated, consumed = chart["production"], chart["consumption"]
-        if any(isinstance(value, bool) or not isinstance(value, (int, float))
-               or not math.isfinite(value) or value < 0 for value in (generated, consumed)):
-            raise ValueError("Historical chart power is missing or invalid")
-        if not all(math.isclose(value, power[field]) for field, value in
-                   (("production", generated), ("consumption", consumed))):
-            raise ValueError("Historical chart power disagrees with interval energy")
-        power = {"production": generated, "consumption": consumed,
-                 "bought": max(0, consumed - generated), "sold": max(0, generated - consumed)}
-    return {
-        **sample, "poll_interval_seconds": 300,
-        "readings": {**sample["readings"],
-                     "production_ac_w": power["production"], "consumption_w": power["consumption"],
-                     "grid_w": power["bought"] - power["sold"],
-                     "grid_import_w": power["bought"], "grid_export_w": power["sold"]},
-        "source_timestamps": {**sample.get("source_timestamps", {}), "powerflow": sample["observed_at"]},
-    }
-
-
-def history_sample(entry, source_id, system_id):
-    stamp = milliseconds(entry["logDateTime"])
-    if stamp % 300_000 or entry.get("logDuration") != 300:
-        raise ValueError("Historical record is not an aligned five-minute interval")
-    channels = {}
-    for channel in entry["channels"]:
-        name, value = channel.get("channelName"), channel.get("value")
-        if name not in CHANNELS.values():
-            continue
-        if name in channels or channel.get("unit") != "Wh" or channel.get("channelType") != "Energy":
-            raise ValueError("Unexpected historical channel metadata")
-        if isinstance(value, bool) or not isinstance(value, (int, float)) or not math.isfinite(value) or not 0 <= value <= 1e18:
-            raise ValueError("Historical energy is missing or invalid")
-        channels[name] = value
-    energy = {field: channels[name] for field, name in CHANNELS.items()}
     when = utc_text(datetime.fromtimestamp(stamp / 1000, timezone.utc))
-    return normalize_history_sample({
+    grid = consumed - generated
+    return {
         "schema_version": 1,
-        # Preserve IDs of intervals already recovered with the former Query API.
+        # This immutable UUID name is part of the record ID format, never a request URL.
         "id": str(uuid5(NAMESPACE_URL, f"https://swqapi.solarweb.com/{system_id}/{source_id}/{when}")),
-        "source_id": source_id, "observed_at": when,
+        "source_id": source_id, "observed_at": when, "poll_interval_seconds": 300,
-        "readings": dict.fromkeys([
+        "readings": {
-            "production_w", "production_ac_w", "consumption_w", "grid_w", "grid_import_w", "grid_export_w",
+            "production_w": None, "production_ac_w": generated, "consumption_w": consumed,
-            "production_ac_total_wh", "grid_import_total_wh", "grid_export_total_wh", "boiler_enabled", "boiler_reason",
+            "grid_w": grid, "grid_import_w": max(0, grid), "grid_export_w": max(0, -grid),
-        ]),
+            "production_ac_total_wh": None, "grid_import_total_wh": None, "grid_export_total_wh": None,
-        "source_timestamps": {},
+            "boiler_enabled": None, "boiler_reason": None,
-        "errors": {"meter": "Cumulative meter readings are unavailable in recovered energy intervals",
+        },
+        "source_timestamps": {"powerflow": when},
+        "errors": {"meter": "Cumulative meter readings are unavailable in recovered chart samples",
                    "boiler": "Historical boiler switch state is unavailable"},
-        "raw": {"solarweb_history": entry},
+        "raw": {"solarweb_history": {"chart_power_w": {"production": generated, "consumption": consumed}}},
-        "history_interval": {"provider": "fronius-solarweb", "start": when, "duration_seconds": 300, "energy_wh": energy},
+    }
-    })
-
-
-def chart_entries(production, consumption):
-    """Pair real chart samples; assume constant W over five minutes, as solar does."""
-    for stamp in sorted(production.keys() & consumption.keys()):
-        generated, consumed = production[stamp], consumption[stamp]
-        values = {"production": generated / 12, "consumption": consumed / 12,
-                  "self_consumption": min(generated, consumed) / 12,
-                  "bought": max(0, consumed - generated) / 12, "sold": max(0, generated - consumed) / 12}
-        yield {"logDateTime": utc_text(datetime.fromtimestamp(stamp / 1000, timezone.utc)), "logDuration": 300,
-               "channels": [{"channelName": name, "channelType": "Energy", "unit": "Wh", "value": values[field]}
-                            for field, name in CHANNELS.items()],
-               "chart_power_w": {"production": generated, "consumption": consumed},
-               "energy_method": "constant_power_over_five_minutes"}
 
 
 class HistoryRecovery:
@@ -158,11 +96,12 @@ class HistoryRecovery:
                 # A day with no usable cloud data must not block recovery of newer gaps.
                 LOG.warning("Solar.web chart data unavailable for %s; leaving its gaps for the next check", day)
                 continue
-            for entry in chart_entries(production, consumption):
+            for stamp in sorted(production.keys() & consumption.keys()):
-                epoch = milliseconds(entry["logDateTime"]) // 1000
+                epoch = stamp // 1000
                 if epoch not in missing or datetime.fromtimestamp(epoch, self.zone).date() != day:
                     continue
-                sample = history_sample(entry, self.source_id, self.config.pv_system_id)
+                sample = history_sample(stamp, production[stamp], consumption[stamp],
+                                        self.source_id, self.config.pv_system_id)
                 self.store.add(sample)
                 missing.remove(epoch)
                 recovered += 1

solar_relay/storage.py

@@ -20,7 +20,6 @@ import pyarrow.parquet as pq
 
 from .config import retention_cutoff, utc_text
 from .coverage import daily_coverage, milliseconds, sample_coverage
-from .history import normalize_history_sample
 
 NUMBERS = (
     "production_w", "production_ac_w", "consumption_w", "grid_w",
@@ -34,7 +33,6 @@ SCHEMA = pa.schema([
     *((field, pa.float64()) for field in NUMBERS),
     ("boiler_enabled", pa.bool_()), ("boiler_reason", pa.string()),
     ("record_kind", pa.string()),
-    ("interval_energy_wh", pa.struct([(field, pa.float64()) for field in ("production", "consumption", "self_consumption", "bought", "sold")])),
     ("power_ms", pa.int64()), ("power_ok", pa.bool_()),
     ("meter_ms", pa.int64()), ("meter_ok", pa.bool_()),
     ("boiler_ms", pa.int64()), ("boiler_ok", pa.bool_()),
@@ -82,9 +80,13 @@ class Store:
         if not path.exists():
             return []
         parquet = pq.ParquetFile(path, page_checksum_verification=True)
-        if not parquet.schema_arrow.equals(SCHEMA, check_metadata=True):
+        schema = parquet.schema_arrow
+        # Validate the fields this version uses; read only those columns.
+        if schema.metadata != SCHEMA.metadata or any(
+            schema.get_field_index(field.name) < 0 or schema.field(field.name) != field for field in SCHEMA
+        ):
             raise StoreError("Unsupported Parquet archive schema")
-        return parquet.read(columns=columns).to_pylist()
+        return parquet.read(columns=SCHEMA.names if columns is None else columns).to_pylist()
 
     @staticmethod
     def _sync_directory(path):
@@ -125,8 +127,7 @@ class Store:
             "payload": json.dumps(sample, allow_nan=False, separators=(",", ":")),
             **{name: readings.get(name) for name in NUMBERS},
             "boiler_enabled": readings.get("boiler_enabled"), "boiler_reason": readings.get("boiler_reason"),
-            "record_kind": "history" if "history_interval" in sample else "live",
+            "record_kind": "history" if "solarweb_history" in sample.get("raw", {}) else "live",
-            "interval_energy_wh": sample.get("history_interval", {}).get("energy_wh"),
             **{name: bool(value) if name.endswith("_ok") else value
                for name, value in zip(COVERAGE_COLUMNS[1:], coverage[3:])},
         }
@@ -137,32 +138,11 @@ class Store:
             rows = self._read(path)
             for existing in rows:
                 if existing["id"] == row["id"]:
-                    if normalize_history_sample(json.loads(existing["payload"])) != normalize_history_sample(sample):
+                    if json.loads(existing["payload"]) != sample:
                         raise StoreError("Sample ID already contains different data")
                     return
             self._write(path, [*rows, row])
 
-    def normalize_history(self):
-        """Backfill common power columns and queued payloads without changing IDs or acknowledgements."""
-        changed = 0
-        with self._locked(write=True):
-            for path in self._files():
-                rows = self._read(path)
-                dirty = False
-                for index, row in enumerate(rows):
-                    if row["record_kind"] != "history":
-                        continue
-                    sample = json.loads(row["payload"])
-                    normalized = normalize_history_sample(sample)
-                    if normalized == sample:
-                        continue
-                    rows[index] = {**self._row(normalized, row["uploaded_at"]), "recorded_ns": row["recorded_ns"]}
-                    changed += 1
-                    dirty = True
-                if dirty:
-                    self._write(path, rows)
-        return changed
-
     def pending(self, limit):
         with self._locked():
             rows = []

tests/test_history.py

@@ -8,7 +8,7 @@ import pyarrow.parquet as pq
 from zoneinfo import ZoneInfo
 
 from solar_relay.config import HistoryConfig, utc_text
-from solar_relay.history import CHANNELS, HistoryRecovery, chart_entries, history_sample
+from solar_relay.history import HistoryRecovery, history_sample
 from solar_relay.http import TransportError
 from solar_relay.solarweb import ChartDataError
 from solar_relay.storage import Store
@@ -17,12 +17,6 @@ from tests.test_status import observed
 SYSTEM = "73733052-6f10-47a5-9746-7f7e76ebcb8c"
 
 
-def entry(when):
-    values = {"production": 100, "consumption": 120, "self_consumption": 80, "bought": 40, "sold": 20}
-    return {"logDateTime": utc_text(when), "logDuration": 300,
-            "channels": [{"channelName": channel, "channelType": "Energy", "unit": "Wh", "value": values[field]}
-                         for field, channel in CHANNELS.items()]}
-
 
 class FakeCharts:
     def __init__(self, when):
@@ -55,9 +49,9 @@ class HistoryTests(unittest.TestCase):
             self.store.add(observed(self.when + timedelta(minutes=offset)))
         self.assertEqual(self.recover(), 1)
         records = list(self.store.records())
-        recovered = [s for s in records if "history_interval" in s]
+        recovered = [s for s in records if "solarweb_history" in s["raw"]]
         self.assertEqual(len(recovered), 1)
-        self.assertEqual(recovered[0]["history_interval"]["energy_wh"]["production"], 100)
+        self.assertEqual(recovered[0]["readings"]["production_ac_w"], 1200)
         self.assertIsNone(recovered[0]["readings"]["boiler_enabled"])
         self.assertNotIn("boiler", recovered[0]["source_timestamps"])
         self.assertIn("boiler", recovered[0]["errors"])
@@ -78,41 +72,20 @@ class HistoryTests(unittest.TestCase):
         self.assertEqual(day["complete"], 0)
         row = pq.ParquetFile(next(self.store.path.glob("date=*/samples.parquet"))).read().to_pylist()[0]
         self.assertEqual(row["record_kind"], "history")
-        self.assertEqual(row["interval_energy_wh"]["production"], 100)
         self.assertEqual(row["production_ac_w"], 1200)
         self.assertEqual(row["consumption_w"], 1440)
         self.assertEqual(row["grid_import_w"], 240)
         self.assertEqual(row["grid_export_w"], 0)
         self.assertEqual(day["slots"][120:123], [1, 1, 1])
 
-    def test_old_archive_normalization_preserves_delivery_state_and_live_records(self):
+    def test_recovered_power_uses_the_same_completeness_checks_as_live_power(self):
-        old = history_sample(entry(self.when), self.source, SYSTEM)
+        sample = history_sample(int(self.when.timestamp() * 1000), 1200, 1440, self.source, SYSTEM)
-        old["readings"] = dict.fromkeys(old["readings"])
+        sample["readings"]["consumption_w"] = None
-        old["source_timestamps"] = {}
+        self.store.add(sample)
-        old.pop("poll_interval_seconds")
+        day = self.store.coverage(self.source, "UTC", 1, self.when + timedelta(days=1))[1]
-        live = observed(self.when + timedelta(minutes=5))
+        self.assertEqual(day["energy_complete"], 0)
-        self.store.add(old)
+        self.assertEqual(day["boiler_complete"], 0)
-        self.store.add(live)
+        self.assertEqual(day["slots"][120], 0)
-        self.store.acknowledge([old["id"]], now=self.when + timedelta(hours=1))
-        path = next(self.store.path.glob("date=*/samples.parquet"))
-        before = {row["id"]: row for row in pq.ParquetFile(path).read().to_pylist()}
-
-        self.assertEqual(self.store.normalize_history(), 1)
-        self.assertEqual(self.store.normalize_history(), 0)
-        after = {row["id"]: row for row in pq.ParquetFile(path).read().to_pylist()}
-        self.assertEqual(after[live["id"]], before[live["id"]])
-        row = after[old["id"]]
-        for field in ("id", "observed_at", "recorded_ns", "uploaded_at"):
-            self.assertEqual(row[field], before[old["id"]][field])
-        self.assertEqual(row["production_ac_w"], 1200)
-        self.assertEqual(row["consumption_w"], 1440)
-        self.assertIsNone(row["production_w"])
-        self.assertIsNone(row["boiler_enabled"])
-        self.assertEqual(self.store.pending(100), [live])
-        self.assertEqual(list(self.store.records())[0]["source_timestamps"]["powerflow"], utc_text(self.when))
-        # Replaying the original legacy payload still identifies the same record.
-        self.store.add(old)
-        self.assertEqual(self.store.status()["samples"], 2)
 
     def test_partial_cloud_records_remain_missing_for_retry(self):
         stamp = int((self.when + timedelta(minutes=5)).timestamp() * 1000)
@@ -123,20 +96,23 @@ class HistoryTests(unittest.TestCase):
         self.assertEqual(self.recover(), 1)
 
     def test_recovered_identifiers_are_stable_for_duplicate_delivery(self):
-        first = history_sample(entry(self.when), self.source, SYSTEM)
+        stamp = int(self.when.timestamp() * 1000)
-        second = history_sample(entry(self.when), self.source, SYSTEM)
+        first = history_sample(stamp, 1200, 1440, self.source, SYSTEM)
+        second = history_sample(stamp, 1200, 1440, self.source, SYSTEM)
         self.assertEqual(first, second)
+        self.store.add(first)
+        self.store.add(second)
+        self.assertEqual(self.store.status()["samples"], 1)
 
-    def test_non_five_minute_and_bad_units_are_rejected(self):
+    def test_unaligned_timestamps_and_invalid_power_are_rejected(self):
-        for change in [{"logDuration": 599}, {"logDateTime": utc_text(self.when + timedelta(seconds=1))}]:
+        stamp = int(self.when.timestamp() * 1000)
-            value = entry(self.when)
+        for bad_stamp in (stamp + 1000, float(stamp), True):
-            value.update(change)
+            with self.subTest(stamp=bad_stamp), self.assertRaises(ValueError):
-            with self.assertRaises(ValueError):
+                history_sample(bad_stamp, 1200, 1440, self.source, SYSTEM)
-                history_sample(value, self.source, SYSTEM)
+        for invalid in (None, True, "1200", -1, float("nan"), float("inf"), 1e19):
-        value = entry(self.when)
+            for generated, consumed in ((invalid, 1440), (1200, invalid)):
-        value["channels"][0]["unit"] = "W"
+                with self.subTest(generated=generated, consumed=consumed), self.assertRaises(ValueError):
-        with self.assertRaises(ValueError):
+                    history_sample(stamp, generated, consumed, self.source, SYSTEM)
-            history_sample(value, self.source, SYSTEM)
 
     def test_missing_private_credentials_fail_before_requests(self):
         with self.assertRaisesRegex(ValueError, "SOLARWEB_USERNAME and SOLARWEB_PASSWORD"):
@@ -190,7 +166,7 @@ class HistoryTests(unittest.TestCase):
         self.client.points = {"production": points, "consumption": points}
         self.store.add(observed(now.replace(minute=0, second=0)))
         self.assertEqual(self.recovery.recover(now=now), 3)
-        recovered = [record for record in self.store.records() if "history_interval" in record]
+        recovered = [record for record in self.store.records() if "solarweb_history" in record["raw"]]
         self.assertEqual([record["observed_at"] for record in recovered], [utc_text(when) for when in recoverable])
         self.assertEqual(min(day for _, _, day in self.client.requests), cutoff.astimezone(self.recovery.zone).date())
 
@@ -266,22 +242,19 @@ class HistoryTests(unittest.TestCase):
                 self.recovery.recover(now=self.when + timedelta(days=2))
         self.assertEqual(client.call_count, 1)
 
-    def test_chart_conversion_matches_solar_and_retains_provenance(self):
+    def test_chart_watts_are_stored_directly_and_retain_provenance(self):
         for generated, consumed in [(1200, 1440), (2400, 1200), (0, 0), (1234.56789, 3456.78901)]:
             with self.subTest(generated=generated, consumed=consumed):
                 stamp = int(self.when.timestamp() * 1000)
-                record = next(chart_entries({stamp: generated}, {stamp: consumed}))
+                sample = history_sample(stamp, generated, consumed, self.source, SYSTEM)
-                sample = history_sample(record, self.source, SYSTEM)
-                values = sample["history_interval"]["energy_wh"]
-                self.assertEqual(values, {"production": generated / 12, "consumption": consumed / 12,
-                                          "self_consumption": min(generated, consumed) / 12,
-                                          "bought": max(0, consumed - generated) / 12,
-                                          "sold": max(0, generated - consumed) / 12})
-                self.assertEqual(record["energy_method"], "constant_power_over_five_minutes")
                 self.assertEqual(sample["readings"]["production_ac_w"], generated)
                 self.assertEqual(sample["readings"]["consumption_w"], consumed)
                 self.assertEqual(sample["readings"]["grid_w"], consumed - generated)
+                self.assertEqual(sample["readings"]["grid_import_w"], max(0, consumed - generated))
+                self.assertEqual(sample["readings"]["grid_export_w"], max(0, generated - consumed))
                 self.assertIsNone(sample["readings"]["production_w"])
+                self.assertEqual(sample["raw"]["solarweb_history"]["chart_power_w"],
+                                 {"production": generated, "consumption": consumed})
 
     def test_dst_days_recover_every_actual_five_minute_slot(self):
         for day, expected in [("2026-03-29", 276), ("2026-10-25", 300)]:

tests/test_parquet.py

@@ -7,11 +7,12 @@ import unittest
 from unittest.mock import patch
 from zoneinfo import ZoneInfo
 
+import pyarrow as pa
 import pyarrow.parquet as pq
 
 from solar_relay.cli import next_poll_at
 from solar_relay.config import retention_cutoff
-from solar_relay.storage import Store, StoreError
+from solar_relay.storage import SCHEMA, Store, StoreError
 from tests.test_status import observed
 
 
@@ -49,6 +50,33 @@ class ParquetTests(unittest.TestCase):
         self.assertEqual(self.store.prune(now=self.now + timedelta(days=31)), 2)
         self.assertEqual(list(self.store.records()), [items[3]])
 
+    def test_additional_archive_columns_do_not_change_records_or_acknowledgements(self):
+        value = observed(self.now)
+        self.store.add(value)
+        path = next(self.path.glob("date=*/samples.parquet"))
+        table = pq.ParquetFile(path).read()
+        table = table.append_column("unused_estimate", pa.array([100.0]))
+        pq.write_table(table, path)
+        self.assertEqual(self.store.pending(100), [value])
+        self.assertEqual(list(self.store.records()), [value])
+        self.store.acknowledge([value["id"]], now=self.now)
+        self.assertEqual(self.store.pending(100), [])
+        self.assertEqual(list(self.store.records()), [value])
+        self.assertTrue(pq.ParquetFile(path).schema_arrow.equals(SCHEMA, check_metadata=True))
+
+    def test_missing_or_mistyped_archive_columns_are_rejected(self):
+        value = observed(self.now)
+        self.store.add(value)
+        path = next(self.path.glob("date=*/samples.parquet"))
+        table = pq.ParquetFile(path).read()
+        index = table.schema.get_field_index("production_ac_w")
+        for invalid in (table.drop_columns(["production_ac_w"]),
+                        table.set_column(index, "production_ac_w", pa.array(["invalid"]))):
+            with self.subTest(schema=invalid.schema):
+                pq.write_table(invalid, path)
+                with self.assertRaises(StoreError):
+                    self.store.pending(100)
+
     def test_retention_boundary_is_local_midnight_regardless_of_execution_time(self):
         zone = ZoneInfo("Europe/Zurich")
         for hour in (0, 12, 23):

tests/test_relay.py

@@ -402,27 +402,28 @@ class ConfigTests(unittest.TestCase):
         with self.assertRaises(ValueError):
             self.load('\n[http]\nbase_path="/../config"\n')
 
-    def test_history_account_settings_and_migration_from_query_api_keys(self):
+    def test_history_account_environment_settings(self):
         config, _ = self.load('\n[history]\nusername_env="BACKUP_USER"\npassword_env="BACKUP_PASSWORD"\n')
         self.assertEqual(config.history.username_env, "BACKUP_USER")
         self.assertEqual(config.history.password_env, "BACKUP_PASSWORD")
         self.assertFalse(config.history.enabled)
-        config, _ = self.load('\n[history]\naccess_key_id_env="OLD_ID"\naccess_key_value_env="OLD_KEY"\n')
-        self.assertEqual(config.history.username_env, "SOLARWEB_USERNAME")
-        self.assertEqual(config.history.password_env, "SOLARWEB_PASSWORD")
         for name in ("username_env", "password_env"):
             with self.subTest(name=name), self.assertRaisesRegex(ValueError, "environment variable"):
                 self.load(f'\n[history]\n{name}="invalid name"\n')
 
-    def test_history_uses_retention_instead_of_legacy_lookback_setting(self):
+    def test_history_uses_collector_retention(self):
         with tempfile.TemporaryDirectory() as directory:
             path = Path(directory) / "synthetic-test.toml"
             path.write_text('[collector]\nretention_days=2\n[history]\nenabled=true\n'
-                            'pv_system_id="73733052-6f10-47a5-9746-7f7e76ebcb8c"\nlookback_days=7\n')
+                            'pv_system_id="73733052-6f10-47a5-9746-7f7e76ebcb8c"\n')
             config = load_config(path)
         self.assertEqual(config.collector.retention_days, 2)
         self.assertTrue(config.history.enabled)
-        self.assertFalse(hasattr(config.history, "lookback_days"))
+
+    def test_obsolete_history_options_are_rejected(self):
+        for name in ("access_key_id_env", "access_key_value_env", "lookback_days"):
+            with self.subTest(name=name), self.assertRaisesRegex(ValueError, "Unknown history settings"):
+                self.load(f'\n[history]\n{name}=7\n')
 
     def test_private_config_accepts_credentials_and_redacts_them_from_repr(self):
         config, _ = self.load('\n[upload]\ntoken="synthetic-shared-token"\n'