fix: publish collection history atomically

src/fenris/collector.py

@@ -5,7 +5,7 @@ This module implements the thinnest complete write path:
 - Acquire controller identity from sysfs
 - Normalize identity exactly once at write time
 - Validate every row against store invariants
-- Commit one well-formed sample
+- Publish the sample and derived evidence in one collection-owned transaction
 
 No code path outside the collector interrogates the device.
 """
@@ -200,7 +200,7 @@ def write_sample(
     """Write one sample to the observation store.
     
     Identity normalization happens exactly once here.
-    Returns segment info for the caller.
+    Returns segment info for the caller. Caller owns the transaction.
     """
     from .segment import find_current_segment, should_open_new_segment, open_segment
     
@@ -262,8 +262,6 @@ def write_sample(
         ),
     )
     
-    conn.commit()
-    
     return {
         "segment_opened": segment_opened,
         "segment_reason": reason,
@@ -284,6 +282,7 @@ def run_collection(
     This is the main entry point for the collector.
     Returns the run outcome.
     """
+    conn = None
     try:
         # Acquire counters and thermal evidence
         counters = acquire_from_smartctl(smartctl_data)
@@ -309,72 +308,55 @@ def run_collection(
         if history_path.exists():
             import_legacy_history(conn, history_path, clock=clock)
 
-        
-        try:
-            # Ensure monitoring period is open (issue #73 AC2)
-            ensure_period_open(conn, clock.utcnow())
-            
-            # Validate invariants
-            validate_sample_invariants(sample, conn)
-            
-            # Write sample and get segment info
-            seg_info = write_sample(sample, identity, conn, clock)
-            
-            # Derive hour observations from interval with previous sample
-            try:
-                # Find the sample we just wrote
-                cursor = conn.execute("SELECT id FROM samples ORDER BY id DESC LIMIT 1")
-                current_id = cursor.fetchone()[0]
+        # Collection owns one transaction for the sample and its evidence.
+        ensure_period_open(conn, clock.utcnow())
 
-                prev = find_previous_sample(conn, seg_info.get("segment_id"), current_id)
-                if prev is not None:
-                    # Build current sample dict for derivation
-                    current = {
-                        "id": current_id,
-                        "ts": sample["ts"],
-                        "bytes_written": sample["bytes_written"],
-                        "bytes_read": sample["bytes_read"],
-                        "power_on_hours": sample["power_on_hours"],
-                        "temperature_c": sample["temperature_c"],
-                        "data_units_written": sample["data_units_written"],
-                        "data_units_read": sample["data_units_read"],
-                    }
-                    derive_hours_from_interval(conn, prev, current)
+        validate_sample_invariants(sample, conn)
+        seg_info = write_sample(sample, identity, conn, clock)
 
-                # Rebuild day aggregates from hour observations
-                from .day_aggregate import derive_all_days, persist_day_aggregate
-                for agg in derive_all_days(conn):
-                    persist_day_aggregate(conn, agg)
-
-                # Derive local-day summary using system timezone (issue #90)
-                try:
-                    from .tz_util import detect_system_tz
-                    from .local_day import derive_local_day_summary, persist_local_day
-                    tz_name = detect_system_tz()
-                    clock_now = clock.utcnow()
-                    local_summary = derive_local_day_summary(conn, tz_name, clock_now)
-                    if local_summary is not None:
-                        persist_local_day(conn, local_summary)
-                except Exception:
-                    # Local-day derivation failure must not prevent publication
-                    pass
-
-                conn.commit()
-            except Exception:
-                # Derivation failure must not prevent sample persistence (issue #73 AC6)
-                pass
-            
-            return {
-                "ok": True,
-                "sample_count": 1,
-                "store_path": str(store_path),
+        cursor = conn.execute("SELECT id FROM samples ORDER BY id DESC LIMIT 1")
+        current_id = cursor.fetchone()[0]
+        prev = find_previous_sample(conn, seg_info.get("segment_id"), current_id)
+        if prev is not None:
+            current = {
+                "id": current_id,
+                "ts": sample["ts"],
+                "bytes_written": sample["bytes_written"],
+                "bytes_read": sample["bytes_read"],
+                "power_on_hours": sample["power_on_hours"],
+                "temperature_c": sample["temperature_c"],
+                "data_units_written": sample["data_units_written"],
+                "data_units_read": sample["data_units_read"],
             }
-        finally:
-            conn.close()
-    
-    except (AcquisitionError, InvariantViolationError) as e:
+            derive_hours_from_interval(conn, prev, current)
+
+        from .day_aggregate import derive_all_days, persist_day_aggregate
+        for agg in derive_all_days(conn):
+            persist_day_aggregate(conn, agg)
+
+        from .tz_util import detect_system_tz
+        from .local_day import derive_local_day_summary, persist_local_day
+        tz_name = detect_system_tz()
+        local_summary = derive_local_day_summary(conn, tz_name, clock.utcnow())
+        if local_summary is not None:
+            persist_local_day(conn, local_summary)
+
+        conn.commit()
+
+        return {
+            "ok": True,
+            "sample_count": 1,
+            "store_path": str(store_path),
+        }
+
+    except Exception as e:
+        if conn is not None:
+            conn.rollback()
         return {
             "ok": False,
             "error": str(e),
             "error_type": type(e).__name__,
         }
+    finally:
+        if conn is not None:
+            conn.close()