feat: add clickhouse micro-batch storage

tests/unit/test_batch.py

@@ -0,0 +1,101 @@
+from datetime import datetime, timezone
+from ipaddress import IPv4Address
+from uuid import uuid4
+
+from core.domain.models import NormalizedCDR
+from core.worker.batch import (
+    BufferedCDR,
+    KafkaPosition,
+    MicroBatch,
+)
+
+
+def normalized_cdr(
+    raw_id: str,
+) -> NormalizedCDR:
+    now = datetime.now(
+        timezone.utc
+    )
+
+    return NormalizedCDR(
+        cdr_id=uuid4(),
+        raw_id=raw_id,
+
+        msisdn="+19875551234",
+        imsi="310410998877665",
+
+        service_type="4G_DATA",
+        radio_access_type="LTE_4G",
+        cell_id="4021_8812",
+
+        bytes_uploaded=100,
+        bytes_downloaded=200,
+        total_bytes=300,
+        duration_seconds=300,
+
+        session_start_time=now,
+        session_end_time=now,
+
+        source_ip=IPv4Address(
+            "10.240.12.1"
+        ),
+    )
+
+
+def buffered_item(
+    raw_id: str,
+    offset: int,
+) -> BufferedCDR:
+    return BufferedCDR(
+        cdr=normalized_cdr(raw_id),
+        position=KafkaPosition(
+            topic="cdr.raw",
+            partition=0,
+            offset=offset,
+        ),
+    )
+
+
+def test_batch_flushes_when_full() -> None:
+    batch = MicroBatch(
+        max_size=2,
+        flush_interval_seconds=10,
+    )
+
+    batch.add(
+        buffered_item(
+            raw_id="raw-1",
+            offset=1,
+        )
+    )
+
+    assert batch.is_full is False
+
+    batch.add(
+        buffered_item(
+            raw_id="raw-2",
+            offset=2,
+        )
+    )
+
+    assert batch.is_full is True
+
+
+def test_batch_clear_resets_state() -> None:
+    batch = MicroBatch(
+        max_size=2,
+        flush_interval_seconds=10,
+    )
+
+    batch.add(
+        buffered_item(
+            raw_id="raw-1",
+            offset=1,
+        )
+    )
+
+    batch.clear()
+
+    assert batch.size == 0
+    assert batch.is_empty is True
+    assert batch.is_due() is False