offline_log_viewer: fix get_control_record_type

Noticed this was broken while parsing some logs.
Added a little regression test for the parsing

tests/rptest/clients/offline_log_viewer.py

@@ -39,6 +39,9 @@ def _json_cmd(self, node, suffix):
             self._redpanda.logger.error(f"Invalid JSON output: {json_out}")
             raise
 
+    def read_kafka_records(self, node, topic):
+        return self._json_cmd(node, f"--type kafka_records --topic {topic}")
+
     def read_controller(self, node):
         return self._json_cmd(node, "--type controller")
 

tests/rptest/tests/transactions_test.py

@@ -23,6 +23,7 @@
 from ducktape.utils.util import wait_until
 from ducktape.errors import TimeoutError
 
+from rptest.clients.offline_log_viewer import OfflineLogViewer
 from rptest.tests.redpanda_test import RedpandaTest
 from rptest.services.admin import Admin
 from rptest.services.redpanda import RedpandaService, SecurityConfig, SaslCredentials
@@ -246,6 +247,12 @@ def simple_test(self):
                 ), f'Records value does not match from input {consumed_from_input_topic[index_from_input].value()}, from output {record.value()}'
                 index_from_input += 1
 
+        log_viewer = OfflineLogViewer(self.redpanda)
+        for node in self.redpanda.started_nodes():
+            records = log_viewer.read_kafka_records(node=node,
+                                                    topic=self.input_t.name)
+            self.logger.info(f"Read {len(records)} from node {node.name}")
+
     @cluster(num_nodes=3)
     def rejoin_member_test(self):
         self.generate_data(self.input_t, self.max_records)

tools/offline_log_viewer/kafka.py

@@ -47,6 +47,14 @@ def spec(rdr, version):
     return rdr.read_envelope(spec, max_version=3)
 
 
+def get_control_record_type(key):
+    rdr = Reader(BytesIO(key))
+    rdr.skip(2)  # skip the 16bit version.
+    # Encoded as big endian
+    type_rdr = Reader(BytesIO(struct.pack(">h", rdr.read_int16())))
+    return KafkaControlRecordType(type_rdr.read_int16()).name
+
+
 def decode_archival_metadata_command(kr, vr):
     key = kr.read_int8()
     if key == ArchivalMetadataCommand.add_segment:
@@ -82,7 +90,7 @@ def decode_record(batch, header, record):
     is_ctrl = attrs["control_batch"]
     is_tx_ctrl = is_txn and is_ctrl
     if is_tx_ctrl:
-        record_dict["type"] = self.get_control_record_type(record.key)
+        record_dict["type"] = get_control_record_type(record.key)
 
     kr = Reader(BytesIO(record.key))
     vr = Reader(BytesIO(record.value))
@@ -102,14 +110,7 @@ def __init__(self, ntp, headers_only):
         self.ntp = ntp
         self.headers_only = headers_only
 
-    def get_control_record_type(self, key):
-        rdr = Reader(BytesIO(key))
-        rdr.skip(2)  # skip the 16bit version.
-        # Encoded as big endian
-        type_rdr = Reader(BytesIO(struct.pack(">h", rdr.read_int16())))
-        return KafkaControlRecordType(type_rdr.read_int16()).name
-
-    def decode(self):
+    def __iter__(self):
         self.results = []
         for batch in self.batches():
             header = batch.header_dict()

tools/offline_log_viewer/viewer.py

@@ -80,8 +80,10 @@ def print_kafka(store, topic, headers_only):
 
             logger.info(f'topic: {ntp.topic}, partition: {ntp.partition}')
             log = KafkaLog(ntp, headers_only=headers_only)
-            for result in log.decode():
-                logger.info(json.dumps(result, indent=2))
+            json_iter = json.JSONEncoder(indent=2).iterencode(
+                SerializableGenerator(log))
+            for record in json_iter:
+                print(record, end='')
 
 
 def print_groups(store):