Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 1 addition & 11 deletions .github/workflows/lint-and-test.yml
Original file line number Diff line number Diff line change
Expand Up @@ -7,18 +7,8 @@ on:
branches: [ main ]
types: [ opened, synchronize, reopened ]

permissions:
contents: read
pull-requests: write

jobs:
lint-and-test:
uses: esclient/tools/.github/workflows/lint-and-test-python.yml@v1.0.1
uses: esclient/tools/.github/workflows/lint-and-test-python.yml@v1.0.2
with:
python-version: "3.13.7"
source: "modservice"
sonar-inclusions: "src/**,Dockerfile"
sonar-exclusions: "**/grpc/**"
sonar-coverage-exclusions: "src/modservice/server.py,src/modservice/settings.py,src/modservice/s3_client.py"
secrets:
SONAR_TOKEN: ${{ secrets.SONAR_TOKEN }}
22 changes: 0 additions & 22 deletions migrations/20260308000000_add_on_moderation_status.sql

This file was deleted.

13 changes: 12 additions & 1 deletion pdm.lock

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

1 change: 1 addition & 0 deletions pyproject.toml
Original file line number Diff line number Diff line change
Expand Up @@ -157,6 +157,7 @@ dev = [
"types-psycopg2==2.9.21.20250915",
"types-aioboto3==15.4.0",
"types-aiofiles==25.1.0.20251011",
"types-confluent-kafka==1.2.2",
"botocore-stubs==1.40.55",
"asyncpg-stubs==0.30.2",
]
7 changes: 6 additions & 1 deletion src/modservice/handler/get_mods.py
Original file line number Diff line number Diff line change
@@ -1,6 +1,11 @@
import grpc

from modservice.constants import STATUS_BANNED, STATUS_HIDDEN, STATUS_ON_MODERATION, STATUS_UPLOADED
from modservice.constants import (
STATUS_BANNED,
STATUS_HIDDEN,
STATUS_ON_MODERATION,
STATUS_UPLOADED,
)
from modservice.grpc import mod_pb2
from modservice.service.service import ModService

Expand Down
63 changes: 37 additions & 26 deletions src/modservice/kafka/config.py
Original file line number Diff line number Diff line change
@@ -1,37 +1,48 @@
import os
from typing import Any

type KafkaClientConfig = dict[str, Any]


class KafkaConfig:

def __init__(self):
self.brokers = os.getenv('KAFKA_BROKERS', 'localhost:9092')
self.request_topic = os.getenv('KAFKA_REQUEST_TOPIC', 'moderation-request')
self.result_topic = os.getenv('KAFKA_RESULT', 'moderation-result')
self.consumer_group_id = os.getenv('KAFKA_CONSUMER_GROUP_ID', 'mod-service-group')
self.max_retries = int(os.getenv('KAFKA_MAX_RETRIES', '3'))
self.retry_backoff_ms = int(os.getenv('KAFKA_RETRY_BACKOFF_MS', '1000'))
self.enable_ssl = os.getenv('KAFKA_ENABLE_SSL', 'false').lower() == 'true'
def __init__(self) -> None:
self.brokers = os.getenv("KAFKA_BROKERS", "localhost:9092")
self.request_topic = os.getenv(
"KAFKA_REQUEST_TOPIC", "moderation-request"
)
self.result_topic = os.getenv("KAFKA_RESULT", "moderation-result")
self.consumer_group_id = os.getenv(
"KAFKA_CONSUMER_GROUP_ID", "mod-service-group"
)
self.max_retries = int(os.getenv("KAFKA_MAX_RETRIES", "3"))
self.retry_backoff_ms = int(
os.getenv("KAFKA_RETRY_BACKOFF_MS", "1000")
)
self.enable_ssl = (
os.getenv("KAFKA_ENABLE_SSL", "false").lower() == "true"
)

def get_producer_config(self):
config = {
'bootstrap.servers': self.brokers,
'client.id': 'mod_service_producer',
'acks': 'all',
'retries': self.max_retries,
'retry.backoff.ms': self.retry_backoff_ms,
'enable.idempotence': True,
'compression.type': 'none'
def get_producer_config(self) -> KafkaClientConfig:
config: KafkaClientConfig = {
"bootstrap.servers": self.brokers,
"client.id": "mod_service_producer",
"acks": "all",
"retries": self.max_retries,
"retry.backoff.ms": self.retry_backoff_ms,
"enable.idempotence": True,
"compression.type": "none",
}
return config

def get_consumer_config(self):
config = {
'bootstrap.servers': self.brokers,
'group.id': self.consumer_group_id,
'auto.offset.reset': 'earliest',
'enable.auto.commit': True,
'auto.commit.interval.ms': 1000,
'session.timeout.ms': 30000,
'heartbeat.interval.ms': 10000
def get_consumer_config(self) -> KafkaClientConfig:
config: KafkaClientConfig = {
"bootstrap.servers": self.brokers,
"group.id": self.consumer_group_id,
"auto.offset.reset": "earliest",
"enable.auto.commit": True,
"auto.commit.interval.ms": 1000,
"session.timeout.ms": 30000,
"heartbeat.interval.ms": 10000,
}
return config
66 changes: 48 additions & 18 deletions src/modservice/kafka/consumer.py
Original file line number Diff line number Diff line change
@@ -1,40 +1,52 @@
"""Kafka consumer for receiving moderation responses."""

from confluent_kafka import Consumer, KafkaException, KafkaError
import logging
import threading
from .config import KafkaConfig
from collections.abc import Callable

from confluent_kafka import Consumer, KafkaError, KafkaException, Message

from modservice.grpc import moderation_pb2

from .config import KafkaConfig

logger = logging.getLogger(__name__)


class KafkaConsumer:

def __init__(self, config: KafkaConfig, callback):
def __init__(
self,
config: KafkaConfig,
callback: (
Callable[[moderation_pb2.ModerateObjectResponse, int], None] | None
),
) -> None:

self.config = config
self.callback = callback
self.consumer = Consumer(config.get_consumer_config())
self.topic = config.result_topic
self.running = False
self.consumer_thread = None
self.consumer_thread: threading.Thread | None = None

self.consumer.subscribe([self.topic])
logger.info(f"KafkaConsumer subscribed to topic: {self.topic}")

def start(self):
def start(self) -> None:

if self.running:
logger.warning("Consumer already running")
return

self.running = True
self.consumer_thread = threading.Thread(target=self._consume_loop, daemon=True)
self.consumer_thread = threading.Thread(
target=self._consume_loop, daemon=True
)
self.consumer_thread.start()
logger.info("KafkaConsumer started")

def stop(self):
def stop(self) -> None:

if not self.running:
return
Expand All @@ -47,7 +59,7 @@ def stop(self):
self.consumer.close()
logger.info("KafkaConsumer stopped")

def _consume_loop(self):
def _consume_loop(self) -> None:

logger.info("Consumer loop started")

Expand All @@ -57,13 +69,16 @@ def _consume_loop(self):

if msg is None:
continue
if msg.error():
if msg.error().code() == KafkaError._PARTITION_EOF:
logger.debug(f"Reached end of partition for topic {msg.topic()}")
elif msg.error().code() == KafkaError._TIMED_OUT:
error = msg.error()
if error is not None:
if error.code() == KafkaError._PARTITION_EOF:
logger.debug(
f"Reached end of partition for topic {msg.topic()}"
)
elif error.code() == KafkaError._TIMED_OUT:
logger.debug("Timeout")
else:
logger.error(f"Consumer error: {msg.error()}")
logger.error(f"Consumer error: {error}")
continue

self._process_message(msg)
Expand All @@ -75,20 +90,35 @@ def _consume_loop(self):

logger.info("Consumer loop ended")

def _process_message(self, msg):
def _process_message(self, msg: Message) -> None:

try:
object_id = 0
if msg.key() is not None:
message_key = msg.key()
if message_key is not None:
try:
key_str = msg.key().decode('utf-8')
key_str = (
message_key.decode("utf-8")
if isinstance(message_key, bytes)
else message_key
)
object_id = int(key_str)
except ValueError as e:
logger.error(f"Failed to parse message key to request ID: {e}")
logger.error(
f"Failed to parse message key to request ID: {e}"
)
return

message_value = msg.value()
if message_value is None:
logger.error("Received message without payload")
return
if not isinstance(message_value, bytes):
logger.error("Received message payload with unexpected type")
return

response = moderation_pb2.ModerateObjectResponse()
response.ParseFromString(msg.value())
response.ParseFromString(message_value)

logger.info(
f"Received moderation response: object_id={object_id}, "
Expand Down
33 changes: 21 additions & 12 deletions src/modservice/kafka/producer.py
Original file line number Diff line number Diff line change
@@ -1,40 +1,47 @@
"""Kafka producer for sending moderation requests."""

from confluent_kafka import Producer, KafkaException
import logging
from .config import KafkaConfig

from confluent_kafka import KafkaError, KafkaException, Message, Producer

from modservice.grpc import moderation_pb2

from .config import KafkaConfig

logger = logging.getLogger(__name__)


class KafkaProducer:

def __init__(self, config: KafkaConfig):
def __init__(self, config: KafkaConfig) -> None:
self.config = config
self.producer = Producer(config.get_producer_config())
self.topic = config.request_topic

logger.info(f"KafkaProducer initialized for topic: {self.topic}")

def send_moderation_request(self, request: moderation_pb2.ModerateObjectRequest, retries: int = 3) -> bool:
def send_moderation_request(
self, request: moderation_pb2.ModerateObjectRequest, retries: int = 3
) -> bool:
for _ in range(retries):
try:
serialized_request = request.SerializeToString()

key = str(request.id).encode('utf-8')
key = str(request.id).encode("utf-8")

self.producer.produce(
topic=self.topic,
value=serialized_request,
key=key,
partition=-1,
callback=self._delivery_callback
on_delivery=self._delivery_callback,
)

self.producer.poll(0)

logger.debug(f"Moderation request queued: ID={request.id}, text_length={len(request.text)}")
logger.debug(
f"Moderation request queued: ID={request.id}, text_length={len(request.text)}"
)
return True

except BufferError as e:
Expand All @@ -48,7 +55,7 @@ def send_moderation_request(self, request: moderation_pb2.ModerateObjectRequest,
return False
return False

def _delivery_callback(self, err, msg):
def _delivery_callback(self, err: KafkaError | None, msg: Message) -> None:

if err is not None:
logger.error(f"Message delivery failed: {err}")
Expand All @@ -58,15 +65,17 @@ def _delivery_callback(self, err, msg):
f"partition={msg.partition()}, offset={msg.offset()}"
)

def flush(self, timeout: float = 10.0):
def flush(self, timeout: float = 10.0) -> None:

remaining = self.producer.flush(timeout)

if remaining > 0:
logger.warning(f"Flush incomplete: {remaining} messages still pending")
logger.warning(
f"Flush incomplete: {remaining} messages still pending"
)
else:
logger.debug("All messages flushed successfully")

def __del__(self):
if hasattr(self, 'producer'):
def __del__(self) -> None:
if hasattr(self, "producer"):
self.flush()
Loading
Loading