feat(sentry_integration): discovered load_runtime exec's the customer's app

[victoria-yining-huang]

The goal of this ticket is to report streams specific errors to our own sentry project, while a customer application that imports sentry_streams should still receive their application specific errors in their separate sentry project.

Intended behaviour:

client app code i tested with:

import uuid
from datetime import date
from functools import partial
from typing import Sequence

from sentry_kafka_schemas.schema_types.events_v1 import EventStreamMessage
from sentry_streams.pipeline import Batch, Map, ParquetSerializer, streaming_source
from sentry_streams.pipeline.message import Message as StreamsMessage
from sentry_streams.pipeline.pipeline import GCSSink

from consumer.streaming_platform.processing_strategies.errors import ErrorsProcessorGCS
import sentry_sdk

sentry_sdk.init(
    dsn="https://20699d75a9ff7c55b2e6a8c5abc55066@o1.ingest.us.sentry.io/4510506713284608",
    # Add data like request headers and IP for users,
    # see https://docs.sentry.io/platforms/python/data-management/data-collected/ for more info
    send_default_pii=True,
)

my_processor = ErrorsProcessorGCS()


def processing_func(
    processor: ErrorsProcessorGCS, msg: StreamsMessage[Sequence[EventStreamMessage]]
):
    return processor.process_streaming_batch(msg)


def generate_files() -> str:
    file_name = (
        "errors-gcs"
        + "/dt="
        + str(date.today())
        + "/data-"
        + str(uuid.uuid4())
        + ".parquet"
    )
    return file_name


pipeline = streaming_source(name="myinput", stream_name="events")

parsed_batches = pipeline.apply(Batch(name="mybatch", batch_size=2))

transformed_batches = parsed_batches.apply(
    Map(name="process", function=partial(processing_func, my_processor))
)
transformed_batches.apply(
    ParquetSerializer(
        name="serializer", schema_fields=my_processor.schema_fields_sentrystreams
    )
).sink(
    GCSSink(name="mysink", bucket="arroyo-artifacts", object_generator=generate_files)
)