thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-10 09:32:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #33037 from kssenii/fix-s3-empty-file

Browse Source 
Fix s3 table function reading empty file
This commit is contained in:
Kseniia Sumarokova 2021-12-22 18:17:29 +03:00 committed by GitHub
commit 01f2e4c00d
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 20 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/IO/ReadBufferFromS3.cpp
View File

@ -239,7 +239,8 @@ std::unique_ptr<ReadBuffer> ReadBufferFromS3::initialize()
}
else
{
req.SetRange(fmt::format("bytes={}-", offset));
if (offset)
req.SetRange(fmt::format("bytes={}-", offset));
LOG_TEST(log, "Read S3 object. Bucket: {}, Key: {}, Offset: {}", bucket, key, offset);
}

20
tests/integration/test_storage_s3/test.py
View File

@ -11,6 +11,7 @@ import helpers.client
import pytest
from helpers.cluster import ClickHouseCluster, ClickHouseInstance, get_instances_dir
from helpers.network import PartitionManager
from helpers.test_tools import exec_query_with_retry
MINIO_INTERNAL_PORT = 9001
@ -809,7 +810,7 @@ def test_seekable_formats(started_cluster):
assert(int(result) == 5000000)
table_function = f"s3(s3_orc, structure='a Int32, b String', format='ORC')"
instance.query(f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(5000000)")
exec_query_with_retry(instance, f"insert into table function {table_function} SELECT number, randomString(100) FROM numbers(5000000)")
result = instance.query(f"SELECT count() FROM {table_function}")
assert(int(result) == 5000000)
@ -832,7 +833,7 @@ def test_seekable_formats_url(started_cluster):
assert(int(result) == 5000000)
table_function = f"s3(s3_orc, structure='a Int32, b String', format='ORC')"
instance.query(f"insert into table function {table_function} select number, randomString(100) from numbers(5000000)")
exec_query_with_retry(instance, f"insert into table function {table_function} select number, randomString(100) from numbers(5000000)")
table_function = f"url('http://{started_cluster.minio_host}:{started_cluster.minio_port}/{bucket}/test_orc', 'ORC', 'a Int32, b String')"
result = instance.query(f"SELECT count() FROM {table_function}")
@ -842,3 +843,18 @@ def test_seekable_formats_url(started_cluster):
result = instance.query(f"SELECT formatReadableSize(memory_usage) FROM system.query_log WHERE startsWith(query, 'SELECT count() FROM url') AND memory_usage > 0 ORDER BY event_time desc")
print(result[:3])
assert(int(result[:3]) < 200)
def test_empty_file(started_cluster):
bucket = started_cluster.minio_bucket
instance = started_cluster.instances["dummy"]
name = "empty"
url = f'http://{started_cluster.minio_ip}:{MINIO_INTERNAL_PORT}/{bucket}/{name}'
minio = started_cluster.minio_client
minio.put_object(bucket, name, io.BytesIO(b""), 0)
table_function = f"s3('{url}', 'CSV', 'id Int32')"
result = instance.query(f"SELECT count() FROM {table_function}")
assert(int(result) == 0)