Skip to content
Open
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
2 changes: 1 addition & 1 deletion tests/unit/test_client.py
Original file line number Diff line number Diff line change
Expand Up @@ -1716,7 +1716,7 @@ def statement_uri(token):
"encoding": "json",
"segments": [{
"type": "inline",
"metadata": {"uncompressedSize": "5", "segmentSize": "5"},
"metadata": {"uncompressedSize": 5, "segmentSize": 5},
"data": encoded_rows,
}],
},
Expand Down
39 changes: 36 additions & 3 deletions tests/unit/test_client_spooling.py
Original file line number Diff line number Diff line change
Expand Up @@ -14,15 +14,19 @@
from unittest import mock

import pytest
import zstandard

from trino.client import _RequestHeartbeat
from trino.client import ClientSession
from trino.client import DecodableSegment
from trino.client import InlineSegment
from trino.client import JsonQueryDataDecoder
from trino.client import SegmentIterator
from trino.client import SpooledSegment
from trino.client import TrinoQuery
from trino.client import TrinoRequest
from trino.client import ZStdQueryDataDecoder
from trino.mapper import NoOpRowMapper


def _mock_trino_request():
Expand Down Expand Up @@ -176,7 +180,7 @@ def _spooled_fetch_response():
"segments": [
{
"type": "inline",
"metadata": {"uncompressedSize": "10", "segmentSize": "10"},
"metadata": {"uncompressedSize": 10, "segmentSize": 10},
"data": "",
}
],
Expand Down Expand Up @@ -210,7 +214,7 @@ def __init__(self, name):
"type": "spooled",
"uri": f"http://storage/{name}",
"ackUri": f"http://storage/{name}/ack",
"metadata": {"uncompressedSize": "10", "segmentSize": "10"},
"metadata": {"uncompressedSize": 10, "segmentSize": 10},
},
request=None,
)
Expand Down Expand Up @@ -261,7 +265,7 @@ def _spooled_segment_with_headers(coordinator_host, custom_headers):
"uri": "https://coordinator/v1/spooled/download/seg1",
"ackUri": "https://coordinator/v1/spooled/ack/seg1",
"headers": {"X-Trino-Spooling-Token": ["token-abc"]},
"metadata": {"segmentSize": "1", "uncompressedSize": "1"},
"metadata": {"segmentSize": 1, "uncompressedSize": 1},
}
request = TrinoRequest(
host="coordinator",
Expand Down Expand Up @@ -327,3 +331,32 @@ def fake_get(uri, headers=None, **kwargs):
segment._send_spooling_request(segment.uri)

assert recorded["headers"]["X-Trino-Spooling-Token"] == "token-abc"


def test_zstd_decoder_decodes_compressed_segment():
rows = [[1, "a"], [2, "b"]]
uncompressed = json.dumps(rows).encode("utf8")
compressed = zstandard.ZstdCompressor().compress(uncompressed)

metadata = {
"segmentSize": len(compressed),
"uncompressedSize": len(uncompressed),
}
decoder = ZStdQueryDataDecoder(JsonQueryDataDecoder(NoOpRowMapper()))

assert decoder.decode(compressed, metadata) == rows


def test_zstd_decoder_rejects_a_segment_whose_size_does_not_match():
rows = [[1, "a"]]
uncompressed = json.dumps(rows).encode("utf8")
compressed = zstandard.ZstdCompressor().compress(uncompressed)

metadata = {
"segmentSize": len(compressed) + 1,
"uncompressedSize": len(uncompressed),
}
decoder = ZStdQueryDataDecoder(JsonQueryDataDecoder(NoOpRowMapper()))

with pytest.raises(RuntimeError, match="Expected to read"):
decoder.decode(compressed, metadata)
6 changes: 3 additions & 3 deletions trino/client.py
Original file line number Diff line number Diff line change
Expand Up @@ -1196,8 +1196,8 @@ class _SpooledProtocolResponseTO(TypedDict):


class _SegmentMetadataTO(TypedDict):
uncompressedSize: str
segmentSize: str
uncompressedSize: int
segmentSize: int


class _SegmentTO(_SegmentMetadataTO):
Expand Down Expand Up @@ -1583,5 +1583,5 @@ def decompress(self, data: bytes, metadata: _SegmentMetadataTO) -> bytes:
class Lz4QueryDataDecoder(CompressedQueryDataDecoder):
def decompress(self, data: bytes, metadata: _SegmentMetadataTO) -> bytes:
expected_uncompressed_size = metadata["uncompressedSize"]
decoded_bytes = lz4.block.decompress(data, uncompressed_size=int(expected_uncompressed_size))
decoded_bytes = lz4.block.decompress(data, uncompressed_size=expected_uncompressed_size)
return decoded_bytes
Loading