Skip to content

Commit 91af0f4

Browse files
authored
Merge pull request #417 from mabel-dev/0.6.26
0.6.26
2 parents 33ea9be + 4bc7adf commit 91af0f4

File tree

5 files changed

+8
-11
lines changed

5 files changed

+8
-11
lines changed

mabel/adapters/google/google_cloud_storage_writer.py

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -46,6 +46,8 @@ def commit(self, byte_data, override_blob_name=None):
4646
# name from the path builder
4747
if override_blob_name:
4848
blob_name = override_blob_name
49+
if blob_name.startswith(self.bucket + "/"):
50+
blob_name = blob_name[len(self.bucket) + 1 :]
4951
else:
5052
blob_name = self._build_path()
5153

mabel/data/internals/group_by.py

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -73,8 +73,7 @@ def _map(self, collect_columns):
7373
for record in self._dictset:
7474
try:
7575
group_key: int = xxh3_64_intdigest(
76-
"".join([str(record[column]) for column in self._columns]),
77-
HASH_SEED
76+
"".join([str(record[column]) for column in self._columns]), HASH_SEED
7877
)
7978
except KeyError:
8079
group_key: int = xxh3_64_intdigest(

mabel/data/writers/internals/blob_writer.py

Lines changed: 0 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -23,7 +23,6 @@ class BlobWriter(object):
2323
# this variable outside the __init__.
2424
buffer = bytearray()
2525
byte_count = 0
26-
manifest = {}
2726

2827
def __init__(
2928
self,
@@ -141,7 +140,6 @@ def commit(self):
141140
if self.records_in_buffer > 0:
142141
lock = threading.Lock()
143142

144-
summary = None
145143
try:
146144
lock.acquire(blocking=True, timeout=10)
147145

@@ -155,10 +153,6 @@ def commit(self):
155153
)
156154

157155
pytable = self.wal.arrow()
158-
try:
159-
summary = self.wal.profile.to_dicts()
160-
except Exception as e:
161-
print(f"[MABEL] Unable to profile morsel - {type(e).__name__} - {e}")
162156

163157
# if we have a schema, make effort to align the parquet file to it
164158
if self.schema:
@@ -179,7 +173,6 @@ def commit(self):
179173
committed_blob_name = self.inner_writer.commit(
180174
byte_data=write_buffer, override_blob_name=None
181175
)
182-
self.manifest[committed_blob_name] = summary
183176

184177
if "BACKOUT" in committed_blob_name:
185178
get_logger().warning(

tests/test_data_group_by.py

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -130,7 +130,9 @@ def test_gappy_set():
130130
{"AVG(key)": 1.0, "value": "one"},
131131
]
132132

133-
assert set(tuple(sorted(d.items())) for d in gs) == set(tuple(sorted(d.items())) for d in expected)
133+
assert set(tuple(sorted(d.items())) for d in gs) == set(
134+
tuple(sorted(d.items())) for d in expected
135+
)
134136

135137

136138
if __name__ == "__main__": # pragma: no cover

tests/test_writer_batch_writer.py

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -6,6 +6,7 @@
66
import pytest
77

88
sys.path.insert(1, os.path.join(sys.path[0], ".."))
9+
910
from mabel.adapters.disk import DiskReader, DiskWriter
1011
from mabel.data import BatchWriter
1112
from mabel.data import Reader
@@ -215,7 +216,7 @@ def get_data():
215216

216217

217218
if __name__ == "__main__": # pragma: no cover
218-
from tests.helpers.runner import run_tests
219+
from helpers.runner import run_tests
219220

220221
test_writer_without_schema_parquet()
221222
run_tests()

0 commit comments

Comments
 (0)