Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

BigQuery: Autofetch table schema on load if not provided #9108

Merged
merged 7 commits into from
Sep 4, 2019
Merged

BigQuery: Autofetch table schema on load if not provided #9108

Show file tree
Hide file tree
Changes from 5 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
8d207c7
Autofetch table schema on load if not provided
plamut Aug 26, 2019
0b9188d
Merge branch 'master' into iss-8142
plamut Aug 28, 2019
8ad6c56
Avoid fetching table schema if WRITE_TRUNCATE job
plamut Aug 28, 2019
2ea14ee
Skip dataframe columns list check
plamut Aug 28, 2019
2e1d137
Raise table NotFound in auto Pandas schema tests
plamut Aug 29, 2019
4e43428
Merge branch 'master' into iss-8142
plamut Sep 4, 2019
2d6e0f5
Use list_columns_and_indexes() for names list
plamut Sep 4, 2019
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
19 changes: 19 additions & 0 deletions bigquery/google/cloud/bigquery/client.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1547,6 +1547,25 @@ def load_table_from_dataframe(
if location is None:
location = self.location

# If table schema is not provided, we try to fetch the existing table
# schema, and check if dataframe schema is compatible with it - except
# for WRITE_TRUNCATE jobs, the existing schema does not matter then.
if (
not job_config.schema
and job_config.write_disposition != job.WriteDisposition.WRITE_TRUNCATE
):
try:
table = self.get_table(destination)
except google.api_core.exceptions.NotFound:
table = None
else:
# schema fields not present in the dataframe are not needed
job_config.schema = [
field
for field in table.schema
if field.name in set(dataframe.columns)
plamut marked this conversation as resolved.
Show resolved Hide resolved
]

job_config.schema = _pandas_helpers.dataframe_to_bq_schema(
dataframe, job_config.schema
)
Expand Down
86 changes: 80 additions & 6 deletions bigquery/tests/unit/test_client.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5279,15 +5279,23 @@ def test_load_table_from_file_bad_mode(self):
def test_load_table_from_dataframe(self):
from google.cloud.bigquery.client import _DEFAULT_NUM_RETRIES
from google.cloud.bigquery import job
from google.cloud.bigquery.schema import SchemaField

client = self._make_client()
records = [{"id": 1, "age": 100}, {"id": 2, "age": 60}]
dataframe = pandas.DataFrame(records)

get_table_patch = mock.patch(
"google.cloud.bigquery.client.Client.get_table",
autospec=True,
return_value=mock.Mock(
schema=[SchemaField("id", "INTEGER"), SchemaField("age", "INTEGER")]
),
)
load_patch = mock.patch(
"google.cloud.bigquery.client.Client.load_table_from_file", autospec=True
)
with load_patch as load_table_from_file:
with load_patch as load_table_from_file, get_table_patch:
client.load_table_from_dataframe(dataframe, self.TABLE_REF)

load_table_from_file.assert_called_once_with(
Expand All @@ -5314,15 +5322,23 @@ def test_load_table_from_dataframe(self):
def test_load_table_from_dataframe_w_client_location(self):
from google.cloud.bigquery.client import _DEFAULT_NUM_RETRIES
from google.cloud.bigquery import job
from google.cloud.bigquery.schema import SchemaField

client = self._make_client(location=self.LOCATION)
records = [{"id": 1, "age": 100}, {"id": 2, "age": 60}]
dataframe = pandas.DataFrame(records)

get_table_patch = mock.patch(
"google.cloud.bigquery.client.Client.get_table",
autospec=True,
return_value=mock.Mock(
schema=[SchemaField("id", "INTEGER"), SchemaField("age", "INTEGER")]
),
)
load_patch = mock.patch(
"google.cloud.bigquery.client.Client.load_table_from_file", autospec=True
)
with load_patch as load_table_from_file:
with load_patch as load_table_from_file, get_table_patch:
client.load_table_from_dataframe(dataframe, self.TABLE_REF)

load_table_from_file.assert_called_once_with(
Expand All @@ -5349,20 +5365,33 @@ def test_load_table_from_dataframe_w_client_location(self):
def test_load_table_from_dataframe_w_custom_job_config(self):
from google.cloud.bigquery.client import _DEFAULT_NUM_RETRIES
from google.cloud.bigquery import job
from google.cloud.bigquery.schema import SchemaField

client = self._make_client()
records = [{"id": 1, "age": 100}, {"id": 2, "age": 60}]
dataframe = pandas.DataFrame(records)
job_config = job.LoadJobConfig()
job_config = job.LoadJobConfig(
write_disposition=job.WriteDisposition.WRITE_TRUNCATE
)

get_table_patch = mock.patch(
"google.cloud.bigquery.client.Client.get_table",
autospec=True,
return_value=mock.Mock(
schema=[SchemaField("id", "INTEGER"), SchemaField("age", "INTEGER")]
),
)
load_patch = mock.patch(
"google.cloud.bigquery.client.Client.load_table_from_file", autospec=True
)
with load_patch as load_table_from_file:
with load_patch as load_table_from_file, get_table_patch as get_table:
client.load_table_from_dataframe(
dataframe, self.TABLE_REF, job_config=job_config, location=self.LOCATION
)

# no need to fetch and inspect table schema for WRITE_TRUNCATE jobs
assert not get_table.called

load_table_from_file.assert_called_once_with(
client,
mock.ANY,
Expand All @@ -5378,6 +5407,7 @@ def test_load_table_from_dataframe_w_custom_job_config(self):

sent_config = load_table_from_file.mock_calls[0][2]["job_config"]
assert sent_config.source_format == job.SourceFormat.PARQUET
assert sent_config.write_disposition == job.WriteDisposition.WRITE_TRUNCATE

@unittest.skipIf(pandas is None, "Requires `pandas`")
@unittest.skipIf(pyarrow is None, "Requires `pyarrow`")
Expand Down Expand Up @@ -5421,7 +5451,12 @@ def test_load_table_from_dataframe_w_automatic_schema(self):
"google.cloud.bigquery.client.Client.load_table_from_file", autospec=True
)

with load_patch as load_table_from_file:
get_table_patch = mock.patch(
plamut marked this conversation as resolved.
Show resolved Hide resolved
"google.cloud.bigquery.client.Client.get_table",
autospec=True,
side_effect=google.api_core.exceptions.NotFound("Table not found"),
)
with load_patch as load_table_from_file, get_table_patch:
client.load_table_from_dataframe(
dataframe, self.TABLE_REF, location=self.LOCATION
)
Expand Down Expand Up @@ -5449,6 +5484,40 @@ def test_load_table_from_dataframe_w_automatic_schema(self):
SchemaField("ts_col", "TIMESTAMP"),
)

@unittest.skipIf(pandas is None, "Requires `pandas`")
@unittest.skipIf(pyarrow is None, "Requires `pyarrow`")
def test_load_table_from_dataframe_unknown_table(self):
from google.cloud.bigquery.client import _DEFAULT_NUM_RETRIES

client = self._make_client()
records = [{"id": 1, "age": 100}, {"id": 2, "age": 60}]
dataframe = pandas.DataFrame(records)

get_table_patch = mock.patch(
"google.cloud.bigquery.client.Client.get_table",
autospec=True,
side_effect=google.api_core.exceptions.NotFound("Table not found"),
)
load_patch = mock.patch(
"google.cloud.bigquery.client.Client.load_table_from_file", autospec=True
)
with load_patch as load_table_from_file, get_table_patch:
# there should be no error
client.load_table_from_dataframe(dataframe, self.TABLE_REF)

load_table_from_file.assert_called_once_with(
client,
mock.ANY,
self.TABLE_REF,
num_retries=_DEFAULT_NUM_RETRIES,
rewind=True,
job_id=mock.ANY,
job_id_prefix=None,
location=None,
project=None,
job_config=mock.ANY,
)

@unittest.skipIf(pandas is None, "Requires `pandas`")
@unittest.skipIf(pyarrow is None, "Requires `pyarrow`")
def test_load_table_from_dataframe_struct_fields_error(self):
Expand Down Expand Up @@ -5741,6 +5810,11 @@ def test_load_table_from_dataframe_wo_pyarrow_custom_compression(self):
records = [{"id": 1, "age": 100}, {"id": 2, "age": 60}]
dataframe = pandas.DataFrame(records)

get_table_patch = mock.patch(
"google.cloud.bigquery.client.Client.get_table",
autospec=True,
side_effect=google.api_core.exceptions.NotFound("Table not found"),
)
load_patch = mock.patch(
"google.cloud.bigquery.client.Client.load_table_from_file", autospec=True
)
Expand All @@ -5749,7 +5823,7 @@ def test_load_table_from_dataframe_wo_pyarrow_custom_compression(self):
dataframe, "to_parquet", wraps=dataframe.to_parquet
)

with load_patch, pyarrow_patch, to_parquet_patch as to_parquet_spy:
with load_patch, get_table_patch, pyarrow_patch, to_parquet_patch as to_parquet_spy:
client.load_table_from_dataframe(
dataframe,
self.TABLE_REF,
Expand Down