Skip to content

fix: use random table for read_pandas #192

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

Merged
merged 5 commits into from
Nov 10, 2023
Merged

fix: use random table for read_pandas #192

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
a436194
fix: use random table for read_pandas
ashleyxuu Nov 10, 2023
03606cd
fix: default to 7 days expiration for `read_csv`, `read_json`, `read_…
tswast Nov 10, 2023
96b8b1d
fix: use random table for read_pandas
ashleyxuu Nov 10, 2023
2d72596
fix: use the real table expression
ashleyxuu Nov 10, 2023
6915c54
Merge remote-tracking branch 'origin/read_pandas-random-table' into r…
ashleyxuu Nov 10, 2023
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 bigframes/core/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -165,7 +165,7 @@ def cached(self, cluster_cols: typing.Sequence[str]) -> ArrayValue:
ibis_expr = compiled_value._to_ibis_expr(
ordering_mode="unordered", expose_hidden_cols=True
)
tmp_table = self.session._ibis_to_session_table(
tmp_table = self.session._ibis_to_temp_table(
ibis_expr, cluster_cols=cluster_cols, api_name="cached"
)

Expand Down
30 changes: 16 additions & 14 deletions bigframes/session/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,6 @@
Tuple,
Union,
)
import uuid
import warnings

import google.api_core.client_info
Expand Down Expand Up @@ -836,11 +835,20 @@ def _read_bigquery_load_job(
)

self._start_generic_job(load_job)
table_id = f"{table.project}.{table.dataset_id}.{table.table_id}"

# Update the table expiration so we aren't limited to the default 24
# hours of the anonymous dataset.
table_expiration = bigquery.Table(table_id)
table_expiration.expires = (
datetime.datetime.now(datetime.timezone.utc) + constants.DEFAULT_EXPIRATION
)
self.bqclient.update_table(table_expiration, ["expires"])

# The BigQuery REST API for tables.get doesn't take a session ID, so we
# can't get the schema for a temp table that way.
return self.read_gbq_table(
f"{table.project}.{table.dataset_id}.{table.table_id}",
table_id,
index_col=index_col,
col_order=col_order,
)
Expand Down Expand Up @@ -977,7 +985,7 @@ def _read_pandas(
job_config.clustering_fields = cluster_cols
job_config.labels = {"bigframes-api": api_name}

load_table_destination = self._create_session_table()
load_table_destination = bigframes_io.random_table(self._anonymous_dataset)
load_job = self.bqclient.load_table_from_dataframe(
pandas_dataframe_copy,
load_table_destination,
Expand All @@ -990,8 +998,9 @@ def _read_pandas(
total_ordering_columns=frozenset([ordering_col]),
integer_encoding=IntegerEncoding(True, is_sequential=True),
)
table_expression = self.ibis_client.sql(
f"SELECT * FROM `{load_table_destination.table_id}`"
table_expression = self.ibis_client.table(
load_table_destination.table_id,
database=f"{load_table_destination.project}.{load_table_destination.dataset_id}",
)

# b/297590178 Potentially a bug in bqclient.load_table_from_dataframe(), that only when the DF is empty, the index columns disappear in table_expression.
Expand Down Expand Up @@ -1269,13 +1278,6 @@ def _check_file_size(self, filepath: str):
"for large files to avoid loading the file into local memory."
)

def _create_session_table(self) -> bigquery.TableReference:
table_name = f"{uuid.uuid4().hex}"
dataset = bigquery.Dataset(
bigquery.DatasetReference(self.bqclient.project, "_SESSION")
)
return dataset.table(table_name)

def _create_empty_temp_table(
self,
schema: Iterable[bigquery.SchemaField],
Expand Down Expand Up @@ -1310,7 +1312,7 @@ def _create_sequential_ordering(
ibis.row_number().cast(ibis_dtypes.int64).name(default_ordering_name)
)
table = table.mutate(**{default_ordering_name: default_ordering_col})
table_ref = self._ibis_to_session_table(
table_ref = self._ibis_to_temp_table(
table,
cluster_cols=list(index_cols) + [default_ordering_name],
api_name=api_name,
Expand All @@ -1326,7 +1328,7 @@ def _create_sequential_ordering(
)
return table, ordering

def _ibis_to_session_table(
def _ibis_to_temp_table(
self,
table: ibis_types.Table,
cluster_cols: Iterable[str],
Expand Down