BUG: Add support to replace partitions in date-partitioned tables (#43) #124
Conversation
Codecov Report
@@ Coverage Diff @@
## master #124 +/- ##
===========================================
- Coverage 75.7% 30.96% -44.75%
===========================================
Files 8 8
Lines 1626 1634 +8
===========================================
- Hits 1231 506 -725
- Misses 395 1128 +733
Continue to review full report at Codecov.
|
mremes
force-pushed
the
dpt-support
branch
3 times, most recently
from
19d0d27 to
3a4f4a7
Compare
mremes
changed the title
| @@ -699,8 +695,9 @@ def delete_and_recreate_table(self, dataset_id, table_id, table_schema): | |||
| table = _Table(self.project_id, dataset_id, | |||
| private_key=self.private_key) | |||
| table.delete(table_id) | |||
There was a problem hiding this comment.
Does this mean that BigQuery lets you delete just a single partition using the table delete operation? I haven't tried that.
There was a problem hiding this comment.
I tried it with the decorated table name and it worked as intended.
| return True | ||
| except NotFound: | ||
| return False | ||
| except self.http_error as ex: | ||
| self.process_http_error(ex) | ||
|
|
||
| def create(self, table_id, schema): | ||
| def create(self, table_id, schema, date_partitioned=False): |
There was a problem hiding this comment.
Rather than pass in date_partitioned here, I'd like to see an argument similar to the configuration argument in read_gbq which can take a dictionary in the JSON API format. That way options like timePartitioning can be set.
There was a problem hiding this comment.
That makes sense to me! I'll look into including the configuration in the to_gbq call to enable this and rest of the API configuration.
|
Closing as pandas-gbq has changed quite a bit since this PR was opened. |
Relates to issue #43
Changes:
Needs: