Hive dialect name error

[hsheth2]

Describe the bug

The Hive dialect name is b'hive', which is of type bytes and not str. This causes issues in the dialect comparison here:

great_expectations/great_expectations/execution_engine/sqlalchemy_dialect.py

Lines 27 to 30 in 1af478a

	def __eq__(self, other: Union[str, GESqlDialect]):
	if isinstance(other, str):
	return self.value.lower() == other.lower()
	return self.value.lower() == other.value.lower()

This issue applies to GE versions 0.15.23+.

This appears to be a regression introduced by #5980. Note that b'foo' == 'foo' is False but hash(b'foo') == hash('foo') is True, which is probably why this code worked before.

To Reproduce

We're running into this issue from the DataHub project, which depends on GE as a library. Hopefully the stack trace below has enough context, but I can try to provide more if needed.

Stack trace that I see:

Traceback (most recent call last):
  File "/home/mayuri/git/acryldata/datahub-fork/metadata-ingestion/src/datahub/ingestion/source/ge_data_profiler.py", line 906, in _generate_single_profile
    batch = self._get_ge_dataset(
  File "/home/mayuri/git/acryldata/datahub-fork/metadata-ingestion/src/datahub/ingestion/source/ge_data_profiler.py", line 970, in _get_ge_dataset
    batch = ge_context.data_context.get_batch(
  File "/home/mayuri/git/acryldata/datahub-fork/metadata-ingestion/venv/lib/python3.8/site-packages/great_expectations/data_context/data_context/base_data_context.py", line 1153, in get_batch
    return self._get_batch_v2(
  File "/home/mayuri/git/acryldata/datahub-fork/metadata-ingestion/venv/lib/python3.8/site-packages/great_expectations/data_context/data_context/base_data_context.py", line 859, in _get_batch_v2
    return validator.get_dataset()
  File "/home/mayuri/git/acryldata/datahub-fork/metadata-ingestion/venv/lib/python3.8/site-packages/great_expectations/validator/validator.py", line 2427, in get_dataset
    return self.expectation_engine(
  File "/home/mayuri/git/acryldata/datahub-fork/metadata-ingestion/venv/lib/python3.8/site-packages/great_expectations/dataset/sqlalchemy_dataset.py", line 568, in __init__
    if self.engine.dialect.name.lower() == GESqlDialect.BIGQUERY:
  File "/home/mayuri/git/acryldata/datahub-fork/metadata-ingestion/venv/lib/python3.8/site-packages/great_expectations/execution_engine/sqlalchemy_dialect.py", line 31, in __eq__
    return self.value.lower() == other.value.lower()

Expected behavior

No crashes.

Environment (please complete the following information):

• Operating System: Linux
• Great Expectations Version: 0.15.23+ (repro'd on .23, .25, and .26)

Additional context

DataHub introduced a temporary fix: datahub-project/datahub#5980.

[AFineDayFor]

Howdy @hsheth2 👋 thanks for reaching out and raising this with us 🙇

We're raising this within the team. Thank you so very much for helping us in identifying this 🦠🔬

[NathanFarmer]

The fix for this is merged into develop and will be included as part of todays release (0.15.27). Thanks @hsheth2!

[hsheth2]

Thanks @NathanFarmer for the quick fix!