Thierry Jean

Hi @coisnepe, from your report you're right that setting these columns hints should enforce the right `pyarrow` decimal precision. ```python {'name': 'short_term_booking_price_with_service_fee', 'nullable': True, 'data_type': 'decimal', 'precision': 38, 'scale': 18}...

Closing. Refer to the linked issue to track progress.

Thank you for the thorough description and for sharing your implementation. At the moment, it makes sense for these features to live outside of the main `dlt` repository as it...

Closing. Will re-open when a clear fix is decided

@rudolfix That all makes sense. I was also thinking we could modify the "transformation resource spec" class. > listing the regular resource does not add _dlt_id I got something wrong...

Because `ReadableDBAPIDataset.__init__(dataset_type=...)` takes the default value `"auto"`, most tests are contaminated by the presence/absence of `ibis-framework` in venv. The explicit value `default` should be passed in tests. We should be...

I would push against subclassing `TypeMapperImpl` for` DataTypeMapper(ABC)` - SQLGlot is not a destination `to_destination_type()`, `from_destination_type()` - in `from_destination_type(db_type, precision, scale)`, there's no situation where `precision` and `scale` should be...

> propagate add_limit step to the upstream resource via connected pipe. this is easy. but will limit parent resource so it may feel a little non intuitive. also you cannot...

Another avenue, which may be simpler (or not), would be that `.add_limit()` applies to the Paginator

Happy to see the refactor `to_sqlglot()` to `to_sqlglot_columns_definition()` merged! Though, maybe the issue should be reopened for the other methods: - `to_sqlglot(self) -> SQLGlotSchema:` - `from_sqlglot(cls, sqlglot_schema: SQLGlotSchema) -> Schema:`...