Beto Dealmeida
Beto Dealmeida
### SUMMARY This PR introduces a new SQLAlchemy dialect, `superset://`, together with a corresponding DB engine spec. The SQLAlchemy dialect is built on top of [shillelagh](https://github.com/betodealmeida/shillelagh/), a framework for building...
### SUMMARY This PR adds a method called `get_metrics` to the `Database` model and the base DB engine spec. It allows DB engine specs to return custom metrics for a...
```python from typing import List, TypeVar from pytypes import is_subtype T = TypeVar("T") is_subtype(List[int], List[T]) ``` Fails with (Python 3.9): ``` Traceback (most recent call last): File "/Users/beto/Projects/datajunction/bug.py", line 5,...
Is it possible to infer the return type based on the input, for generic functions? Eg, with this [classic example](https://mypy.readthedocs.io/en/stable/generics.html#generic-functions): ```python from typing import TypeVar, Sequence T = TypeVar('T') #...
PyHive defines a "trino" entry point for a SQLAlchemy dialect: https://github.com/dropbox/PyHive/blob/master/setup.py#L72 If you have both PyHive and `sqlalchemy-trino` installed both will register the same dialect, and each request will randomly...
The `CountStar` processor should handle more cases, eg: ```sql SELECT country , COUNT(*) FROM table WHERE name = 'John' GROUP BY country HAVING COUNT(*) > 100 ```
**Is your feature request related to a problem? Please describe.** Adapters could benefit from knowing which columns were requested, so they can fetch less data. **Describe the solution you'd like**...
Implement adapters: - [x] WeatherAPI - [x] CSV files - [x] Google Sheets - [x] Socrata - [x] Pandas - [x] Datasette - [x] Github repositories - [x] [XML](https://github.com/betodealmeida/shillelagh/issues/389) -...
For some adapters like Pandas it would be more efficient to return a buffer that points to the data, instead of an iterator.
# Summary # Testing instructions