Daniel K

### Description There are multiple situations in datasets calling for introducing a new file object. For one example, reading a dataset like Google OpenImage produces a URI column which needs...

### Description It is a common pattern to simply read the contents of a storage file by filename. Currently this requires a lengthy construct like this: ``` next(DataChain.from_storage('gs://datachain-demo/dogs-and-cats/cat.1009.jpg', type='image').collect("file")).read() ```...

### Description So far we have focused on supporting only cloud and local storages. It seems like there is a need to support the https:// transport as well (optional username/password)

Aggregator and generator inputs should be Iterators, not lists

1

### Description Here is a sample generator from LLM tutorial: ``` from typing import List class Dialog(BaseModel): id: int text: str def text_block(id: List[int], sender: List[str], text: List[str]) -> Dialog:...

### Description There is a number of improvements that we need to implement at some point: 1. currently, the code does not validate the data against the automatically created schema....