ValueError: dictionary update sequence element #0 when using dict on a FeatureCollection

[rsemlal-murmuration]

Description

We have an error ValueError: dictionary update sequence element #0 has length X; 2 is required raised when using dict() on a valid instance of FeatureCollection.

How to reproduce

from geojson_pydantic import FeatureCollection

fc = FeatureCollection.model_validate({'type': 'FeatureCollection', 'features': [{'type': 'Feature', 'properties': {}, 'geometry': {'type': 'Point', 'coordinates': [1, 2]}}]})

dict(fc) # <-- here

Expected

The call to dict should return a dict

Observed

The call to dict raises a ValueError

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: dictionary update sequence element #0 has length 5; 2 is required

Why is this an issue

We are using FeatureCollection as an input parameter type for a Prefect flow.

It looks something like this

# Flow definition
@flow(name="my_flow")
async def my_flow(area: FeatureCollection):
   return "yes"

my_flow(....) # Flow usage

It turns out that Prefect, since version 3.2.16 we believe, start applying a dict on flows input parameters if these parameters are Pydantic models (see here)

Since the dict fails on FeatureCollections, it makes unusable in this context.

[vincentsarago]

🤔 I wonder if this is due to

geojson-pydantic/geojson_pydantic/features.py

Lines 42 to 52 in 11ff35b

	def __iter__(self) -> Iterator[Feat]: # type: ignore [override]
	"""iterate over features"""
	return iter(self.features)
	def __len__(self) -> int:
	"""return features length"""
	return len(self.features)
	def __getitem__(self, index: int) -> Feat:
	"""get feature at a given index"""
	return self.features[index]

ref python/cpython#116938

this is unfortunate, sadly I'm not quite sure we want to remove those method in geojson-pydantic

I don't know prefect by IMO they should be using model_dump() instead of dict() 🤷

[rsemlal-murmuration]

Thanks for the quick resolution, everything works fine with geojson-pydantic 2.0.0 👌