How to convert existing static STAC collection to stac-geoparquet?

[scottyhq]

I'm trying to create stac-geoparquet for some existing static STAC collections (~1000 items), the docs currently are focused on API search results and I'm wondering about a couple things

1. Is there a way to just use stacrs here (no pystac)?

path = 'collection.json'
c = pystac.read_file(path)
items = [i.to_dict() for i in c.get_items(recursive=True)]
stacrs.write(path.replace('.json', '.geoparquet'), items)

2. How to add collection metadata to the output parquet file?
   According to the spec

To make a stac-geoparquet file a fully self-contained representation, you can include the Collection JSON in the Parquet metadata. If present in the Parquet file metadata, the key must be stac:collection and the value must be a JSON string with the Collection JSON.

I'm not sure this can currently be done with stacrs.write / I'm finding this actually quite challenging to figure out 😅 . I tried the following:

items = [i.to_dict() for i in c.get_items(recursive=True)]
table = stacrs.to_arrow(items)
collection_metadata = {'stac:collection': json.dumps(collection.to_dict())}
updated_table = table.replace_schema_metadata(collection_metadata)
# how to write to geoparquet w/ additional metadata ??

Side note: I also tried stac-geoparquet, but it overwrites schema metadata to only have the geo key (https://github.com/stac-utils/stac-geoparquet/blob/121f6485fd188179f1c05d53ec26c69d2518c683/stac_geoparquet/arrow/_to_parquet.py#L144)

[gadomski]

Is there a way to just use stacrs here (no pystac)?

Not yet, but it wouldn't be hard to add ... we've got https://docs.rs/stac/latest/stac/struct.Node.html#method.into_values in the underlying Rust crate, so we'll just need to expose that in a stacrs.walk function: #68

How to add collection metadata to the output parquet file?

Unsupported at this time: stac-utils/rustac#428

[gadomski]

@scottyhq for the first one (walking all items) try https://pypi.org/project/stacrs/0.5.9b0/ (python -m pip install --pre --upgrade stacrs). Iterating all items would look like:

catalog = await stacrs.read("https://stacrs.test/catalog.json")
all_items = []
async for _, _, items in stacrs.walk(catalog):
    all_items.extend(items)