Documentation Index
Fetch the complete documentation index at: https://mintlify.com/terrafloww/rasteret/llms.txt
Use this file to discover all available pages before exploring further.
Class Definition
from rasteret import BandRegistry
Class Methods
register()
BandRegistry.register(collection_id: str, band_map: dict[str, str]) -> None
Collection identifier (e.g. "my-collection").
Mapping of band codes to band names (e.g. {"B1": "red", "B2": "green"}).
get()
BandRegistry.get(
collection_id: str,
default: dict[str, str] | None = None
) -> dict[str, str]
default
dict[str, str] | None
default:"None"
Default value to return if collection ID is not found.
The registered band mapping, or the default value if not found.
list_registered()
BandRegistry.list_registered() -> list[str]
List of registered collection identifiers.
Examples
Register Custom Bands
from rasteret import BandRegistry
# Register band mapping for a custom collection
BandRegistry.register(
"my-collection",
{
"B1": "red",
"B2": "green",
"B3": "blue",
"B4": "nir"
}
)
Register Multi-Band Asset
# For a collection with multi-band GeoTIFFs
BandRegistry.register(
"drone-imagery",
{
"1": "red",
"2": "green",
"3": "blue",
"4": "nir",
"5": "rededge"
}
)
Lookup Bands
# Get band mapping for a collection
sentinel_bands = BandRegistry.get("sentinel-2-l2a")
print(sentinel_bands)
# {
# "B01": "coastal",
# "B02": "blue",
# "B03": "green",
# "B04": "red",
# ...
# }
# With default fallback
custom_bands = BandRegistry.get(
"unknown-collection",
default={"B1": "band1"}
)
List All Collections
# See all registered collections
collections = BandRegistry.list_registered()
print(collections)
# ['sentinel-2-l2a', 'landsat-c2-l2', 'my-collection']
Using with build_from_stac
import rasteret
# Register bands before building
BandRegistry.register(
"my-stac-collection",
{
"red": "B04",
"green": "B03",
"blue": "B02"
}
)
# Band map is used during ingestion
collection = rasteret.build_from_stac(
name="test",
stac_api="https://api.example.com",
collection="my-stac-collection",
data_source="my-stac-collection",
bbox=bbox,
date_range=date_range
)
# Access registered bands
data = collection.get_numpy(
geometries=bbox,
bands=["red", "green", "blue"]
)
Override Built-in Mapping
# Override Sentinel-2 bands with custom names
BandRegistry.register(
"sentinel-2-l2a",
{
"B02": "blue_10m",
"B03": "green_10m",
"B04": "red_10m",
"B08": "nir_10m"
}
)
Built-in Mappings
Sentinel-2 L2A
{
"B01": "coastal",
"B02": "blue",
"B03": "green",
"B04": "red",
"B05": "rededge1",
"B06": "rededge2",
"B07": "rededge3",
"B08": "nir",
"B8A": "nir08",
"B09": "nir09",
"B11": "swir16",
"B12": "swir22",
"SCL": "scl"
}
Landsat Collection 2 Level-2
{
"B1": "coastal",
"B2": "blue",
"B3": "green",
"B4": "red",
"B5": "nir08",
"B6": "swir16",
"B7": "swir22",
"qa_aerosol": "qa_aerosol",
"qa_pixel": "qa_pixel",
"qa_radsat": "qa_radsat"
}
Integration Points
The BandRegistry is used by:
- build_from_stac(): Maps STAC asset keys to band codes during ingestion
- build_from_table(): Maps band codes to asset URLs when constructing the assets column
- Collection methods: Validates band codes in
get_numpy(), get_xarray(), etc.
Notes
- Band mappings are stored in a class-level registry
- Built-in collections are pre-registered at import time
- Collection IDs are case-sensitive
- You can override built-in mappings by re-registering
- Band codes are arbitrary strings (e.g. “B04”, “red”, “band_4”)
- Used primarily for STAC asset key resolution during ingestion