S14Amazonas preview

[JanssenBrm]

No description provided.

[JanssenBrm]

Preview for #245

[algorithm-services-catalogue]

🔍 Catalogue's Preview Site Deployed

Your changes have been deployed to the preview site:

🔗 Preview URL: https://esa-apex.github.io/apex-algorithms-catalogue-web/pr-preview/pr-372/

This preview will be updated automatically when you push new changes to your PR.

[sivasanarul]

I am trying to update the resto api inside the udf to Odata api. I am using this function which works offline when I call it like this
feats = fetch_s1_features(bbox_4326, temporal_dict["start"], temporal_dict["end"])
But it doesnt work when I use it inside udf:
Could you help me fix this:

temporal_dict = {
"start": pd.Timestamp("2020-01-17 00:00:00"),
"end": pd.Timestamp("2020-05-04 00:00:00"),
}
bbox_4326 = [-11.753685105, -55.165335994, -12.73714945, -54.158521354]
ODATA_URL = "https://datahub.creodias.eu/odata/v1/Products"
def fetch_s1_features(bbox: List[float], start_iso: str, end_iso: str) -> List[dict]:
"""
Query the Copernicus Dataspace OData API for Sentinel-1 features within bbox and datetime range.
Returns features in a format compatible with the rest of the code (OpenSearch-like structure).
"""
# bbox is [south, west, north, east] - convert to WKT polygon
south, west, north, east = bbox
polygon_wkt = f"POLYGON(({west} {south},{east} {south},{east} {north},{west} {north},{west} {south}))"

# Build OData filter parts
filters = [
    "Collection/Name eq 'SENTINEL-1'",
    f"ContentDate/Start gt {start_iso.strftime('%Y-%m-%dT00:00:00.000Z')}",
    f"ContentDate/Start lt {end_iso.strftime('%Y-%m-%dT23:59:59.999Z')}",
    f"OData.CSC.Intersects(area=geography'SRID=4326;{polygon_wkt}')",
    "Attributes/OData.CSC.StringAttribute/any(att:att/Name eq 'productType' and att/OData.CSC.StringAttribute/Value eq 'IW_GRDH_1S-COG')",
]

filter_str = " and ".join(filters)

params = {
    "$filter": filter_str,
    "$top": 1000,
    "$expand": "Attributes",
    "$orderby": "ContentDate/Start asc",
}

r = requests.get(ODATA_URL, params=params, timeout=60)
r.raise_for_status()

odata_products = r.json().get("value", [])

# Transform OData response to OpenSearch-like format for compatibility
features = []
for prod in odata_products:
    # Extract orbitDirection from Attributes
    orbit_direction = "UNKNOWN"
    for attr in prod.get("Attributes", []):
        if attr.get("Name") == "orbitDirection":
            orbit_direction = attr.get("Value", "UNKNOWN")
            break
    
    # Build compatible feature structure
    feature = {
        "properties": {
            "title": prod.get("Name", ""),
            "orbitDirection": orbit_direction,
        },
        "geometry": prod.get("GeoFootprint", {}),
    }
    features.append(feature)
logger.info(f"-- fetched features {features} --")
return features

[JanssenBrm]

@VictorVerhaert - Could you help with @sivasanarul question? @sivasanarul - maybe it is best to iterate on the GitHub repo where the actual code is located through a GitHub issue? Otherwise, we will have too many locations where discussions are tracked.

[sivasanarul]

@VictorVerhaert https://github.com/Gisat/dev-S14amazonas/blob/main/apex_integration_dev.py
https://github.com/Gisat/dev-S14amazonas/blob/main/udf_apex_S1backscatter_changedetection.py

These are the relevant scripts that is updated with the Odata api. You already make a pull request there.

this script is used to test only the fetch_s1_features function:
apex_udf_call.py