Skip to content

Convert Shapefile to GeoParquet

Convert ESRI shapefiles (a .shp file, a .zip archive containing one or more, or an HTTP(S) URL of such a .zip) into GeoParquet — one file per .shp, each with its own schema, CRS (as PROJJSON), and geometry types (GeoParquet 1.1.0, WKB encoding, ZSTD). Files are read and written in chunks, so memory stays flat on large shapefiles. No reprojection. Target is S3 by default (one file per shapefile under an s3://<bucket>/<target>/ prefix); local_output writes to a directory on the EPT service filesystem instead.

A bare remote .shp URL is rejected: a shapefile needs its sidecar files (.shx, .dbf, .prj), so remote sources must be .zip archives.

Request

To S3, using your personal bucket (oidc-<username>, derived from your bearer token) and its default endpoint:

curl --fail-with-body \
  -X POST "${EPT_API_URL}/v1/conversions/shp/parquet" \
  -H "Authorization: Bearer ${ACCESS_TOKEN}" \
  -H "Content-Type: application/json" \
  -d '{
    "source_uri": "/data/mangrove_eov_poly_2025.zip",
    "target": "arco-data/mangrove"
  }'

To a specific bucket/project prefix, with request-scoped S3 credentials from POST /v1/auth:

{
  "source_uri": "https://files.example.org/data/mangrove_eov_poly_2025.zip",
  "bucket": "project-<projectid>",
  "target": "arco-data/mangrove",
  "endpoint_url": "https://minio.dive.edito.eu",
  "access_key_id": "<s3_credentials.access_key_id>",
  "secret_access_key": "<s3_credentials.secret_access_key>",
  "session_token": "<s3_credentials.session_token>"
}

To a directory on the EPT service filesystem (mutually exclusive with target/bucket and S3 credentials):

{
  "source_uri": "/data/mangrove_eov_poly_2025.zip",
  "local_output": "/data/out"
}

Convert only specific shapefiles (by stem; defaults to every .shp found):

{
  "source_uri": "/data/pkg.zip",
  "local_output": "/data/out",
  "layers": ["mangrove_eov_poly"],
  "overwrite": true
}

Response

{
  "source_uri": "/data/mangrove_eov_poly_2025.zip",
  "target_uri": "s3://oidc-<your-username>/arco-data/mangrove",
  "layers": ["mangrove_eov_poly"],
  "feature_counts": {"mangrove_eov_poly": 15432},
  "crs": {"mangrove_eov_poly": "EPSG:4326"},
  "outputs": ["s3://oidc-<your-username>/arco-data/mangrove/mangrove_eov_poly.parquet"],
  "status": "completed"
}

target_uri and outputs appear only in the response; the request takes target/bucket/local_output. When endpoint_url is provided, target_uri/outputs are full https:// URLs; s3:// URIs otherwise.

Notes

  • Request validation is extra="forbid" — unknown fields fail with 422.
  • target is a prefix relative to bucket and must not include s3:// or the bucket name; bucket must be a bare bucket name. bucket defaults to your personal oidc-<username> bucket.
  • Sources are local paths or HTTP(S) URLs, never s3://.
  • The target must not exist (local) or must not collide with existing objects (S3) unless overwrite is true.
  • Shapefile stems are sanitized into file names ([^A-Za-z0-9._-] → _); collisions fail the request.
  • Shapefiles without a usable .prj convert with crs: null; no reprojection is attempted.
  • S3 credentials are request-scoped; they never enter the process environment.