Skip to main content

GeoJSON Reader

The GeoJSON reader provides support for reading GeoJSON and GeoJSONSeq (newline-delimited GeoJSON) formats.

Format Name​

geojson_ogr

Overview​

This is a named OGR Reader that intelligently switches between GeoJSON and GeoJSONSeq drivers based on the multi option.

Supported Formats​

  • GeoJSON (.geojson, .json) - Standard GeoJSON FeatureCollection
  • GeoJSONSeq (.geojsonl, .geojsons) - Newline-delimited GeoJSON (default)

Basic Usage​

Python​

# Read standard GeoJSON (sample-data Volumes path)
df = spark.read.format("geojson_ogr") \
    .option("multi", "false") \
    .load("/Volumes/main/default/geobrix_samples/geobrix-examples/nyc/boroughs/nyc_boroughs.geojson")
df.show()
Example output
+--------------------+-----------+---------+
|geom_0              |geom_0_srid|BoroName |
+--------------------+-----------+---------+
|[BINARY]            |4326       |Manhattan|
|...                 |...        |...      |
+--------------------+-----------+---------+

Scala​

val df = spark.read.format("geojson_ogr")
      |  .option("multi", "false")
      |  .load("/Volumes/main/default/geobrix_samples/geobrix-examples/nyc/boroughs/nyc_boroughs.geojson")
Example output
+--------------------+-----------+---------+
|geom_0              |geom_0_srid|BoroName |
+--------------------+-----------+---------+
|[BINARY]            |4326       |Manhattan|
|...                 |...        |...      |
+--------------------+-----------+---------+

SQL​

-- Read GeoJSON in SQL (sample-data Volumes path)
SELECT * FROM geojson_ogr.`/Volumes/main/default/geobrix_samples/geobrix-examples/nyc/boroughs/nyc_boroughs.geojson`;
Example output
+--------------------+-----------+---------+
|geom_0              |geom_0_srid|BoroName |
+--------------------+-----------+---------+
|[BINARY]            |4326       |...      |
|...                 |...        |...      |
+--------------------+-----------+---------+

Options​

multi​

Default: "true"

Controls which GeoJSON driver to use:

  • "true" → Uses GeoJSONSeq driver (newline-delimited, better for large files)
  • "false" → Uses GeoJSON driver (standard FeatureCollection format)
# Read GeoJSONSeq (newline-delimited, sample-data path)
df = spark.read.format("geojson_ogr").load("/Volumes/main/default/geobrix_samples/geobrix-examples/nyc/boroughs/nyc_boroughs.geojsonl")
# Or explicitly: .option("multi", "true")
df.show()
Example output
+--------------------+-----------+---------+
|geom_0              |geom_0_srid|BoroName |
+--------------------+-----------+---------+
|[BINARY]            |4326       |...      |
|...                 |...        |...      |
+--------------------+-----------+---------+

Other Options​

All OGR reader options are available, e.g.:

  • chunkSize - Records per chunk (default: "10000")
  • asWKB - Output as WKB vs WKT (default: "true")

Output Schema​

The output maintains attribute columns and adds three columns for geometry:

root
 |-- geom_0: binary (geometry in WKB format)
 |-- geom_0_srid: integer (spatial reference ID)
 |-- geom_0_srid_proj: string (projection definition)
 |-- <properties>: various types (GeoJSON properties)

GeoJSON vs GeoJSONSeq​

Standard GeoJSON (FeatureCollection)​

Format:

{
  "type": "FeatureCollection",
  "features": [
    {
      "type": "Feature",
      "geometry": { "type": "Point", "coordinates": [0, 0] },
      "properties": { "name": "Feature 1" }
    }
  ]
}

When to use: Small to medium files, API responses, standard GeoJSON files

Read with: option("multi", "false")

GeoJSONSeq (Newline-Delimited)​

Format:

{"type":"Feature","geometry":{"type":"Point","coordinates":[0,0]},"properties":{"name":"Feature 1"}}
{"type":"Feature","geometry":{"type":"Point","coordinates":[1,1]},"properties":{"name":"Feature 2"}}

When to use: Large files, streaming data, parallel processing, better Spark performance

Read with: option("multi", "true") (default) or omit the option

Next Steps​