apache/iceberg · error · java.lang.UnsupportedOperationException

Iceberg does not support Spark geography edge algorithm:

Error message

Iceberg does not support Spark geography edge algorithm: 

What it means

When converting a Spark GeographyType to Iceberg, convertAlgorithm maps Spark's edge-interpolation algorithm to Iceberg's EdgeAlgorithm. Spark only supports SPHERICAL today, so any other algorithm value hits the default branch and throws this UnsupportedOperationException. The loud failure avoids silently defaulting a new Spark algorithm to the wrong semantics.

Source

Thrown at spark/v4.2/spark/src/main/java/org/apache/iceberg/spark/SparkTypeToType.java:200

      }
      return Types.GeographyType.of(geography.crs(), convertAlgorithm(geography.algorithm()));
    } else if (atomic instanceof NullType) {
      return Types.UnknownType.get();
    }

    throw new UnsupportedOperationException("Not a supported type: " + atomic.catalogString());
  }

  // Translates Spark's edge-interpolation algorithm to Iceberg's, mirroring
  // TypeToSparkType#convertAlgorithm. Spark supports only the spherical algorithm today; anything
  // else is rejected loudly rather than silently defaulting, so a new Spark algorithm surfaces here
  // instead of being dropped.
  private static EdgeAlgorithm convertAlgorithm(EdgeInterpolationAlgorithm algorithm) {
    switch (algorithm.toString().toUpperCase(Locale.ROOT)) {
      case "SPHERICAL":
        return EdgeAlgorithm.SPHERICAL;
      default:
        throw new UnsupportedOperationException(
            "Iceberg does not support Spark geography edge algorithm: " + algorithm);
    }
  }
}

View on GitHub (pinned to 86d9c8fc54)

Solutions

1. Upgrade iceberg-spark to a version supporting the Spark geography edge algorithm
2. Recreate the geography column with the SPHERICAL algorithm (the only supported one)
3. Convert the column to geometry (crs only) if edge semantics don't matter for your use
4. Check algorithm values with st_ functions/DESCRIBE before writing to Iceberg

Example fix

// before
// geography(geom, 'SOME_NEW_ALGORITHM') -> throws on conversion
// after
df.withColumn("geom", functions.expr("st_geogfromwkb(st_aswkb(geom), true /* spherical */)")); // SPHERICAL

if (!"SPHERICAL".equalsIgnoreCase(String.valueOf(sparkGeogType.algorithm()))) {
  throw new IllegalStateException("Only SPHERICAL geography is supported by Iceberg");
}

if (dt instanceof org.apache.spark.sql.types.GeographyType g
    && !"SPHERICAL".equalsIgnoreCase(String.valueOf(g.algorithm()))) {
  return false; // not convertible until recreated with SPHERICAL
}

try {
  icebergType = converter.convert(sparkSchema);
} catch (UnsupportedOperationException e) {
  if (e.getMessage().contains("edge algorithm")) {
    // recreate the column with SPHERICAL or convert to geometry
  }
}

When it happens

Trigger: Schema conversion of a geography column whose EdgeInterpolationAlgorithm is not SPHERICAL (e.g. a new/unknown algorithm introduced by a Spark upgrade or plugin).

Common situations: Running a newer Spark with a new geography algorithm against an older iceberg-spark; corrupted or exotic algorithm values from third-party spatial libraries; copy-pasted DDL with an algorithm string that isn't SPHERICAL.

Related errors

• Iceberg does not support Spark geography edge algorithm:
• Cannot convert Spark geography with mixed SRID to Iceberg
• Cannot process unknown snapshot operation: ${op.toLowerCase(
• Cannot convert Spark geography with mixed SRID to Iceberg
• Unexpected command:

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/5b74e0736952f9c3. Report an issue: GitHub.