refactor: rename vortex-geo to vortex-spatial

.github/workflows/codspeed.yml

@@ -57,7 +57,7 @@ jobs:
           - { shard: 6, name: "Encodings 3", packages: "vortex-pco vortex-runend vortex-sequence" }
           - { shard: 7, name: "Encodings 4", packages: "vortex-sparse vortex-zigzag vortex-zstd" }
           - { shard: 8, name: "Storage formats & row encoding", packages: "vortex-flatbuffers vortex-proto vortex-btrblocks vortex-row" }
-          - { shard: 9, name: "Tensor & geo", packages: "vortex-tensor vortex-geo" }
+          - { shard: 9, name: "Tensor & spatial", packages: "vortex-tensor vortex-spatial" }
     name: "Benchmark with Codspeed (Shard #${{ matrix.shard }})"
     timeout-minutes: 30
     runs-on: >-

Cargo.toml

@@ -69,7 +69,7 @@ members = [
     "benchmarks/duckdb-bench",
     "benchmarks/random-access-bench",
     "benchmarks/string-bench",
-    "vortex-geo",
+    "vortex-spatial",
 ]
 exclude = ["java/testfiles", "wasm-test"]
 resolver = "2"
@@ -310,7 +310,6 @@ vortex-fastlanes = { version = "0.1.0", path = "./encodings/fastlanes", default-
 vortex-file = { version = "0.1.0", path = "./vortex-file", default-features = false }
 vortex-flatbuffers = { version = "0.1.0", path = "./vortex-flatbuffers", default-features = false }
 vortex-fsst = { version = "0.1.0", path = "./encodings/fsst", default-features = false }
-vortex-geo = { version = "0.1.0", path = "./vortex-geo", default-features = false }
 vortex-io = { version = "0.1.0", path = "./vortex-io", default-features = false }
 vortex-ipc = { version = "0.1.0", path = "./vortex-ipc", default-features = false }
 vortex-json = { version = "0.1.0", path = "./vortex-json", default-features = false }
@@ -327,6 +326,7 @@ vortex-scan = { version = "0.1.0", path = "./vortex-scan", default-features = fa
 vortex-sequence = { version = "0.1.0", path = "encodings/sequence", default-features = false }
 vortex-session = { version = "0.1.0", path = "./vortex-session", default-features = false }
 vortex-sparse = { version = "0.1.0", path = "./encodings/sparse", default-features = false }
+vortex-spatial = { version = "0.1.0", path = "./vortex-spatial", default-features = false }
 vortex-tensor = { version = "0.1.0", path = "./vortex-tensor", default-features = false }
 vortex-utils = { version = "0.1.0", path = "./vortex-utils", default-features = false }
 vortex-zigzag = { version = "0.1.0", path = "./encodings/zigzag", default-features = false }

bench-orchestrator/bench_orchestrator/ci_matrix/render.py

@@ -12,7 +12,7 @@
     Format.PARQUET,
     Format.VORTEX,
     Format.VORTEX_COMPACT,
-    Format.VORTEX_NATIVE,
+    Format.VORTEX_SPATIAL_NATIVE,
     Format.DUCKDB,
     Format.LANCE,
 )

bench-orchestrator/bench_orchestrator/config.py

@@ -43,7 +43,7 @@ class Format(Enum):
     PARQUET = "parquet"
     VORTEX = "vortex"
     VORTEX_COMPACT = "vortex-compact"
-    VORTEX_NATIVE = "vortex-geo-native"
+    VORTEX_SPATIAL_NATIVE = "vortex-spatial-native"
     DUCKDB = "duckdb"
     LANCE = "lance"
 
@@ -77,7 +77,7 @@ class Benchmark(Enum):
         Format.PARQUET,
         Format.VORTEX,
         Format.VORTEX_COMPACT,
-        Format.VORTEX_NATIVE,
+        Format.VORTEX_SPATIAL_NATIVE,
         Format.DUCKDB,
     ],
     Engine.LANCE: [Format.LANCE],

bench-orchestrator/tests/test_config.py

@@ -26,21 +26,21 @@ def test_parse_formats_json_accepts_ci_format_arrays() -> None:
     assert formats == [Format.PARQUET, Format.VORTEX, Format.DUCKDB]
 
 
-def test_parse_formats_json_accepts_vortex_native() -> None:
-    formats = parse_formats_json('["parquet","vortex","vortex-geo-native"]')
+def test_parse_formats_json_accepts_vortex_spatial_native() -> None:
+    formats = parse_formats_json('["parquet","vortex","vortex-spatial-native"]')
 
-    assert formats == [Format.PARQUET, Format.VORTEX, Format.VORTEX_NATIVE]
+    assert formats == [Format.PARQUET, Format.VORTEX, Format.VORTEX_SPATIAL_NATIVE]
 
 
-def test_resolve_axis_targets_offers_vortex_native_on_duckdb_only() -> None:
-    # vortex-geo-native is a DuckDB-only lane; the DataFusion axis is dropped as unsupported.
+def test_resolve_axis_targets_offers_vortex_spatial_native_on_duckdb_only() -> None:
+    # vortex-spatial-native is a DuckDB-only lane; the DataFusion axis is dropped as unsupported.
     targets, warnings = resolve_axis_targets(
         [Engine.DATAFUSION, Engine.DUCKDB],
-        [Format.VORTEX_NATIVE],
+        [Format.VORTEX_SPATIAL_NATIVE],
     )
 
-    assert targets == [BenchmarkTarget(engine=Engine.DUCKDB, format=Format.VORTEX_NATIVE)]
-    assert warnings == ["Format vortex-geo-native is not supported by engine datafusion"]
+    assert targets == [BenchmarkTarget(engine=Engine.DUCKDB, format=Format.VORTEX_SPATIAL_NATIVE)]
+    assert warnings == ["Format vortex-spatial-native is not supported by engine datafusion"]
 
 
 def test_resolve_axis_targets_filters_unsupported_combinations() -> None:
@@ -72,19 +72,19 @@ def test_resolve_axis_targets_skips_engines_a_benchmark_cannot_run() -> None:
     assert warnings == ["Benchmark spatialbench does not support engine datafusion"]
 
 
-def test_resolve_axis_targets_expands_spatialbench_three_lanes() -> None:
+def test_resolve_axis_targets_supports_spatialbench_three_lanes() -> None:
     # The single-command three-lane comparison: parquet, WKB vortex, and native-geometry vortex, all
     # on DuckDB.
     targets, warnings = resolve_axis_targets(
         [Engine.DUCKDB],
-        [Format.PARQUET, Format.VORTEX, Format.VORTEX_NATIVE],
+        [Format.PARQUET, Format.VORTEX, Format.VORTEX_SPATIAL_NATIVE],
         Benchmark.SPATIALBENCH,
     )
 
     assert targets == [
         BenchmarkTarget(engine=Engine.DUCKDB, format=Format.PARQUET),
         BenchmarkTarget(engine=Engine.DUCKDB, format=Format.VORTEX),
-        BenchmarkTarget(engine=Engine.DUCKDB, format=Format.VORTEX_NATIVE),
+        BenchmarkTarget(engine=Engine.DUCKDB, format=Format.VORTEX_SPATIAL_NATIVE),
     ]
     assert warnings == []
 

bench-orchestrator/tests/test_executor.py

@@ -33,17 +33,17 @@ def test_build_command_adds_duckdb_cleanup_flag() -> None:
     assert "scale-factor=1.0" in cmd
 
 
-def test_build_command_serializes_vortex_native_format() -> None:
+def test_build_command_serializes_vortex_spatial_native_format() -> None:
     executor = BenchmarkExecutor(Path("/tmp/duckdb-bench"), Engine.DUCKDB)
 
     cmd = executor.build_command(
         benchmark=Benchmark.SPATIALBENCH,
-        formats=[Format.PARQUET, Format.VORTEX, Format.VORTEX_NATIVE],
+        formats=[Format.PARQUET, Format.VORTEX, Format.VORTEX_SPATIAL_NATIVE],
         iterations=1,
         options={"scale-factor": "1.0"},
     )
 
-    assert "parquet,vortex,vortex-geo-native" in cmd
+    assert "parquet,vortex,vortex-spatial-native" in cmd
 
 
 def test_build_command_omits_formats_for_lance_backend() -> None:

benchmarks/datafusion-bench/src/lib.rs

@@ -109,7 +109,7 @@ pub fn format_to_df_format(format: Format) -> Arc<dyn FileFormat> {
     match format {
         Format::Csv => Arc::new(CsvFormat::default()) as _,
         Format::Parquet => Arc::new(ParquetFormat::new()),
-        Format::OnDiskVortex | Format::VortexCompact | Format::VortexNative => Arc::new(
+        Format::OnDiskVortex | Format::VortexCompact | Format::VortexSpatialNative => Arc::new(
             VortexFormat::new_with_options(SESSION.clone(), vortex_table_options()),
         ),
         Format::OnDiskDuckDB | Format::Lance => {

benchmarks/duckdb-bench/src/lib.rs

@@ -183,7 +183,7 @@ impl DuckClient {
             Format::Parquet
             | Format::OnDiskVortex
             | Format::VortexCompact
-            | Format::VortexNative => "VIEW",
+            | Format::VortexSpatialNative => "VIEW",
             Format::OnDiskDuckDB => "TABLE",
             Format::Lance => {
                 anyhow::bail!(

docs/concepts/expressions.md

@@ -12,7 +12,7 @@ such as whether the function is strict, and the actual logic for executing the f
 The built-in scalar functions can be found in the `vortex-array::expr` module, with additional use-case specific
 functions provided by integration and plugin crates.
 
-In the future, we plan to add full support for geospatial functions, date-time functions, and more.
+In the future, we plan to add full support for spatial functions, date-time functions, and more.
 
 ## Scalar Function Arrays
 

java/vortex-jni/src/test/java/dev/vortex/api/GeoTypesTest.java → java/vortex-jni/src/test/java/dev/vortex/api/SpatialTypesTest.java

@@ -33,11 +33,11 @@
 import org.junit.jupiter.api.io.TempDir;
 
 /**
- * Round-trips a Vortex geo extension column ({@code vortex.geo.wkb}) through the JNI boundary. Geo columns cross the
- * boundary as Arrow fields tagged with the GeoArrow extension name ({@code geoarrow.wkb}) and JSON metadata carrying
- * the CRS.
+ * Round-trips a Vortex spatial extension column ({@code vortex.st.wkb}) through the JNI boundary. Spatial columns cross
+ * the boundary as Arrow fields tagged with the GeoArrow extension name ({@code geoarrow.wkb}) and JSON metadata
+ * carrying the CRS.
  */
-public final class GeoTypesTest {
+public final class SpatialTypesTest {
     private static final String EXTENSION_NAME_KEY = "ARROW:extension:name";
     private static final String EXTENSION_METADATA_KEY = "ARROW:extension:metadata";
     private static final String GEOARROW_WKB = "geoarrow.wkb";
@@ -57,7 +57,7 @@ public static void loadLibrary() {
 
     @BeforeAll
     static void setup() throws IOException {
-        writePath = tempDir.resolve("geo.vortex").toAbsolutePath().toUri().toString();
+        writePath = tempDir.resolve("spatial.vortex").toAbsolutePath().toUri().toString();
 
         WKB_POINTS.add(wkbPoint(1.0, 2.0));
         WKB_POINTS.add(wkbPoint(-111.7610, 34.8697));

scripts/compare-benchmark-jsons.py

@@ -551,7 +551,7 @@ def build_statistical_analysis(df: pd.DataFrame, threshold_pct: int) -> dict[str
     }
 
 
-def calculate_geo_mean(df: pd.DataFrame) -> float:
+def calculate_geometric_mean(df: pd.DataFrame) -> float:
     """Geometric mean of positive ratios from a DataFrame ratio column."""
 
     valid_ratios = [r for r in df["ratio"] if r > 0 and not pd.isna(r)]
@@ -1019,8 +1019,8 @@ def main() -> None:
     vortex_df = headline_df[headline_df["file_format"].str.startswith("vortex")]
     parquet_df = headline_df[headline_df["file_format"].eq(CONTROL_FORMAT)]
 
-    vortex_geo_mean_ratio = calculate_geo_mean(vortex_df)
-    parquet_geo_mean_ratio = calculate_geo_mean(parquet_df)
+    vortex_geometric_mean_ratio = calculate_geometric_mean(vortex_df)
+    parquet_geometric_mean_ratio = calculate_geometric_mean(parquet_df)
 
     statistical_analysis = build_statistical_analysis(query_df, threshold_pct)
     verdict = build_verdict(statistical_analysis) if statistical_analysis is not None else None
@@ -1037,15 +1037,15 @@ def main() -> None:
 
     if len(vortex_df) > 0:
         vortex_performance = format_performance(
-            vortex_geo_mean_ratio,
+            vortex_geometric_mean_ratio,
             improvement_threshold,
             regression_threshold,
             "vortex",
         )
         summary_fields.append(f"**Vortex (geomean)**: {vortex_performance}")
     if len(parquet_df) > 0:
         parquet_performance = format_performance(
-            parquet_geo_mean_ratio,
+            parquet_geometric_mean_ratio,
             improvement_threshold,
             regression_threshold,
             "parquet",
@@ -1079,7 +1079,7 @@ def main() -> None:
     for engine, file_format, unit in sorted(grouped_tables.groups.keys(), key=group_sort_key):
         group_df = grouped_tables.get_group((engine, file_format, unit)).sort_values("name")
         group_performance = format_performance(
-            calculate_geo_mean(group_df),
+            calculate_geometric_mean(group_df),
             improvement_threshold,
             regression_threshold,
             "group",

vortex-array/src/expr/exprs.rs

@@ -610,7 +610,7 @@ where
 /// whose result is null exactly when any operand is null.
 ///
 /// This is the `ScalarFnVTable::validity` for kernels that propagate nulls and never produce a
-/// null from non-null inputs (comparisons, arithmetic, most geo and tensor ops). Returning it lets
+/// null from non-null inputs (comparisons, arithmetic, most spatial and tensor operations). Returning it lets
 /// the planner derive the output's null mask without executing the kernel. Yields `None` when the
 /// expression has no children.
 pub fn union_child_validities(expression: &Expression) -> VortexResult<Option<Expression>> {

vortex-bench/Cargo.toml

@@ -28,7 +28,7 @@ vortex = { workspace = true, features = [
     "zstd",
 ] }
 vortex-arrow = { workspace = true }
-vortex-geo = { workspace = true }
+vortex-spatial = { workspace = true }
 vortex-tensor = { workspace = true } # TODO(connor): In the future, this might be inside vortex.
 
 anyhow = { workspace = true }

vortex-bench/sql/spatialbench.md

@@ -1,6 +1,6 @@
 # SpatialBench benchmark
 
-The [Apache Sedona SpatialBench](https://sedona.apache.org/spatialbench/) geospatial
+The [Apache Sedona SpatialBench](https://sedona.apache.org/spatialbench/) spatial
 analytics benchmark: twelve queries (Q1 ... Q12 in [`spatialbench.sql`](./spatialbench.sql),
 DuckDB dialect) over a trips/zones schema, exercising spatial predicates and functions such
 as `ST_DWithin`, `ST_Intersects`, and `ST_Distance`. The query logic matches upstream
@@ -13,3 +13,16 @@ The harness lives in [`src/spatialbench`](../src/spatialbench).
 ```bash
 vx-bench run spatialbench
 ```
+
+The default command compares the Parquet and Vortex WKB representations with DuckDB. To run the
+native Vortex spatial representation explicitly:
+
+```bash
+vx-bench run spatialbench --engine duckdb --format vortex-spatial-native
+```
+
+To compare all three representations in one run:
+
+```bash
+vx-bench run spatialbench --engine duckdb --format parquet,vortex,vortex-spatial-native
+```