add vortex-geo crate and WKB extension type

Cargo.toml

@@ -69,6 +69,7 @@ members = [
     # Benchmarks website v3 (alpha) - leaf binary, not part of vortex-* API
     "benchmarks-website/server",
     "benchmarks-website/migrate",
+    "vortex-geo",
 ]
 exclude = ["java/testfiles", "wasm-test"]
 resolver = "2"
@@ -161,6 +162,9 @@ flatbuffers = "25.2.10"
 fsst-rs = "0.5.11"
 futures = { version = "0.3.31", default-features = false }
 fuzzy-matcher = "0.3"
+geo-traits = "0.3.0"
+geo-types = "0.7.19"
+geoarrow = "0.8.0"
 get_dir = "0.5.0"
 glob = "0.3.2"
 goldenfile = "1"
@@ -267,6 +271,7 @@ tracing-subscriber = "0.3"
 url = "2.5.7"
 uuid = { version = "1.23", features = ["js"] }
 wasm-bindgen-futures = "0.4.54"
+wkb = "0.9.2"
 xshell = "0.2.6"
 zigzag = "0.1.0"
 zip = "8.0.0"
@@ -290,6 +295,7 @@ 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 }

vortex-duckdb/Cargo.toml

@@ -40,16 +40,19 @@ tracing = { workspace = true }
 tracing-subscriber = { workspace = true }
 url = { workspace = true }
 vortex = { workspace = true, features = ["files", "tokio", "object_store"] }
+vortex-geo = { workspace = true }
 vortex-utils = { workspace = true, features = ["dashmap"] }
 
 [dev-dependencies]
 anyhow = { workspace = true }
+geo-types = { workspace = true }
 jiff = { workspace = true }
 rstest = { workspace = true }
 tempfile = { workspace = true }
 vortex-array = { workspace = true, features = ["_test-harness"] }
 vortex-runend = { workspace = true }
 vortex-sequence = { workspace = true }
+wkb = { workspace = true }
 
 [lints]
 workspace = true

vortex-duckdb/cpp/include/duckdb_vx/logical_type.h

@@ -12,6 +12,10 @@ extern "C" {
 char *duckdb_vx_logical_type_stringify(duckdb_logical_type ty);
 duckdb_logical_type duckdb_vx_logical_type_copy(duckdb_logical_type ty);
 
+/// Creates a GEOMETRY logical type with the given CRS (Coordinate Reference System).
+/// `crs` must be a NUL-terminated UTF-8 string. Pass an empty string for no CRS.
+duckdb_logical_type duckdb_vx_create_geometry(const char *crs);
+
 #ifdef __cplusplus /* End C ABI */
 }
 #endif

vortex-duckdb/cpp/include/duckdb_vx/value.h

@@ -3,13 +3,28 @@
 
 #pragma once
 
+#include "duckdb.h"
+
 #ifdef __cplusplus /* If compiled as C++, use C ABI */
 extern "C" {
 #endif
 
 // Create a null value with a reference to a logical type.
 duckdb_value duckdb_vx_value_create_null(duckdb_logical_type ty);
 
+/// Creates a GEOMETRY value containing the given WKB bytes and CRS.
+///
+/// `wkb` points to `len` bytes of well-known-binary geometry data; the bytes are not validated.
+/// `crs` must be a NUL-terminated UTF-8 string; pass NULL or an empty string for no CRS.
+duckdb_value duckdb_vx_value_create_geometry(const uint8_t *wkb, idx_t len, const char *crs);
+
+/// Extracts the raw WKB bytes from a GEOMETRY value as a duckdb_blob.
+///
+/// This bypasses the GEOMETRY -> BLOB default cast (which would require the spatial extension to
+/// be loaded). The returned `data` pointer must be freed with `duckdb_free`. Returns `{nullptr, 0}`
+/// if `value` is null or not a GEOMETRY value.
+duckdb_blob duckdb_vx_value_get_geometry(duckdb_value value);
+
 #ifdef __cplusplus /* End C ABI */
 }
 #endif

vortex-duckdb/cpp/logical_type.cpp

@@ -4,6 +4,7 @@
 #include "duckdb_vx/logical_type.h"
 #include "duckdb/common/types.hpp"
 #include <cassert>
+#include <string>
 
 duckdb_logical_type duckdb_vx_logical_type_copy(duckdb_logical_type ty) {
     D_ASSERT(ty);
@@ -19,3 +20,11 @@ char *duckdb_vx_logical_type_stringify(duckdb_logical_type c_type) {
     memcpy(result, str.c_str(), str.size() + 1);
     return result;
 }
+
+duckdb_logical_type duckdb_vx_create_geometry(const char *crs) {
+    D_ASSERT(crs);
+    auto geom =
+        (*crs == '\0') ? duckdb::LogicalType::GEOMETRY() : duckdb::LogicalType::GEOMETRY(std::string(crs));
+    auto copy = duckdb::make_uniq<duckdb::LogicalType>(std::move(geom));
+    return reinterpret_cast<duckdb_logical_type>(copy.release());
+}

vortex-duckdb/cpp/value.cpp

@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright the Vortex contributors
 
+#include "duckdb/common/types/geometry_crs.hpp"
 #include "duckdb/common/types/value.hpp"
 
 #include "duckdb_vx.h"
@@ -10,3 +11,30 @@ extern "C" duckdb_value duckdb_vx_value_create_null(duckdb_logical_type ty) {
     auto value = duckdb::make_uniq<duckdb::Value>(*logical_type);
     return reinterpret_cast<duckdb_value>(value.release());
 }
+
+extern "C" duckdb_value duckdb_vx_value_create_geometry(const uint8_t *wkb, idx_t len, const char *crs) {
+    const auto bytes = reinterpret_cast<duckdb::const_data_ptr_t>(wkb);
+    auto value =
+        (crs == nullptr || *crs == '\0')
+            ? duckdb::Value::GEOMETRY(bytes, len)
+            : duckdb::Value::GEOMETRY(bytes, len, duckdb::CoordinateReferenceSystem(std::string(crs)));
+    auto owned = duckdb::make_uniq<duckdb::Value>(std::move(value));
+    return reinterpret_cast<duckdb_value>(owned.release());
+}
+
+extern "C" duckdb_blob duckdb_vx_value_get_geometry(duckdb_value value) {
+    if (value == nullptr) {
+        return {nullptr, 0};
+    }
+    const auto val = reinterpret_cast<duckdb::Value *>(value);
+    if (val->type().id() != duckdb::LogicalTypeId::GEOMETRY) {
+        return {nullptr, 0};
+    }
+    const auto &str = duckdb::StringValue::Get(*val);
+    const auto size = str.size();
+    auto buf = reinterpret_cast<void *>(duckdb_malloc(size));
+    if (size > 0) {
+        memcpy(buf, str.c_str(), size);
+    }
+    return {buf, size};
+}