Merge pull request #90 from samansmink/bump-delta-to-0.3.1

bump kernel to v0.3.1

.github/workflows/LocalTesting.yml

@@ -113,7 +113,6 @@ jobs:
         uses: hendrikmuhs/ccache-action@main
         with:
           key: ${{ github.job }}
-          save: ${{ github.ref == 'refs/heads/main' || github.repository != 'duckdb/duckdb' }}
 
       - name: Setup vcpkg
         uses: lukka/[email protected]
@@ -281,13 +280,13 @@ jobs:
         if: always()
         shell: bash
         run: |
-          python3 ./scripts/regression_test_runner.py --old=duckdb_delta/build/release/benchmark/benchmark_runner --new=build/release/benchmark/benchmark_runner --benchmarks=.github/regression/tpch_sf1_local.csv --verbose --threads=2 --root-dir=.
+          python3 ./duckdb/scripts/regression_test_runner.py --old=duckdb_delta/build/release/benchmark/benchmark_runner --new=build/release/benchmark/benchmark_runner --benchmarks=.github/regression/tpch_sf1_local.csv --verbose --threads=2 --root-dir=.
 
       - name: Regression Test TPC-DS
         if: always()
         shell: bash
         run: |
-          python ./scripts/regression_test_runner.py --old=duckdb_delta/build/release/benchmark/benchmark_runner --new=build/release/benchmark/benchmark_runner --benchmarks=.github/regression/tpcds_sf1_local.csv --verbose --threads=2 --root-dir=.
+          python ./duckdb/scripts/regression_test_runner.py --old=duckdb_delta/build/release/benchmark/benchmark_runner --new=build/release/benchmark/benchmark_runner --benchmarks=.github/regression/tpcds_sf1_local.csv --verbose --threads=2 --root-dir=.
 
       - name: Test benchmark makefile
         shell: bash

CMakeLists.txt

@@ -99,7 +99,7 @@ ExternalProject_Add(
         GIT_REPOSITORY "https://github.com/delta-incubator/delta-kernel-rs"
         # WARNING: the FFI headers are currently pinned due to the C linkage issue of the c++ headers. Currently, when bumping
         # the kernel version, the produced header in ./src/include/delta_kernel_ffi.hpp should be also bumped, applying the fix
-        GIT_TAG v0.2.0
+        GIT_TAG v0.3.1
         # Prints the env variables passed to the cargo build to the terminal, useful in debugging because passing them
         # through CMake is an error-prone mess
         CONFIGURE_COMMAND ${CMAKE_COMMAND} -E env ${RUST_UNSET_ENV_VARS} ${RUST_ENV_VARS} env

src/include/delta_kernel_ffi.hpp

@@ -50,6 +50,7 @@ enum class KernelError {
   InvalidDecimalError,
   InvalidStructDataError,
   InternalError,
+  InvalidExpression,
 };
 
 struct CStringMap;
@@ -93,6 +94,13 @@ struct KernelBoolSlice {
   uintptr_t len;
 };
 
+/// An owned slice of u64 row indexes allocated by the kernel. The engine is responsible for
+/// freeing this slice by calling [super::free_row_indexes] once.
+struct KernelRowIndexArray {
+  uint64_t *ptr;
+  uintptr_t len;
+};
+
 /// Represents an object that crosses the FFI boundary and which outlives the scope that created
 /// it. It can be passed freely between rust code and external code. The
 ///
@@ -382,7 +390,7 @@ using CScanCallback = void(*)(NullableCvoid engine_context,
                               const DvInfo *dv_info,
                               const CStringMap *partition_map);
 
-// This trickery is from https://github.com/mozilla/cbindgen/issues/402#issuecomment-578680163
+    // This trickery is from https://github.com/mozilla/cbindgen/issues/402#issuecomment-578680163
 struct im_an_unused_struct_that_tricks_msvc_into_compilation {
     ExternResult<KernelBoolSlice> field;
     ExternResult<bool> field2;
@@ -394,6 +402,7 @@ struct im_an_unused_struct_that_tricks_msvc_into_compilation {
     ExternResult<Handle<SharedScanDataIterator>> field8;
     ExternResult<Handle<SharedScan>> field9;
     ExternResult<Handle<ExclusiveFileReadResultIterator>> field10;
+    ExternResult<KernelRowIndexArray> field11;
 };
 
 extern "C" {
@@ -403,6 +412,11 @@ extern "C" {
 /// Caller is responsible for passing a valid handle.
 void free_bool_slice(KernelBoolSlice slice);
 
+/// # Safety
+///
+/// Caller is responsible for passing a valid handle.
+void free_row_indexes(KernelRowIndexArray slice);
+
 /// Drop an `ExclusiveEngineData`.
 ///
 /// # Safety
@@ -705,6 +719,14 @@ ExternResult<KernelBoolSlice> selection_vector_from_dv(const DvInfo *dv_info,
                                                        Handle<SharedExternEngine> engine,
                                                        Handle<SharedGlobalScanState> state);
 
+/// Get a vector of row indexes out of a [`DvInfo`] struct
+///
+/// # Safety
+/// Engine is responsible for providing valid pointers for each argument
+ExternResult<KernelRowIndexArray> row_indexes_from_dv(const DvInfo *dv_info,
+                                                      Handle<SharedExternEngine> engine,
+                                                      Handle<SharedGlobalScanState> state);
+
 /// Shim for ffi to call visit_scan_data. This will generally be called when iterating through scan
 /// data which provides the data handle and selection vector as each element in the iterator.
 ///
@@ -715,6 +737,6 @@ void visit_scan_data(Handle<ExclusiveEngineData> data,
                      NullableCvoid engine_context,
                      CScanCallback callback);
 
-} // extern "C"
+}  // extern "C"
 
-} // namespace ffi
+}  // namespace ffi