feat: add StructLikeSet (#598)

Adds StructLikeSet<bool kValidate = true>, a hash set for StructLike
rows backed by an internal arena allocator.

Key design points:
- Deep-copies inserted rows into a monotonic_buffer_resource arena;
string data and nested struct/list/map scalars are fully materialized so
the set owns its memory independently of the caller
- Transparent heterogeneous lookup: Contains() does not allocate a
temporary key
- Hash and equality semantics match the Java reference implementation
(String.hashCode, StructLikeHash, ListHash; float/double use canonical
NaN bits and distinguish ±0.0)
- Schema validation (field count + scalar type) on Insert/Contains; can
be disabled via kValidate=false (UncheckedStructLikeSet) when the caller
guarantees conformance
- Internal Arena wrapper containers use std::pmr::vector

Author: wgtmac

src/iceberg/CMakeLists.txt

@@ -112,6 +112,7 @@ set(ICEBERG_SOURCES
     util/property_util.cc
     util/snapshot_util.cc
     util/string_util.cc
+    util/struct_like_set.cc
     util/temporal_util.cc
     util/timepoint.cc
     util/transform_util.cc

src/iceberg/iceberg_export.h

@@ -27,8 +27,31 @@
 #  else
 #    define ICEBERG_EXPORT __declspec(dllimport)
 #  endif
-#else  // Not Windows
+
+#  define ICEBERG_TEMPLATE_EXPORT ICEBERG_EXPORT
+
+// For template class declarations. Empty on MSVC: dllexport on a class template
+// declaration combined with extern template triggers C4910.
+#  if defined(_MSC_VER)
+#    define ICEBERG_TEMPLATE_CLASS_EXPORT
+#  else
+#    define ICEBERG_TEMPLATE_CLASS_EXPORT ICEBERG_EXPORT
+#  endif
+
+// For extern template declarations. Empty when building the DLL on MSVC:
+// `extern` + `dllexport` is contradictory and triggers C4910.
+#  if defined(_MSC_VER) && defined(ICEBERG_EXPORTING) && !defined(ICEBERG_STATIC)
+#    define ICEBERG_EXTERN_TEMPLATE_CLASS_EXPORT
+#  else
+#    define ICEBERG_EXTERN_TEMPLATE_CLASS_EXPORT ICEBERG_TEMPLATE_EXPORT
+#  endif
+
+#else  // Non-Windows
 #  ifndef ICEBERG_EXPORT
 #    define ICEBERG_EXPORT __attribute__((visibility("default")))
 #  endif
+
+#  define ICEBERG_TEMPLATE_EXPORT
+#  define ICEBERG_TEMPLATE_CLASS_EXPORT ICEBERG_EXPORT
+#  define ICEBERG_EXTERN_TEMPLATE_CLASS_EXPORT ICEBERG_TEMPLATE_EXPORT
 #endif

src/iceberg/meson.build

@@ -130,6 +130,7 @@ iceberg_sources = files(
     'util/property_util.cc',
     'util/snapshot_util.cc',
     'util/string_util.cc',
+    'util/struct_like_set.cc',
     'util/temporal_util.cc',
     'util/timepoint.cc',
     'util/transform_util.cc',

src/iceberg/test/CMakeLists.txt

@@ -118,6 +118,7 @@ add_iceberg_test(util_test
                  roaring_position_bitmap_test.cc
                  position_delete_index_test.cc
                  string_util_test.cc
+                 struct_like_set_test.cc
                  transform_util_test.cc
                  truncate_util_test.cc
                  url_encoder_test.cc

src/iceberg/test/meson.build

@@ -93,6 +93,7 @@ iceberg_tests = {
             'position_delete_index_test.cc',
             'roaring_position_bitmap_test.cc',
             'string_util_test.cc',
+            'struct_like_set_test.cc',
             'transform_util_test.cc',
             'truncate_util_test.cc',
             'url_encoder_test.cc',