feat(file-index): support additional types and validate decimal scale

include/paimon/file_index/file_index_reader.h

@@ -34,6 +34,8 @@ namespace paimon {
 /// `std::shared_ptr<FileIndexResult>` objects. It reads pre-built file-level index data
 /// (e.g., bitmap, bsi or bloom filters) from index file and evaluates
 /// whether a given data file may contain rows matching a specific predicate.
+/// @note Callers of `Visit*` for DECIMAL fields must ensure each literal's scale matches the scale
+/// of the indexed data; otherwise, index filtering results may be incorrect.
 class PAIMON_EXPORT FileIndexReader : public FunctionVisitor<std::shared_ptr<FileIndexResult>> {
  public:
     Result<std::shared_ptr<FileIndexResult>> VisitIsNotNull() override;

include/paimon/global_index/global_index_reader.h

@@ -35,6 +35,8 @@ namespace paimon {
 /// Derived classes are expected to implement the visitor methods (e.g., `VisitEqual`,
 /// `VisitIsNull`, etc.) to return index-based results that indicate which
 /// rows satisfy the given predicate.
+/// @note Callers of `Visit*` for DECIMAL fields must ensure each literal's scale matches the scale
+/// of the indexed data; otherwise, index filtering results may be incorrect.
 class PAIMON_EXPORT GlobalIndexReader : public FunctionVisitor<std::shared_ptr<GlobalIndexResult>> {
  public:
     /// VisitVectorSearch performs approximate vector similarity search.

src/paimon/CMakeLists.txt

@@ -572,6 +572,7 @@ if(PAIMON_BUILD_TESTS)
                     common/file_index/rangebitmap/dictionary/chunked_dictionary_test.cpp
                     common/file_index/rangebitmap/range_bitmap_file_index_test.cpp
                     common/file_index/rangebitmap/range_bitmap_io_test.cpp
+                    common/file_index/rangebitmap/range_bitmap_type_adapter_test.cpp
                     common/file_index/bloomfilter/bloom_filter_file_index_test.cpp
                     common/file_index/bloomfilter/fast_hash_test.cpp
                     common/global_index/complete_index_score_batch_reader_test.cpp

src/paimon/common/file_index/CMakeLists.txt

@@ -29,11 +29,13 @@ set(PAIMON_FILE_INDEX_SRC
     rangebitmap/dictionary/chunked_dictionary.cpp
     rangebitmap/dictionary/fixed_length_chunk.cpp
     rangebitmap/dictionary/key_factory.cpp
+    rangebitmap/dictionary/variable_length_chunk.cpp
     rangebitmap/utils/literal_serialization_utils.cpp
     rangebitmap/bit_slice_index_bitmap.cpp
     rangebitmap/range_bitmap.cpp
     rangebitmap/range_bitmap_file_index.cpp
-    rangebitmap/range_bitmap_file_index_factory.cpp)
+    rangebitmap/range_bitmap_file_index_factory.cpp
+    rangebitmap/range_bitmap_type_adapter.cpp)
 
 add_paimon_lib(paimon_file_index
                SOURCES

src/paimon/common/file_index/bitmap/bitmap_file_index_meta.cpp

@@ -18,11 +18,13 @@
 
 #include "paimon/common/file_index/bitmap/bitmap_file_index_meta.h"
 
+#include <cstdint>
 #include <string>
 #include <utility>
 
 #include "fmt/format.h"
 #include "paimon/common/utils/field_type_utils.h"
+#include "paimon/common/utils/math.h"
 #include "paimon/defs.h"
 #include "paimon/io/data_input_stream.h"
 #include "paimon/memory/bytes.h"
@@ -80,6 +82,18 @@ Result<std::function<void(const Literal&)>> BitmapFileIndexMeta::GetValueWriter(
                 [output_stream](const Literal& literal) -> void {
                     output_stream->WriteValue<int64_t>(literal.GetValue<int64_t>());
                 });
+        case FieldType::FLOAT:
+            return std::function<void(const Literal&)>(
+                [output_stream](const Literal& literal) -> void {
+                    const float value = CanonicalizeFloatingPoint(literal.GetValue<float>());
+                    output_stream->WriteValue<float>(value);
+                });
+        case FieldType::DOUBLE:
+            return std::function<void(const Literal&)>(
+                [output_stream](const Literal& literal) -> void {
+                    const double value = CanonicalizeFloatingPoint(literal.GetValue<double>());
+                    output_stream->WriteValue<double>(value);
+                });
         case FieldType::STRING:
             return std::function<void(const Literal&)>(
                 [output_stream](const Literal& literal) -> void {
@@ -155,6 +169,24 @@ Result<std::function<Result<Literal>()>> BitmapFileIndexMeta::GetValueReader(
             };
             return func;
         }
+        case FieldType::FLOAT: {
+            std::function<Result<Literal>()> func = [&in, move_body_start,
+                                                     this]() -> Result<Literal> {
+                PAIMON_ASSIGN_OR_RAISE(float value,
+                                       ReadAndMoveBodyStart<float>(in, move_body_start));
+                return Literal(value);
+            };
+            return func;
+        }
+        case FieldType::DOUBLE: {
+            std::function<Result<Literal>()> func = [&in, move_body_start,
+                                                     this]() -> Result<Literal> {
+                PAIMON_ASSIGN_OR_RAISE(double value,
+                                       ReadAndMoveBodyStart<double>(in, move_body_start));
+                return Literal(value);
+            };
+            return func;
+        }
         case FieldType::DATE: {
             std::function<Result<Literal>()> func = [&in, move_body_start,
                                                      this]() -> Result<Literal> {