GH-33596: [C++][Parquet] Parquet page index read support

cpp/src/parquet/CMakeLists.txt

@@ -130,9 +130,12 @@ set(PARQUET_STATIC_TEST_LINK_LIBS ${PARQUET_MIN_TEST_LIBS} parquet_static thrift
 
 #
 # Generated Thrift sources
-set_source_files_properties(src/generated/parquet_types.cpp src/generated/parquet_types.h
-                            src/generated/parquet_constants.cpp
-                            src/generated/parquet_constants.h
+set(PARQUET_THRIFT_SOURCE_DIR "${ARROW_SOURCE_DIR}/src/generated/")
+
+set_source_files_properties("${PARQUET_THRIFT_SOURCE_DIR}/parquet_types.cpp"
+                            "${PARQUET_THRIFT_SOURCE_DIR}/parquet_types.h"
+                            "${PARQUET_THRIFT_SOURCE_DIR}/parquet_constants.cpp"
+                            "${PARQUET_THRIFT_SOURCE_DIR}/parquet_constants.h"
                             PROPERTIES SKIP_PRECOMPILE_HEADERS ON
                                        SKIP_UNITY_BUILD_INCLUSION ON)
 
@@ -167,8 +170,8 @@ set(PARQUET_SRCS
     metadata.cc
     xxhasher.cc
     page_index.cc
-    "${ARROW_SOURCE_DIR}/src/generated/parquet_constants.cpp"
-    "${ARROW_SOURCE_DIR}/src/generated/parquet_types.cpp"
+    "${PARQUET_THRIFT_SOURCE_DIR}/parquet_constants.cpp"
+    "${PARQUET_THRIFT_SOURCE_DIR}/parquet_types.cpp"
     platform.cc
     printer.cc
     properties.cc
@@ -277,8 +280,8 @@ add_arrow_lib(parquet
 
 if(WIN32 AND NOT (ARROW_TEST_LINKAGE STREQUAL "static"))
   add_library(parquet_test_support STATIC
-              "${ARROW_SOURCE_DIR}/src/generated/parquet_constants.cpp"
-              "${ARROW_SOURCE_DIR}/src/generated/parquet_types.cpp")
+              "${PARQUET_THRIFT_SOURCE_DIR}/parquet_constants.cpp"
+              "${PARQUET_THRIFT_SOURCE_DIR}/parquet_types.cpp")
   target_link_libraries(parquet_test_support thrift::thrift)
   set(PARQUET_SHARED_TEST_LINK_LIBS ${PARQUET_SHARED_TEST_LINK_LIBS} parquet_test_support)
   set(PARQUET_LIBRARIES ${PARQUET_LIBRARIES} parquet_test_support)

cpp/src/parquet/file_reader.cc

@@ -40,6 +40,7 @@
 #include "parquet/exception.h"
 #include "parquet/file_writer.h"
 #include "parquet/metadata.h"
+#include "parquet/page_index.h"
 #include "parquet/platform.h"
 #include "parquet/properties.h"
 #include "parquet/schema.h"
@@ -302,6 +303,22 @@ class SerializedFile : public ParquetFileReader::Contents {
 
   std::shared_ptr<FileMetaData> metadata() const override { return file_metadata_; }
 
+  std::shared_ptr<PageIndexReader> GetPageIndexReader() override {
+    if (!file_metadata_) {
+      // Usually this won't happen if user calls one of the static Open() functions
+      // to create a ParquetFileReader instance. But if user calls the constructor
+      // directly and calls GetPageIndexReader() before Open() then this could happen.
+      throw ParquetException(
+          "Cannot call GetPageIndexReader() due to missing file metadata. Did you "
+          "forget to call ParquetFileReader::Open() first?");
+    }
+    if (!page_index_reader_) {
+      page_index_reader_ = PageIndexReader::Make(source_.get(), file_metadata_,
+                                                 properties_, file_decryptor_);
+    }
+    return page_index_reader_;
+  }
+
   void set_metadata(std::shared_ptr<FileMetaData> metadata) {
     file_metadata_ = std::move(metadata);
   }
@@ -522,7 +539,7 @@ class SerializedFile : public ParquetFileReader::Contents {
   int64_t source_size_;
   std::shared_ptr<FileMetaData> file_metadata_;
   ReaderProperties properties_;
-
+  std::shared_ptr<PageIndexReader> page_index_reader_;
   std::shared_ptr<InternalFileDecryptor> file_decryptor_;
 
   // \return The true length of the metadata in bytes
@@ -784,6 +801,10 @@ std::shared_ptr<FileMetaData> ParquetFileReader::metadata() const {
   return contents_->metadata();
 }
 
+std::shared_ptr<PageIndexReader> ParquetFileReader::GetPageIndexReader() {
+  return contents_->GetPageIndexReader();
+}
+
 std::shared_ptr<RowGroupReader> ParquetFileReader::RowGroup(int i) {
   if (i >= metadata()->num_row_groups()) {
     std::stringstream ss;

cpp/src/parquet/file_reader.h

@@ -32,6 +32,7 @@ namespace parquet {
 
 class ColumnReader;
 class FileMetaData;
+class PageIndexReader;
 class PageReader;
 class RowGroupMetaData;
 
@@ -98,6 +99,7 @@ class PARQUET_EXPORT ParquetFileReader {
     virtual void Close() = 0;
     virtual std::shared_ptr<RowGroupReader> GetRowGroup(int i) = 0;
     virtual std::shared_ptr<FileMetaData> metadata() const = 0;
+    virtual std::shared_ptr<PageIndexReader> GetPageIndexReader() = 0;
   };
 
   ParquetFileReader();
@@ -133,6 +135,17 @@ class PARQUET_EXPORT ParquetFileReader {
   // Returns the file metadata. Only one instance is ever created
   std::shared_ptr<FileMetaData> metadata() const;
 
+  /// Returns the PageIndexReader. Only one instance is ever created.
+  ///
+  /// If the file does not have the page index, nullptr may be returned.
+  /// Because it pays to check existence of page index in the file, it
+  /// is possible to return a non null value even if page index does
+  /// not exist. It is the caller's responsibility to check the return
+  /// value and follow-up calls to PageIndexReader.
+  ///
+  /// WARNING: The returned PageIndexReader must not outlive the ParquetFileReader.
+  std::shared_ptr<PageIndexReader> GetPageIndexReader();
+
   /// Pre-buffer the specified column indices in all row groups.
   ///
   /// Readers can optionally call this to cache the necessary slices