#include <mmap_reader.hpp>

Classes
struct	RowGroupInfo
	Summary information for a single row group. More...

Public Member Functions
const Schema &	schema () const
	The file's column schema.

int64_t	num_rows () const
	Total number of rows across all row groups.

int64_t	num_row_groups () const
	Number of row groups in the file.

const std::string &	created_by () const
	The "created by" string from the file footer (may be empty).

const std::vector< thrift::KeyValue > &	key_value_metadata () const
	User-defined key-value metadata from the file footer.

RowGroupInfo	row_group (size_t index) const
	Retrieve summary information for a specific row group.

const thrift::Statistics *	column_statistics (size_t row_group_index, size_t column_index) const
	Retrieve the Thrift Statistics for a column chunk.

template<typename T >
expected< std::vector< T > >	read_column (size_t row_group_index, size_t column_index)
	Read an entire column from a row group as a typed vector.

expected< std::vector< std::string > >	read_column_as_strings (size_t row_group_index, size_t column_index)
	Read a column and convert all values to their string representations.

expected< std::vector< std::vector< std::string > > >	read_all ()
	Read all rows from all row groups as a vector of string rows.

const MmapReader &	mmap () const
	Direct access to the memory-mapped file data.

	~MmapParquetReader ()=default
	Default destructor.

	MmapParquetReader (MmapParquetReader &&) noexcept=default
	Move-constructible.

MmapParquetReader &	operator= (MmapParquetReader &&) noexcept=default
	Move-assignable.

Static Public Member Functions
static expected< MmapParquetReader >	open (const std::filesystem::path &path)
	Open a Parquet file with memory-mapped I/O.

Detailed Description

Definition at line 266 of file mmap_reader.hpp.

Constructor & Destructor Documentation

◆ ~MmapParquetReader()

signet::forge::MmapParquetReader::~MmapParquetReader ( )

default

Default destructor.

◆ MmapParquetReader()

signet::forge::MmapParquetReader::MmapParquetReader ( MmapParquetReader && )

defaultnoexcept

Move-constructible.

Member Function Documentation

◆ column_statistics()

const thrift::Statistics * signet::forge::MmapParquetReader::column_statistics	(	size_t	row_group_index,
		size_t	column_index
	)		const

inline

Retrieve the Thrift Statistics for a column chunk.

Parameters

row_group_index	Zero-based row group index.
column_index	Zero-based column index.

Returns: Pointer to the Statistics struct, or nullptr if unavailable.

Definition at line 442 of file mmap_reader.hpp.

◆ created_by()

const std::string & signet::forge::MmapParquetReader::created_by ( ) const

inline

The "created by" string from the file footer (may be empty).

Definition at line 399 of file mmap_reader.hpp.

◆ key_value_metadata()

const std::vector< thrift::KeyValue > & signet::forge::MmapParquetReader::key_value_metadata ( ) const

inline

User-defined key-value metadata from the file footer.

Returns: A reference to the metadata vector, or an empty vector if none.

Definition at line 403 of file mmap_reader.hpp.

◆ mmap()

const MmapReader & signet::forge::MmapParquetReader::mmap ( ) const

inline

Direct access to the memory-mapped file data.

Definition at line 708 of file mmap_reader.hpp.

◆ num_row_groups()

int64_t signet::forge::MmapParquetReader::num_row_groups ( ) const

inline

Number of row groups in the file.

Definition at line 394 of file mmap_reader.hpp.

◆ num_rows()

int64_t signet::forge::MmapParquetReader::num_rows ( ) const

inline

Total number of rows across all row groups.

Definition at line 391 of file mmap_reader.hpp.

◆ open()

static expected< MmapParquetReader > signet::forge::MmapParquetReader::open ( const std::filesystem::path & path )

inlinestatic

Open a Parquet file with memory-mapped I/O.

Validates the file structure (PAR1 magic, footer length), deserializes the FileMetaData, and reconstructs the column schema.

Parameters

path	Filesystem path to the Parquet file.

Returns: An MmapParquetReader on success, or an Error.

Definition at line 275 of file mmap_reader.hpp.

◆ operator=()

MmapParquetReader & signet::forge::MmapParquetReader::operator= ( MmapParquetReader && )

defaultnoexcept

Move-assignable.

◆ read_all()

expected< std::vector< std::vector< std::string > > > signet::forge::MmapParquetReader::read_all ( )

inline

Read all rows from all row groups as a vector of string rows.

Each inner vector represents one row, with columns converted to strings via read_column_as_strings(). Useful for inspection and debugging, but not recommended for high-performance paths.

Returns: A 2D vector [row][column] of string values, or an Error.

Definition at line 666 of file mmap_reader.hpp.

◆ read_column()

template<typename T >

expected< std::vector< T > > signet::forge::MmapParquetReader::read_column	(	size_t	row_group_index,
		size_t	column_index
	)

inline

Read an entire column from a row group as a typed vector.

Automatically detects the encoding strategy (PLAIN, dictionary, DELTA_BINARY_PACKED, BYTE_STREAM_SPLIT, or RLE boolean) and dispatches to the appropriate decoder.

Template Parameters

T	The C++ type to decode into (bool, int32_t, int64_t, float, double, std::string).

Parameters

row_group_index	Zero-based row group index.
column_index	Zero-based column index.

Returns: A vector of decoded values, or an Error.

Definition at line 467 of file mmap_reader.hpp.

◆ read_column_as_strings()

expected< std::vector< std::string > > signet::forge::MmapParquetReader::read_column_as_strings	(	size_t	row_group_index,
		size_t	column_index
	)

inline

Read a column and convert all values to their string representations.

Booleans become "true"/"false", numerics use std::to_string(), BYTE_ARRAY values are returned as-is, and FIXED_LEN_BYTE_ARRAY values are hex-encoded.

Parameters

row_group_index	Zero-based row group index.
column_index	Zero-based column index.

Returns: A vector of string-converted values, or an Error.

Definition at line 597 of file mmap_reader.hpp.

◆ row_group()

RowGroupInfo signet::forge::MmapParquetReader::row_group ( size_t index ) const

inline

Retrieve summary information for a specific row group.

Parameters

index Zero-based row group index.

Returns: A RowGroupInfo struct.

Exceptions

std::out_of_range if index >= num_row_groups().

Definition at line 423 of file mmap_reader.hpp.

◆ schema()

const Schema & signet::forge::MmapParquetReader::schema ( ) const

inline

The file's column schema.

Definition at line 388 of file mmap_reader.hpp.

The documentation for this class was generated from the following file:

include/signet/mmap_reader.hpp

Classes

Public Member Functions

Static Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ ~MmapParquetReader()

◆ MmapParquetReader()

Member Function Documentation

◆ column_statistics()

◆ created_by()

◆ key_value_metadata()

◆ mmap()

◆ num_row_groups()

◆ num_rows()

◆ open()

◆ operator=()

◆ read_all()

◆ read_column()

◆ read_column_as_strings()

◆ row_group()

◆ schema()