#include <hdfs-parquet-scanner.h>

Inheritance diagram for impala::HdfsParquetScanner:

Collaboration diagram for impala::HdfsParquetScanner:

Classes
class	BaseColumnReader

class	BoolColumnReader

class	ColumnReader

struct	FileVersion

struct	SchemaNode
	Internal representation of a column schema (including nested-type columns). More...

Public Member Functions
	HdfsParquetScanner (HdfsScanNode scan_node, RuntimeState state)

virtual	~HdfsParquetScanner ()

virtual Status	Prepare (ScannerContext *context)
	One-time initialisation of state that is constant across scan ranges. More...

virtual void	Close ()

virtual Status	ProcessSplit ()

Static Public Member Functions
static Status	IssueInitialRanges (HdfsScanNode scan_node, const std::vector< HdfsFileDesc > &files)

Static Public Attributes
static const int	FILE_BLOCK_SIZE = 4096

static const char *	LLVM_CLASS_NAME = "class.impala::HdfsScanner"

Protected Types
typedef int(*	WriteTuplesFn )(HdfsScanner , MemPool , TupleRow , int, FieldLocation , int, int, int, int)

Protected Member Functions
Status	InitializeWriteTuplesFn (HdfsPartitionDescriptor *partition, THdfsFileFormat::type type, const std::string &scanner_name)

void	StartNewRowBatch ()
	Set batch_ to a new row batch and update tuple_mem_ accordingly. More...

int	GetMemory (MemPool pool, Tuple tuple_mem, TupleRow **tuple_row_mem)

Status	CommitRows (int num_rows)

void	AddFinalRowBatch ()

void	AttachPool (MemPool *pool, bool commit_batch)

bool IR_ALWAYS_INLINE	EvalConjuncts (TupleRow *row)

int	WriteEmptyTuples (RowBatch *row_batch, int num_tuples)

int	WriteEmptyTuples (ScannerContext context, TupleRow tuple_row, int num_tuples)
	Write empty tuples and commit them to the context object. More...

int	WriteAlignedTuples (MemPool pool, TupleRow tuple_row_mem, int row_size, FieldLocation *fields, int num_tuples, int max_added_tuples, int slots_per_tuple, int row_start_indx)

Status	UpdateDecompressor (const THdfsCompression::type &compression)

Status	UpdateDecompressor (const std::string &codec)

bool	ReportTupleParseError (FieldLocation fields, uint8_t errors, int row_idx)

virtual void	LogRowParseError (int row_idx, std::stringstream *)

bool	WriteCompleteTuple (MemPool pool, FieldLocation fields, Tuple tuple, TupleRow tuple_row, Tuple template_tuple, uint8_t error_fields, uint8_t *error_in_row)

void	ReportColumnParseError (const SlotDescriptor desc, const char data, int len)

void	InitTuple (Tuple template_tuple, Tuple tuple)

Tuple *	next_tuple (Tuple *t) const

TupleRow *	next_row (TupleRow *r) const

ExprContext *	GetConjunctCtx (int idx) const

Static Protected Member Functions
static llvm::Function *	CodegenWriteCompleteTuple (HdfsScanNode , LlvmCodeGen , const std::vector< ExprContext * > &conjunct_ctxs)

static llvm::Function *	CodegenWriteAlignedTuples (HdfsScanNode , LlvmCodeGen , llvm::Function *write_tuple_fn)

Protected Attributes
HdfsScanNode *	scan_node_
	The scan node that started this scanner. More...

RuntimeState *	state_
	RuntimeState for error reporting. More...

ScannerContext *	context_
	Context for this scanner. More...

ScannerContext::Stream *	stream_
	The first stream for context_. More...

std::vector< ExprContext * >	conjunct_ctxs_

Tuple *	template_tuple_

int	tuple_byte_size_
	Fixed size of each tuple, in bytes. More...

Tuple *	tuple_
	Current tuple pointer into tuple_mem_. More...

RowBatch *	batch_

uint8_t *	tuple_mem_
	The tuple memory of batch_. More...

int	num_errors_in_file_
	number of errors in current file More...

boost::scoped_ptr< TextConverter >	text_converter_
	Helper class for converting text to other types;. More...

int32_t	num_null_bytes_
	Number of null bytes in the tuple. More...

boost::scoped_ptr< Codec >	decompressor_
	Decompressor class to use, if any. More...

THdfsCompression::type	decompression_type_
	The most recently used decompression type. More...

boost::scoped_ptr< MemPool >	data_buffer_pool_

RuntimeProfile::Counter *	decompress_timer_
	Time spent decompressing bytes. More...

WriteTuplesFn	write_tuples_fn_
	Jitted write tuples function pointer. Null if codegen is disabled. More...

Private Member Functions
Status	AssembleRows (int row_group_idx)

Status	ProcessFooter (bool *eosr)

Status	CreateColumnReaders ()

BaseColumnReader *	CreateReader (const SchemaNode &node)

Status	InitColumns (int row_group_idx)

Status	ValidateFileMetadata ()
	Validates the file metadata. More...

Status	ValidateColumn (const BaseColumnReader &col_reader, int row_group_idx)

Status	InitNewRange ()
	Part of the HdfsScanner interface, not used in Parquet. More...

Status	CreateSchemaTree (const std::vector< parquet::SchemaElement > &schema, SchemaNode *node) const

Status	CreateSchemaTree (const std::vector< parquet::SchemaElement > &schema, int max_def_level, int idx, int col_idx, SchemaNode *node) const
	Recursive implementation used internally by the above CreateSchemaTree() function. More...

Private Attributes
std::vector< BaseColumnReader * >	column_readers_
	Column reader for each materialized columns for this file. More...

parquet::FileMetaData	file_metadata_
	File metadata thrift object. More...

FileVersion	file_version_
	Version of the application that wrote this file. More...

SchemaNode	schema_
	The root schema node for this file. More...

const DiskIoMgr::ScanRange *	metadata_range_
	Scan range for the metadata. More...

Status	parse_status_
	Returned in ProcessSplit. More...

boost::scoped_ptr< MemPool >	dictionary_pool_

ScopedTimer< MonotonicStopWatch >	assemble_rows_timer_
	Timer for materializing rows. This ignores time getting the next buffer. More...

RuntimeProfile::Counter *	num_cols_counter_
	Number of cols that need to be read. More...

Static Private Attributes
static const int	FOOTER_SIZE = 100 * 1024

Friends
class	BaseColumnReader

template<typename T >
class	ColumnReader

class	BoolColumnReader

Detailed Description

This scanner parses Parquet files located in HDFS, and writes the content as tuples in the Impala in-memory representation of data, e.g. (tuples, rows, row batches). For the file format spec, see: github.com/Parquet/parquet-format Parquet (and other columnar formats) use scanner ranges differently than other formats. Each materialized column maps to a single ScanRange. For streaming reads, all the columns need to be read in parallel. This is done by issuing one ScanRange (in IssueInitialRanges()) for the file footer as the other scanners do. This footer range is processed in ProcessSplit(). ProcessSplit() then computes the column ranges and submits them to the IoMgr for immediate scheduling (so they don't surface in DiskIoMgr::GetNextRange()). Scheduling them immediately also guarantees they are all read at once. Like the other scanners, each parquet scanner object is one to one with a ScannerContext. Unlike the other scanners though, the context will have multiple streams, one for each column.

Definition at line 43 of file hdfs-parquet-scanner.h.

Member Typedef Documentation

typedef int(* impala::HdfsScanner::WriteTuplesFn)(HdfsScanner *, MemPool *, TupleRow *, int, FieldLocation *, int, int, int, int)

protectedinherited

Matching typedef for WriteAlignedTuples for codegen. Refer to comments for that function.

Definition at line 212 of file hdfs-scanner.h.

Constructor & Destructor Documentation

impala::HdfsParquetScanner::HdfsParquetScanner	(	HdfsScanNode *	scan_node,
		RuntimeState *	state
	)

Definition at line 125 of file hdfs-parquet-scanner.cc.

References assemble_rows_timer_.

impala::HdfsParquetScanner::~HdfsParquetScanner ( )

virtual

Definition at line 133 of file hdfs-parquet-scanner.cc.

Member Function Documentation

void HdfsScanner::AddFinalRowBatch ( )

protectedinherited

Attach all remaining resources from context_ to batch_ and send batch_ to the scan node. This must be called after all rows have been committed and no further resources are needed from context_ (in practice this will happen in each scanner subclass's Close() implementation).

Definition at line 145 of file hdfs-scanner.cc.

References impala::HdfsScanNode::AddMaterializedRowBatch(), impala::HdfsScanner::batch_, impala::HdfsScanner::context_, impala::ScannerContext::ReleaseCompletedResources(), and impala::HdfsScanner::scan_node_.

Referenced by impala::HdfsTextScanner::Close(), impala::BaseSequenceScanner::Close(), and Close().

Status HdfsParquetScanner::AssembleRows ( int row_group_idx )

private

Reads data from all the columns (in parallel) and assembles rows into the context object. Returns when the entire row group is complete or an error occurred.

Definition at line 805 of file hdfs-parquet-scanner.cc.

Referenced by ProcessSplit().

void impala::HdfsScanner::AttachPool	(	MemPool *	pool,
		bool	commit_batch
	)

inlineprotectedinherited

Release all memory in 'pool' to batch_. If commit_batch is true, the row batch will be committed. commit_batch should be true if the attached pool is expected to be non-trivial (i.e. a decompression buffer) to minimize scanner mem usage.

Definition at line 256 of file hdfs-scanner.h.

References impala::MemPool::AcquireData(), impala::HdfsScanner::batch_, impala::HdfsScanner::CommitRows(), and impala::RowBatch::tuple_data_pool().

Referenced by impala::HdfsTextScanner::Close(), impala::BaseSequenceScanner::Close(), Close(), impala::HdfsTextScanner::FillByteBufferGzip(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsSequenceScanner::ReadCompressedBlock(), impala::HdfsParquetScanner::BaseColumnReader::ReadDataPage(), and impala::HdfsRCFileScanner::ResetRowGroup().

void HdfsParquetScanner::Close ( )

virtual

Release all resources the scanner has allocated. This is the last chance for the scanner to attach any resources to the ScannerContext object.

Reimplemented from impala::HdfsScanner.

Definition at line 449 of file hdfs-parquet-scanner.cc.

References impala::HdfsScanner::AddFinalRowBatch(), assemble_rows_timer_, impala::HdfsScanner::AttachPool(), impala::HdfsScanner::Close(), column_readers_, dictionary_pool_, impala::HdfsScanNode::RangeComplete(), and impala::HdfsScanner::scan_node_.

Function * HdfsScanner::CodegenWriteAlignedTuples	(	HdfsScanNode *	,
		LlvmCodeGen *	,
		llvm::Function *	write_tuple_fn
	)

staticprotectedinherited

Codegen function to replace WriteAlignedTuples. WriteAlignedTuples is cross compiled to IR. This function loads the precompiled IR function, modifies it and returns the resulting function.

Definition at line 495 of file hdfs-scanner.cc.

References impala::LlvmCodeGen::codegen_timer(), impala::LlvmCodeGen::FinalizeFunction(), impala::LlvmCodeGen::GetFunction(), impala::LlvmCodeGen::ReplaceCallSites(), and SCOPED_TIMER.

Referenced by impala::HdfsTextScanner::Codegen(), and impala::HdfsSequenceScanner::Codegen().

Function * HdfsScanner::CodegenWriteCompleteTuple	(	HdfsScanNode *	,
		LlvmCodeGen *	,
		const std::vector< ExprContext * > &	conjunct_ctxs
	)

staticprotectedinherited

Codegen function to replace WriteCompleteTuple. Should behave identically to WriteCompleteTuple.

Definition at line 296 of file hdfs-scanner.cc.

Referenced by impala::HdfsTextScanner::Codegen(), and impala::HdfsSequenceScanner::Codegen().

Status HdfsScanner::CommitRows ( int num_rows )

protectedinherited

Commit num_rows to the current row batch. If this completes, the row batch is enqueued with the scan node and StartNewRowBatch() is called. Returns Status::OK if the query is not cancelled and hasn't exceeded any mem limits. Scanner can call this with 0 rows to flush any pending resources (attached pools and io buffers) to minimize memory consumption.

Definition at line 124 of file hdfs-scanner.cc.

References impala::HdfsScanNode::AddMaterializedRowBatch(), impala::RowBatch::AtCapacity(), impala::HdfsScanner::batch_, impala::TupleDescriptor::byte_size(), impala::Status::CANCELLED, impala::ScannerContext::cancelled(), impala::RowBatch::capacity(), impala::RuntimeState::CheckQueryState(), impala::RowBatch::CommitRows(), impala::HdfsScanner::conjunct_ctxs_, impala::HdfsScanner::context_, impala::ExprContext::FreeLocalAllocations(), impala::ScannerContext::num_completed_io_buffers(), impala::RowBatch::num_rows(), impala::Status::OK, impala::ScannerContext::ReleaseCompletedResources(), RETURN_IF_ERROR, impala::HdfsScanner::scan_node_, impala::HdfsScanner::StartNewRowBatch(), impala::HdfsScanner::state_, impala::HdfsScanNode::tuple_desc(), and impala::HdfsScanner::tuple_mem_.

Referenced by AssembleRows(), impala::HdfsScanner::AttachPool(), impala::HdfsTextScanner::FinishScanRange(), impala::HdfsSequenceScanner::ProcessDecompressedBlock(), ProcessFooter(), impala::HdfsTextScanner::ProcessRange(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsSequenceScanner::ProcessRange(), impala::HdfsRCFileScanner::ProcessRange(), and ProcessSplit().

Status HdfsParquetScanner::CreateColumnReaders ( )

private

Populates column_readers_ from the file schema. Schema resolution is handled in this function as well. We allow additional columns at the end in either the table or file schema. If there are extra columns in the file schema, it is simply ignored. If there are extra in the table schema, we return NULLs for those columns.

Definition at line 1050 of file hdfs-parquet-scanner.cc.

References impala::HdfsParquetScanner::SchemaNode::children, impala::SlotDescriptor::col_path(), column_readers_, CreateReader(), impala::HdfsParquetScanner::SchemaNode::DebugString(), impala::ScannerContext::Stream::filename(), gen_ir_descriptions::idx, impala::HdfsScanNode::InitEmptyTemplateTuple(), impala::HdfsScanNode::materialized_slots(), impala::SlotDescriptor::null_indicator_offset(), impala::HdfsScanNode::num_partition_keys(), impala::Status::OK, path(), impala::PrintPath(), impala::HdfsScanner::scan_node_, schema_, impala::Tuple::SetNull(), impala::HdfsParquetScanner::SchemaNode::slot_desc, impala::HdfsScanner::stream_, impala::HdfsScanner::template_tuple_, VLOG_FILE, and VLOG_QUERY.

Referenced by ProcessSplit().

HdfsParquetScanner::BaseColumnReader * HdfsParquetScanner::CreateReader ( const SchemaNode & node )

private

Creates a reader for node. node must refer to a non-nested column and node.slot_desc must be non-NULL. The reader is added to the runtime state's object pool.

Definition at line 472 of file hdfs-parquet-scanner.cc.

References impala::ObjectPool::Add(), BoolColumnReader, impala::ColumnType::GetByteSize(), impala::RuntimeState::obj_pool(), impala::HdfsScanNode::runtime_state(), impala::HdfsScanner::scan_node_, impala::HdfsParquetScanner::SchemaNode::slot_desc, impala::ColumnType::type, impala::SlotDescriptor::type(), impala::TYPE_BIGINT, impala::TYPE_BOOLEAN, impala::TYPE_CHAR, impala::TYPE_DECIMAL, impala::TYPE_DOUBLE, impala::TYPE_FLOAT, impala::TYPE_INT, impala::TYPE_SMALLINT, impala::TYPE_STRING, impala::TYPE_TIMESTAMP, impala::TYPE_TINYINT, and impala::TYPE_VARCHAR.

Referenced by CreateColumnReaders().

Status impala::HdfsParquetScanner::CreateSchemaTree	(	const std::vector< parquet::SchemaElement > &	schema,
		SchemaNode *	node
	)		const

private

Unflattens the schema metadata from a Parquet file metadata and converts it to our SchemaNode representation. Returns the result in 'n' unless an error status is returned. Does not set the slot_desc field of any SchemaNode.

Referenced by ProcessFooter().

Status impala::HdfsParquetScanner::CreateSchemaTree	(	const std::vector< parquet::SchemaElement > &	schema,
		int	max_def_level,
		int *	idx,
		int *	col_idx,
		SchemaNode *	node
	)		const

private

Recursive implementation used internally by the above CreateSchemaTree() function.

bool IR_ALWAYS_INLINE impala::HdfsScanner::EvalConjuncts ( TupleRow * row )

inlineprotectedinherited

Convenience function for evaluating conjuncts using this scanner's ExprContexts. This must always be inlined so we can correctly replace the call to ExecNode::EvalConjuncts() during codegen.

Definition at line 266 of file hdfs-scanner.h.

References impala::HdfsScanner::conjunct_ctxs_, and impala::ExecNode::EvalConjuncts().

Referenced by AssembleRows(), impala::HdfsAvroScanner::DecodeAvroData(), impala::HdfsRCFileScanner::ProcessRange(), impala::HdfsScanner::WriteCompleteTuple(), impala::HdfsScanner::WriteEmptyTuples(), and impala::HdfsTextScanner::WriteFields().

ExprContext * HdfsScanner::GetConjunctCtx ( int idx ) const

protectedinherited

Simple wrapper around conjunct_ctxs_. Used in the codegen'd version of WriteCompleteTuple() because it's easier than writing IR to access conjunct_ctxs_.

Definition at line 79 of file hdfs-scanner-ir.cc.

References impala::HdfsScanner::conjunct_ctxs_, and gen_ir_descriptions::idx.

int HdfsScanner::GetMemory	(	MemPool **	pool,
		Tuple **	tuple_mem,
		TupleRow **	tuple_row_mem
	)

protectedinherited

Gets memory for outputting tuples into batch_. *pool is the mem pool that should be used for memory allocated for those tuples. *tuple_mem should be the location to output tuples, and *tuple_row_mem for outputting tuple rows. Returns the maximum number of tuples/tuple rows that can be output (before the current row batch is complete and a new one is allocated). Memory returned from this call is invalidated after calling CommitRows. Callers must call GetMemory again after calling this function.

Definition at line 115 of file hdfs-scanner.cc.

References impala::RowBatch::AddRow(), impala::HdfsScanner::batch_, impala::RowBatch::capacity(), impala::RowBatch::GetRow(), impala::RowBatch::num_rows(), impala::RowBatch::tuple_data_pool(), and impala::HdfsScanner::tuple_mem_.

Referenced by AssembleRows(), impala::HdfsTextScanner::FinishScanRange(), impala::HdfsSequenceScanner::ProcessDecompressedBlock(), ProcessFooter(), impala::HdfsTextScanner::ProcessRange(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsSequenceScanner::ProcessRange(), and impala::HdfsRCFileScanner::ProcessRange().

Status HdfsParquetScanner::InitColumns ( int row_group_idx )

private

Walks file_metadata_ and initiates reading the materialized columns. This initializes column_readers_ and issues the reads for the columns.

Definition at line 1093 of file hdfs-parquet-scanner.cc.

References impala::DiskIoMgr::AddScanRanges(), impala::ScannerContext::AddStream(), impala::HdfsScanNode::AllocateScanRange(), impala::HdfsParquetScanner::FileVersion::application, column_readers_, impala::HdfsScanner::context_, impala::DiskIoMgr::RequestRange::disk_id(), impala::DiskIoMgr::ScanRange::expected_local(), impala::DiskIoMgr::RequestRange::file(), impala::HdfsFileDesc::file_length, file_metadata_, file_version_, impala::HdfsFileDesc::filename, impala::DiskIoMgr::RequestRange::fs(), impala::HdfsScanNode::GetFileDesc(), impala::RuntimeState::io_mgr(), impala::HdfsScanNode::materialized_slots(), MAX_DICT_HEADER_SIZE, metadata_range_, impala::HdfsFileDesc::mtime, impala::Status::OK, impala::HdfsScanNode::reader_context(), RETURN_IF_ERROR, impala::HdfsScanNode::runtime_state(), impala::HdfsScanner::scan_node_, impala::ScannerContext::Stream::set_contains_tuple_data(), impala::DiskIoMgr::ScanRange::try_cache(), ValidateColumn(), and impala::HdfsParquetScanner::FileVersion::VersionLt().

Referenced by ProcessSplit().

Status HdfsScanner::InitializeWriteTuplesFn	(	HdfsPartitionDescriptor *	partition,
		THdfsFileFormat::type	type,
		const std::string &	scanner_name
	)

protectedinherited

Initializes write_tuples_fn_ to the jitted function if codegen is possible.

partition - partition descriptor for this scanner/scan range
type - type for this scanner
scanner_name - debug string name for this scanner (e.g. HdfsTextScanner)

Definition at line 87 of file hdfs-scanner.cc.

References impala::HdfsPartitionDescriptor::escape_char(), impala::HdfsScanNode::GetCodegenFn(), impala::ExecNode::id(), impala::HdfsScanNode::IncNumScannersCodegenDisabled(), impala::HdfsScanNode::IncNumScannersCodegenEnabled(), impala::Status::OK, impala::HdfsScanner::scan_node_, impala::TupleDescriptor::string_slots(), impala::HdfsScanNode::tuple_desc(), and impala::HdfsScanner::write_tuples_fn_.

Referenced by impala::HdfsSequenceScanner::InitNewRange(), and impala::HdfsTextScanner::ResetScanner().

Status impala::HdfsParquetScanner::InitNewRange ( )

inlineprivatevirtual

Part of the HdfsScanner interface, not used in Parquet.

Implements impala::HdfsScanner.

Definition at line 184 of file hdfs-parquet-scanner.h.

References impala::Status::OK.

void impala::HdfsScanner::InitTuple	(	Tuple *	template_tuple,
		Tuple *	tuple
	)

inlineprotectedinherited

Initialize a tuple. TODO: only copy over non-null slots. TODO: InitTuple is called frequently, avoid the if, perhaps via templatization.

Definition at line 355 of file hdfs-scanner.h.

References impala::HdfsScanner::num_null_bytes_, and impala::HdfsScanner::tuple_byte_size_.

Referenced by AssembleRows(), impala::HdfsAvroScanner::DecodeAvroData(), impala::HdfsRCFileScanner::ProcessRange(), impala::HdfsScanner::WriteCompleteTuple(), and impala::HdfsTextScanner::WriteFields().

Status HdfsParquetScanner::IssueInitialRanges	(	HdfsScanNode *	scan_node,
		const std::vector< HdfsFileDesc * > &	files
	)

static

Issue just the footer range for each file. We'll then parse the footer and pick out the columns we want.

Definition at line 83 of file hdfs-parquet-scanner.cc.

References impala::HdfsScanNode::AddDiskIoRanges(), impala::HdfsScanNode::AllocateScanRange(), impala::DiskIoMgr::RequestRange::disk_id(), impala::DiskIoMgr::ScanRange::expected_local(), impala::RuntimeState::LogError(), impala::DiskIoMgr::ScanRange::meta_data(), impala::DiskIoMgr::RequestRange::offset(), impala::Status::OK, impala::ScanRangeMetadata::partition_id, impala::HdfsScanNode::RangeComplete(), RETURN_IF_ERROR, impala::HdfsScanNode::runtime_state(), and impala::DiskIoMgr::ScanRange::try_cache().

Referenced by impala::HdfsScanNode::GetNext().

void HdfsScanner::LogRowParseError	(	int	row_idx,
		std::stringstream *
	)

protectedvirtualinherited

Utility function to append an error message for an invalid row. This is called from ReportTupleParseError() row_idx is the index of the row in the current batch. Subclasses should override this function (i.e. text needs to join boundary rows). Since this is only in the error path, vtable overhead is acceptable.

Reimplemented in impala::HdfsSequenceScanner, and impala::HdfsTextScanner.

Definition at line 572 of file hdfs-scanner.cc.

Referenced by impala::HdfsScanner::ReportTupleParseError().

TupleRow* impala::HdfsScanner::next_row ( TupleRow * r ) const

inlineprotectedinherited

Definition at line 368 of file hdfs-scanner.h.

References impala::HdfsScanner::batch_, and impala::RowBatch::row_byte_size().

Referenced by AssembleRows(), impala::HdfsAvroScanner::DecodeAvroData(), impala::HdfsRCFileScanner::ProcessRange(), impala::HdfsScanner::WriteEmptyTuples(), and impala::HdfsTextScanner::WriteFields().

Tuple* impala::HdfsScanner::next_tuple ( Tuple * t ) const

inlineprotectedinherited

Definition at line 363 of file hdfs-scanner.h.

References impala::HdfsScanner::tuple_byte_size_.

Referenced by AssembleRows(), impala::HdfsAvroScanner::DecodeAvroData(), impala::HdfsRCFileScanner::ProcessRange(), and impala::HdfsTextScanner::WriteFields().

Status HdfsParquetScanner::Prepare ( ScannerContext * context )

virtual

One-time initialisation of state that is constant across scan ranges.

Reimplemented from impala::HdfsScanner.

Definition at line 440 of file hdfs-parquet-scanner.cc.

References ADD_COUNTER, impala::HdfsScanNode::IncNumScannersCodegenDisabled(), num_cols_counter_, impala::Status::OK, impala::HdfsScanner::Prepare(), RETURN_IF_ERROR, impala::ExecNode::runtime_profile(), and impala::HdfsScanner::scan_node_.

Status HdfsParquetScanner::ProcessFooter ( bool * eosr )

private

Process the file footer and parse file_metadata_. This should be called with the last FOOTER_SIZE bytes in context_. *eosr is a return value. If true, the scan range is complete (e.g. select count(*))

Definition at line 916 of file hdfs-parquet-scanner.cc.

References impala::HdfsScanNode::AllocateScanRange(), impala::DiskIoMgr::BufferDescriptor::buffer(), impala::HdfsScanner::CommitRows(), impala::HdfsScanner::context_, COUNTER_ADD, CreateSchemaTree(), impala::DeserializeThriftMsg(), impala::DiskIoMgr::RequestRange::disk_id(), impala::ScannerContext::Stream::eosr(), impala::DiskIoMgr::ScanRange::expected_local(), impala::DiskIoMgr::RequestRange::file(), impala::HdfsFileDesc::file_length, file_metadata_, impala::ScannerContext::Stream::filename(), impala::DiskIoMgr::RequestRange::fs(), impala::ScannerContext::Stream::GetBuffer(), impala::Status::GetDetail(), impala::HdfsScanNode::GetFileDesc(), impala::HdfsScanner::GetMemory(), impala::RuntimeState::io_mgr(), impala::DiskIoMgr::BufferDescriptor::len(), impala::HdfsScanNode::MarkFileDescIssued(), impala::HdfsScanNode::materialized_slots(), impala::DiskIoMgr::max_read_buffer_size(), metadata_range_, impala::HdfsFileDesc::mtime, impala::Status::OK, impala::Status::ok(), impala::PARQUET_VERSION_NUMBER, pool, impala::DiskIoMgr::Read(), impala::HdfsScanNode::reader_context(), impala::DiskIoMgr::BufferDescriptor::Return(), RETURN_IF_ERROR, impala::ScanNode::rows_read_counter(), impala::HdfsScanNode::runtime_state(), impala::HdfsScanner::scan_node_, impala::ScannerContext::Stream::scan_range(), schema_, impala::HdfsScanner::stream_, impala::DiskIoMgr::ScanRange::try_cache(), UNLIKELY, ValidateFileMetadata(), and impala::HdfsScanner::WriteEmptyTuples().

Referenced by ProcessSplit().

Status HdfsParquetScanner::ProcessSplit ( )

virtual

Process an entire split, reading bytes from the context's streams. Context is initialized with the split data (e.g. template tuple, partition descriptor, etc). This function should only return on error or end of scan range.

Implements impala::HdfsScanner.

Definition at line 769 of file hdfs-parquet-scanner.cc.

References AssembleRows(), impala::HdfsScanner::batch_, column_readers_, impala::HdfsScanner::CommitRows(), impala::HdfsScanner::context_, COUNTER_SET, CreateColumnReaders(), file_metadata_, InitColumns(), num_cols_counter_, impala::Status::OK, ProcessFooter(), impala::ScannerContext::ReleaseCompletedResources(), RETURN_IF_ERROR, and impala::HdfsScanner::stream_.

void HdfsScanner::ReportColumnParseError	(	const SlotDescriptor *	desc,
		const char *	data,
		int	len
	)

protectedinherited

Report parse error for column @ desc. If abort_on_error is true, sets parse_status_ to the error message.

Definition at line 577 of file hdfs-scanner.cc.

References impala::RuntimeState::abort_on_error(), impala::SlotDescriptor::col_pos(), impala::RuntimeState::LogError(), impala::RuntimeState::LogHasSpace(), impala::HdfsScanNode::num_partition_keys(), impala::Status::ok(), impala::HdfsScanner::parse_status_, impala::HdfsScanner::scan_node_, impala::HdfsScanner::state_, and impala::SlotDescriptor::type().

Referenced by impala::HdfsRCFileScanner::ProcessRange(), impala::HdfsScanner::ReportTupleParseError(), and impala::HdfsTextScanner::WritePartialTuple().

bool HdfsScanner::ReportTupleParseError	(	FieldLocation *	fields,
		uint8_t *	errors,
		int	row_idx
	)

protectedinherited

Utility function to report parse errors for each field. If errors[i] is nonzero, fields[i] had a parse error. row_idx is the idx of the row in the current batch that had the parse error Returns false if parsing should be aborted. In this case parse_status_ is set to the error. This is called from WriteAlignedTuples.

Definition at line 546 of file hdfs-scanner.cc.

References impala::RuntimeState::abort_on_error(), impala::ScannerContext::Stream::filename(), impala::RuntimeState::LogError(), impala::RuntimeState::LogHasSpace(), impala::HdfsScanner::LogRowParseError(), impala::HdfsScanNode::materialized_slots(), impala::HdfsScanner::num_errors_in_file_, impala::Status::ok(), impala::HdfsScanner::parse_status_, impala::HdfsScanner::ReportColumnParseError(), impala::RuntimeState::ReportFileErrors(), impala::HdfsScanner::scan_node_, impala::HdfsScanner::state_, and impala::HdfsScanner::stream_.

Referenced by impala::HdfsSequenceScanner::ProcessRange(), and impala::HdfsScanner::WriteAlignedTuples().

void HdfsScanner::StartNewRowBatch ( )

protectedinherited

Set batch_ to a new row batch and update tuple_mem_ accordingly.

Definition at line 108 of file hdfs-scanner.cc.

References impala::MemPool::Allocate(), impala::HdfsScanner::batch_, impala::RuntimeState::batch_size(), impala::ExecNode::mem_tracker(), impala::ExecNode::row_desc(), impala::HdfsScanner::scan_node_, impala::HdfsScanner::state_, impala::HdfsScanner::tuple_byte_size_, impala::RowBatch::tuple_data_pool(), and impala::HdfsScanner::tuple_mem_.

Referenced by impala::HdfsScanner::CommitRows(), and impala::HdfsScanner::Prepare().

Status HdfsScanner::UpdateDecompressor ( const THdfsCompression::type & compression )

protectedinherited

Update the decompressor_ object given a compression type or codec name. Depending on the old compression type and the new one, it may close the old decompressor and/or create a new one of different type.

Definition at line 513 of file hdfs-scanner.cc.

References impala::Codec::CreateDecompressor(), impala::HdfsScanner::data_buffer_pool_, impala::HdfsScanner::decompression_type_, impala::HdfsScanner::decompressor_, impala::Status::OK, RETURN_IF_ERROR, impala::HdfsScanner::scan_node_, impala::TupleDescriptor::string_slots(), and impala::HdfsScanNode::tuple_desc().

Referenced by impala::HdfsAvroScanner::InitNewRange(), impala::HdfsSequenceScanner::InitNewRange(), and impala::HdfsTextScanner::ProcessSplit().

Status impala::HdfsScanner::UpdateDecompressor ( const std::string & codec )

protectedinherited

Status HdfsParquetScanner::ValidateColumn	(	const BaseColumnReader &	col_reader,
		int	row_group_idx
	)

private

Validates the column metadata to make sure this column is supported (e.g. encoding, type, etc) and matches the type of col_reader's slot desc.

Definition at line 1290 of file hdfs-parquet-scanner.cc.

References impala::HdfsParquetScanner::BaseColumnReader::col_idx(), impala::ColumnType::DebugString(), impala::ParquetPlainEncoder::DecimalSize(), impala::DiskIoMgr::RequestRange::file(), file_metadata_, impala::IMPALA_TO_PARQUET_TYPES, IsEncodingSupported(), LOG_OR_RETURN_ON_ERROR, metadata_range_, impala::Status::OK, impala::ColumnType::precision, impala::PrintEncoding(), impala::ColumnType::scale, impala::HdfsParquetScanner::BaseColumnReader::schema_element(), impala::HdfsParquetScanner::BaseColumnReader::slot_desc(), impala::HdfsScanner::state_, impala::ColumnType::type, impala::SlotDescriptor::type(), and impala::TYPE_DECIMAL.

Referenced by InitColumns().

Status HdfsParquetScanner::ValidateFileMetadata ( )

private

Validates the file metadata.

Definition at line 1263 of file hdfs-parquet-scanner.cc.

References file_metadata_, file_version_, impala::ScannerContext::Stream::filename(), impala::Status::OK, impala::PARQUET_CURRENT_VERSION, and impala::HdfsScanner::stream_.

Referenced by ProcessFooter().

int HdfsScanner::WriteAlignedTuples	(	MemPool *	pool,
		TupleRow *	tuple_row_mem,
		int	row_size,
		FieldLocation *	fields,
		int	num_tuples,
		int	max_added_tuples,
		int	slots_per_tuple,
		int	row_start_indx
	)

protectedinherited

Processes batches of fields and writes them out to tuple_row_mem.

'pool' mempool to allocate from for auxiliary tuple memory
'tuple_row_mem' preallocated tuple_row memory this function must use.
'fields' must start at the beginning of a tuple.
'num_tuples' number of tuples to process
'max_added_tuples' the maximum number of tuples that should be added to the batch.
'row_start_index' is the number of rows that have already been processed as part of WritePartialTuple. Returns the number of tuples added to the row batch. This can be less than num_tuples/tuples_till_limit because of failed conjuncts. Returns -1 if parsing should be aborted due to parse errors.

Definition at line 33 of file hdfs-scanner-ir.cc.

References impala::HdfsScanner::ReportTupleParseError(), impala::HdfsScanner::template_tuple_, impala::HdfsScanner::tuple_, impala::HdfsScanner::tuple_byte_size_, UNLIKELY, and impala::HdfsScanner::WriteCompleteTuple().

Referenced by impala::HdfsSequenceScanner::ProcessDecompressedBlock(), and impala::HdfsTextScanner::WriteFields().

bool HdfsScanner::WriteCompleteTuple	(	MemPool *	pool,
		FieldLocation *	fields,
		Tuple *	tuple,
		TupleRow *	tuple_row,
		Tuple *	template_tuple,
		uint8_t *	error_fields,
		uint8_t *	error_in_row
	)

protectedinherited

Writes out all slots for 'tuple' from 'fields'. 'fields' must be aligned to the start of the tuple (e.g. fields[0] maps to slots[0]). After writing the tuple, it will be evaluated against the conjuncts.

error_fields is an out array. error_fields[i] will be set to true if the ith field had a parse error
error_in_row is an out bool. It is set to true if any field had parse errors Returns whether the resulting tuplerow passed the conjuncts. The parsing of the fields and evaluating against conjuncts is combined in this function. This is done so it can be possible to evaluate conjuncts as slots are materialized (on partial tuples). This function is replaced by a codegen'd function at runtime. This is the reason that the out error parameters are typed uint8_t instead of bool. We need to be able to match this function's signature identically for the codegen'd function. Bool's as out parameters can get converted to bytes by the compiler and rather than implicitly depending on that to happen, we will explicitly type them to bytes. TODO: revisit this

Definition at line 217 of file hdfs-scanner.cc.

References impala::HdfsScanner::EvalConjuncts(), impala::HdfsScanner::InitTuple(), impala::FieldLocation::len, impala::HdfsScanNode::materialized_slots(), impala::HdfsScanner::scan_node_, impala::TupleRow::SetTuple(), impala::HdfsScanner::text_converter_, impala::HdfsScanNode::tuple_idx(), and UNLIKELY.

Referenced by impala::HdfsSequenceScanner::ProcessRange(), and impala::HdfsScanner::WriteAlignedTuples().

int HdfsScanner::WriteEmptyTuples	(	RowBatch *	row_batch,
		int	num_tuples
	)

protectedinherited

Utility method to write out tuples when there are no materialized fields (e.g. select count(*) or only partition keys). num_tuples - Total number of tuples to write out. Returns the number of tuples added to the row batch.

Definition at line 157 of file hdfs-scanner.cc.

References impala::RowBatch::AddRow(), impala::RowBatch::AddRows(), impala::RowBatch::AtCapacity(), impala::RowBatch::capacity(), impala::RowBatch::CommitLastRow(), impala::RowBatch::CommitRows(), impala::HdfsScanner::EvalConjuncts(), impala::RowBatch::GetRow(), impala::RowBatch::INVALID_ROW_INDEX, impala::RowBatch::num_rows(), impala::HdfsScanner::scan_node_, impala::TupleRow::SetTuple(), impala::HdfsScanner::template_tuple_, and impala::HdfsScanNode::tuple_idx().

Referenced by impala::HdfsSequenceScanner::ProcessDecompressedBlock(), ProcessFooter(), impala::HdfsTextScanner::ProcessRange(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsSequenceScanner::ProcessRange(), and impala::HdfsRCFileScanner::ProcessRange().

int HdfsScanner::WriteEmptyTuples	(	ScannerContext *	context,
		TupleRow *	tuple_row,
		int	num_tuples
	)

protectedinherited

Write empty tuples and commit them to the context object.

Definition at line 195 of file hdfs-scanner.cc.

References impala::HdfsScanner::EvalConjuncts(), impala::HdfsScanner::next_row(), impala::HdfsScanner::scan_node_, impala::TupleRow::SetTuple(), impala::HdfsScanner::template_tuple_, and impala::HdfsScanNode::tuple_idx().

Friends And Related Function Documentation

friend class BaseColumnReader

friend

Definition at line 116 of file hdfs-parquet-scanner.h.

friend class BoolColumnReader

friend

Definition at line 121 of file hdfs-parquet-scanner.h.

Referenced by CreateReader().

template<typename T >

friend class ColumnReader

friend

Definition at line 119 of file hdfs-parquet-scanner.h.

Member Data Documentation

ScopedTimer<MonotonicStopWatch> impala::HdfsParquetScanner::assemble_rows_timer_

private

Timer for materializing rows. This ignores time getting the next buffer.

Definition at line 147 of file hdfs-parquet-scanner.h.

Referenced by AssembleRows(), Close(), and HdfsParquetScanner().

RowBatch* impala::HdfsScanner::batch_

protectedinherited

The current row batch being populated. Creating new row batches, attaching context resources, and handing off to the scan node is handled by this class in CommitRows(), but AttachPool() must be called by scanner subclasses to attach any memory allocated by that subclass. All row batches created by this class are transferred to the scan node (i.e., all batches are ultimately owned by the scan node).

Definition at line 177 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::AddFinalRowBatch(), impala::HdfsScanner::AttachPool(), impala::HdfsScanner::CommitRows(), impala::HdfsScanner::GetMemory(), impala::HdfsScanner::next_row(), impala::HdfsSequenceScanner::ProcessDecompressedBlock(), ProcessSplit(), impala::HdfsScanner::StartNewRowBatch(), impala::HdfsTextScanner::WriteFields(), and impala::HdfsScanner::~HdfsScanner().

std::vector<BaseColumnReader*> impala::HdfsParquetScanner::column_readers_

private

Column reader for each materialized columns for this file.

Definition at line 125 of file hdfs-parquet-scanner.h.

Referenced by AssembleRows(), Close(), CreateColumnReaders(), InitColumns(), and ProcessSplit().

std::vector<ExprContext*> impala::HdfsScanner::conjunct_ctxs_

protectedinherited

ExprContext for each conjunct. Each scanner has its own ExprContexts so the conjuncts can be safely evaluated in parallel.

Definition at line 154 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::Close(), impala::HdfsScanner::CommitRows(), impala::HdfsScanner::EvalConjuncts(), impala::HdfsScanner::GetConjunctCtx(), and impala::HdfsScanner::Prepare().

ScannerContext* impala::HdfsScanner::context_

protectedinherited

Context for this scanner.

Definition at line 147 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::AddFinalRowBatch(), AssembleRows(), impala::HdfsScanner::CommitRows(), impala::HdfsTextScanner::FillByteBufferCompressedFile(), impala::HdfsTextScanner::FillByteBufferGzip(), InitColumns(), impala::HdfsTextScanner::InitNewRange(), impala::HdfsSequenceScanner::InitNewRange(), impala::HdfsScanner::Prepare(), impala::HdfsSequenceScanner::ProcessDecompressedBlock(), ProcessFooter(), impala::HdfsTextScanner::ProcessRange(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsSequenceScanner::ProcessRange(), impala::HdfsRCFileScanner::ProcessRange(), ProcessSplit(), and impala::HdfsTextScanner::ResetScanner().

boost::scoped_ptr<MemPool> impala::HdfsScanner::data_buffer_pool_

protectedinherited

Pool to allocate per data block memory. This should be used with the decompressor and any other per data block allocations.

Definition at line 205 of file hdfs-scanner.h.

Referenced by impala::HdfsTextScanner::Close(), impala::BaseSequenceScanner::Close(), impala::HdfsTextScanner::FillByteBufferGzip(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsSequenceScanner::ReadCompressedBlock(), impala::HdfsRCFileScanner::ReadRowGroup(), impala::HdfsRCFileScanner::ResetRowGroup(), impala::HdfsScanner::UpdateDecompressor(), and impala::HdfsTextScanner::WritePartialTuple().

RuntimeProfile::Counter* impala::HdfsScanner::decompress_timer_

protectedinherited

Time spent decompressing bytes.

Definition at line 208 of file hdfs-scanner.h.

Referenced by impala::HdfsTextScanner::FillByteBufferCompressedFile(), impala::HdfsTextScanner::FillByteBufferGzip(), impala::HdfsSequenceScanner::GetRecord(), impala::HdfsScanner::Prepare(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsRCFileScanner::ReadColumnBuffers(), impala::HdfsSequenceScanner::ReadCompressedBlock(), impala::HdfsParquetScanner::BaseColumnReader::ReadDataPage(), and impala::HdfsRCFileScanner::ReadKeyBuffers().

THdfsCompression::type impala::HdfsScanner::decompression_type_

protectedinherited

The most recently used decompression type.

Definition at line 201 of file hdfs-scanner.h.

Referenced by impala::HdfsTextScanner::FillByteBuffer(), impala::HdfsTextScanner::FillByteBufferCompressedFile(), and impala::HdfsScanner::UpdateDecompressor().

boost::scoped_ptr<Codec> impala::HdfsScanner::decompressor_

protectedinherited

Decompressor class to use, if any.

Definition at line 198 of file hdfs-scanner.h.

Referenced by impala::HdfsTextScanner::Close(), impala::BaseSequenceScanner::Close(), impala::HdfsScanner::Close(), impala::HdfsTextScanner::FillByteBuffer(), impala::HdfsTextScanner::FillByteBufferCompressedFile(), impala::HdfsTextScanner::FillByteBufferGzip(), impala::HdfsTextScanner::FinishScanRange(), impala::HdfsSequenceScanner::GetRecord(), impala::HdfsRCFileScanner::InitNewRange(), impala::HdfsAvroScanner::ProcessRange(), impala::HdfsRCFileScanner::ReadColumnBuffers(), impala::HdfsSequenceScanner::ReadCompressedBlock(), impala::HdfsRCFileScanner::ReadKeyBuffers(), and impala::HdfsScanner::UpdateDecompressor().

boost::scoped_ptr<MemPool> impala::HdfsParquetScanner::dictionary_pool_

private

Pool to copy dictionary page buffer into. This pool is shared across all the pages in a column chunk.

Definition at line 144 of file hdfs-parquet-scanner.h.

Referenced by Close(), and impala::HdfsParquetScanner::BaseColumnReader::ReadDataPage().

const int impala::HdfsScanner::FILE_BLOCK_SIZE = 4096

staticinherited

Assumed size of an OS file block. Used mostly when reading file format headers, etc. This probably ought to be a derived number from the environment.

Definition at line 95 of file hdfs-scanner.h.

parquet::FileMetaData impala::HdfsParquetScanner::file_metadata_

private

File metadata thrift object.

Definition at line 128 of file hdfs-parquet-scanner.h.

Referenced by AssembleRows(), InitColumns(), ProcessFooter(), ProcessSplit(), ValidateColumn(), and ValidateFileMetadata().

FileVersion impala::HdfsParquetScanner::file_version_

private

Version of the application that wrote this file.

Definition at line 131 of file hdfs-parquet-scanner.h.

Referenced by impala::HdfsParquetScanner::ColumnReader< T >::ColumnReader(), InitColumns(), impala::HdfsParquetScanner::BaseColumnReader::ReadDataPage(), and ValidateFileMetadata().

const int impala::HdfsParquetScanner::FOOTER_SIZE = 100 * 1024

staticprivate

Size of the file footer. This is a guess. If this value is too little, we will need to issue another read.

Definition at line 113 of file hdfs-parquet-scanner.h.

const char * HdfsScanner::LLVM_CLASS_NAME = "class.impala::HdfsScanner"

staticinherited

Scanner subclasses must implement these static functions as well. Unfortunately, c++ does not allow static virtual functions. Issue the initial ranges for 'files'. HdfsFileDesc groups all the splits assigned to this scan node by file. This is called before any of the scanner subclasses are created to process splits in 'files'. The strategy on how to parse the scan ranges depends on the file format.

For simple text files, all the splits are simply issued to the io mgr and one split == one scan range.
For formats with a header, the metadata is first parsed, and then the ranges are issued to the io mgr. There is one scan range for the header and one range for each split.
For columnar formats, the header is parsed and only the relevant byte ranges should be issued to the io mgr. This is one range for the metadata and one range for each column, for each split. This function is how scanners can pick their strategy. void IssueInitialRanges(HdfsScanNode* scan_node, const std::vector<HdfsFileDesc*>& files); Codegen all functions for this scanner. The codegen'd function is specific to the scanner subclass but not specific to each scanner object. We don't want to codegen the functions for each scanner object. llvm::Function* Codegen(HdfsScanNode* scan_node);

Definition at line 137 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::CodegenWriteCompleteTuple().

const DiskIoMgr::ScanRange* impala::HdfsParquetScanner::metadata_range_

private

Scan range for the metadata.

Definition at line 137 of file hdfs-parquet-scanner.h.

Referenced by InitColumns(), ProcessFooter(), and ValidateColumn().

RuntimeProfile::Counter* impala::HdfsParquetScanner::num_cols_counter_

private

Number of cols that need to be read.

Definition at line 150 of file hdfs-parquet-scanner.h.

Referenced by Prepare(), and ProcessSplit().

int impala::HdfsScanner::num_errors_in_file_

protectedinherited

number of errors in current file

Definition at line 183 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::ReportTupleParseError().

int32_t impala::HdfsScanner::num_null_bytes_

protectedinherited

Number of null bytes in the tuple.

Definition at line 189 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::InitTuple().

Status impala::HdfsParquetScanner::parse_status_

private

Returned in ProcessSplit.

Definition at line 140 of file hdfs-parquet-scanner.h.

Referenced by AssembleRows(), and impala::HdfsParquetScanner::BoolColumnReader::ReadSlot().

HdfsScanNode* impala::HdfsScanner::scan_node_

protectedinherited

The scan node that started this scanner.

Definition at line 141 of file hdfs-scanner.h.

SchemaNode impala::HdfsParquetScanner::schema_

private

The root schema node for this file.

Definition at line 134 of file hdfs-parquet-scanner.h.

Referenced by CreateColumnReaders(), and ProcessFooter().

RuntimeState* impala::HdfsScanner::state_

protectedinherited

RuntimeState for error reporting.

Definition at line 144 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::Close(), impala::HdfsScanner::CommitRows(), impala::HdfsTextScanner::FillByteBufferGzip(), impala::HdfsTextScanner::FinishScanRange(), impala::HdfsTextScanner::Prepare(), impala::HdfsScanner::Prepare(), impala::HdfsSequenceScanner::Prepare(), impala::HdfsSequenceScanner::ProcessBlockCompressedScanRange(), impala::HdfsRCFileScanner::ProcessRange(), impala::BaseSequenceScanner::ProcessSplit(), impala::HdfsSequenceScanner::ReadCompressedBlock(), impala::BaseSequenceScanner::ReadPastSize(), impala::HdfsRCFileScanner::ReadRowGroup(), impala::HdfsScanner::ReportColumnParseError(), impala::HdfsScanner::ReportTupleParseError(), impala::HdfsAvroScanner::ResolveSchemas(), impala::HdfsScanner::StartNewRowBatch(), ValidateColumn(), and impala::HdfsTextScanner::WriteFields().

ScannerContext::Stream* impala::HdfsScanner::stream_

protectedinherited

The first stream for context_.

Definition at line 150 of file hdfs-scanner.h.

Tuple* impala::HdfsScanner::template_tuple_

protectedinherited

A partially materialized tuple with only partition key slots set. The non-partition key slots are set to NULL. The template tuple must be copied into tuple_ before any of the other slots are materialized. Pointer is NULL if there are no partition key slots. This template tuple is computed once for each file and valid for the duration of that file. It is owned by the HDFS scan node.

Definition at line 164 of file hdfs-scanner.h.

Referenced by impala::HdfsAvroScanner::AllocateFileHeader(), AssembleRows(), CreateColumnReaders(), impala::HdfsAvroScanner::DecodeAvroData(), impala::HdfsAvroScanner::InitNewRange(), impala::HdfsScanner::Prepare(), impala::HdfsSequenceScanner::ProcessRange(), impala::HdfsRCFileScanner::ProcessRange(), impala::HdfsAvroScanner::ResolveSchemas(), impala::HdfsScanner::WriteAlignedTuples(), impala::HdfsScanner::WriteEmptyTuples(), and impala::HdfsTextScanner::WriteFields().

boost::scoped_ptr<TextConverter> impala::HdfsScanner::text_converter_

protectedinherited

Helper class for converting text to other types;.

Definition at line 186 of file hdfs-scanner.h.

Referenced by impala::HdfsTextScanner::InitNewRange(), impala::HdfsSequenceScanner::InitNewRange(), impala::HdfsRCFileScanner::Prepare(), impala::HdfsRCFileScanner::ProcessRange(), impala::HdfsScanner::WriteCompleteTuple(), and impala::HdfsTextScanner::WritePartialTuple().

Tuple* impala::HdfsScanner::tuple_

protectedinherited

Current tuple pointer into tuple_mem_.

Definition at line 170 of file hdfs-scanner.h.

Referenced by impala::HdfsTextScanner::FinishScanRange(), impala::HdfsSequenceScanner::ProcessDecompressedBlock(), impala::HdfsTextScanner::ProcessRange(), impala::HdfsSequenceScanner::ProcessRange(), impala::HdfsScanner::WriteAlignedTuples(), and impala::HdfsTextScanner::WriteFields().

int impala::HdfsScanner::tuple_byte_size_

protectedinherited

Fixed size of each tuple, in bytes.

Definition at line 167 of file hdfs-scanner.h.

Referenced by AssembleRows(), impala::HdfsScanner::InitTuple(), impala::HdfsScanner::next_tuple(), impala::HdfsScanner::StartNewRowBatch(), and impala::HdfsScanner::WriteAlignedTuples().

uint8_t* impala::HdfsScanner::tuple_mem_

protectedinherited

The tuple memory of batch_.

Definition at line 180 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::CommitRows(), impala::HdfsScanner::GetMemory(), and impala::HdfsScanner::StartNewRowBatch().

WriteTuplesFn impala::HdfsScanner::write_tuples_fn_

protectedinherited

Jitted write tuples function pointer. Null if codegen is disabled.

Definition at line 215 of file hdfs-scanner.h.

Referenced by impala::HdfsScanner::InitializeWriteTuplesFn(), impala::HdfsSequenceScanner::ProcessDecompressedBlock(), and impala::HdfsTextScanner::WriteFields().

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

be/src/exec/hdfs-parquet-scanner.h
be/src/exec/hdfs-parquet-scanner.cc

Classes

Public Member Functions

Static Public Member Functions

Static Public Attributes

Protected Types

Protected Member Functions

Static Protected Member Functions

Protected Attributes

Private Member Functions

Private Attributes

Static Private Attributes

Friends

Detailed Description

Member Typedef Documentation

Constructor & Destructor Documentation

Member Function Documentation

Friends And Related Function Documentation

Member Data Documentation