#include <pattern.h>

Public Member Functions
	PatternPointer ()

	PatternPointer (unsigned char *dataref, const unsigned int bytesize)

	PatternPointer (const Pattern &ref)

	PatternPointer (const Pattern *ref)

	PatternPointer (const PatternPointer &ref)

	PatternPointer (const PatternPointer *ref)

PatternPointer &	operator= (const PatternPointer &other)

	PatternPointer (std::istream in, bool ignoreeol=false, const unsigned char version=2, unsigned char corpusstart=NULL, bool debug=false)

void	write (std::ostream out, const unsigned char corpusstart=NULL) const

	PatternPointer (unsigned char *, unsigned int, unsigned int)

	PatternPointer (const PatternPointer &, unsigned int, unsigned int)

	PatternPointer (const Pattern &, unsigned int, unsigned int)

uint32_t	computemask () const

const size_t	n () const

const size_t	bytesize () const

const size_t	size () const

const size_t	hash () const

const PatternCategory	category () const

const bool	isskipgram () const

const bool	isflexgram () const

std::string	tostring (const ClassDecoder &classdecoder) const

std::string	decode (const ClassDecoder &classdecoder) const

bool	out () const

bool	operator== (const PatternPointer &other) const

bool	operator!= (const PatternPointer &other) const

bool	operator== (const Pattern &other) const

bool	operator!= (const Pattern &other) const

PatternPointer	toflexgram () const

bool	isgap (int i) const

PatternPointer &	operator++ ()

bool	operator< (const PatternPointer &other) const

int	ngrams (std::vector< PatternPointer > &container, const int n) const

int	subngrams (std::vector< PatternPointer > &container, int minn=1, int maxn=9) const

int	ngrams (std::vector< std::pair< PatternPointer, int >> &container, const int n) const

int	subngrams (std::vector< std::pair< PatternPointer, int >> &container, int minn=1, int maxn=9) const

int	parts (std::vector< PatternPointer > &container) const

int	parts (std::vector< std::pair< int, int > > &container) const

int	gaps (std::vector< std::pair< int, int > > &container) const

const unsigned int	skipcount () const

PatternPointer	addskip (const std::pair< int, int > &gap) const

PatternPointer	addskips (const std::vector< std::pair< int, int > > &gaps) const

int	flexcollapse (unsigned char *collapseddata) const

bool	instanceof (const Pattern &skipgram) const

	operator Pattern ()

Pattern	pattern () const

Public Attributes
unsigned char *	data

uint32_t	bytes

uint32_t	mask

Static Public Attributes
static const int	patterntype = PATTERNPOINTER

Constructor & Destructor Documentation

PatternPointer::PatternPointer ( )

inline

PatternPointer::PatternPointer	(	unsigned char *	dataref,
		const unsigned int	bytesize
	)

inline

PatternPointer::PatternPointer ( const Pattern & ref )

inline

PatternPointer::PatternPointer ( const Pattern * ref )

inline

PatternPointer::PatternPointer ( const PatternPointer & ref )

inline

PatternPointer::PatternPointer ( const PatternPointer * ref )

inline

PatternPointer::PatternPointer	(	std::istream *	in,
		bool	ignoreeol = `false`,
		const unsigned char	version = `2`,
		unsigned char *	corpusstart = `NULL`,
		bool	debug = `false`
	)

PatternPointer::PatternPointer	(	unsigned char *	ref,
		unsigned int	begin,
		unsigned int	length
	)

PatternPointer::PatternPointer	(	const PatternPointer &	ref,
		unsigned int	begin,
		unsigned int	length
	)

PatternPointer::PatternPointer	(	const Pattern &	,
		unsigned	int,
		unsigned	int
	)

Member Function Documentation

PatternPointer PatternPointer::addskip ( const std::pair< int, int > & gap ) const

Replaces a series of tokens with a skip/gap of a particular size. Effectively turns a pattern into a skipgram.

Parameters

gap	The position and size of the skip/gap: a pair consisting of a begin index (0-indexed) and a length, i.e. the size of the skip

PatternPointer PatternPointer::addskips ( const std::vector< std::pair< int, int > > & gaps ) const

Replaces multiple series of tokens with skips/gaps of particular sizes. Effectively turns a pattern into a skipgram.

Parameters

gaps	The positions and sizes of the gaps: a vector of pairs, each pair consisting of a begin index (0-indexed) and a length, indicating where to place the gap

Returns: A skipgram

const size_t PatternPointer::bytesize ( ) const

inline

const PatternCategory PatternPointer::category ( ) const

uint32_t PatternPointer::computemask ( ) const

std::string PatternPointer::decode ( const ClassDecoder & classdecoder ) const

inline

int PatternPointer::flexcollapse ( unsigned char * collapseddata ) const

Low-level function for flexgrams, that returns a collapsed comparable representation of the flexgram in collapseddata (has to be pre-allocated). Return value is the number of bytes of the representation. In the collapsed representation adjacent flexgrams are removed.

int PatternPointer::gaps ( std::vector< std::pair< int, int > > & container ) const

Finds all the gaps of a skipgram or flexgram., parts are the portions that are not skips and adds them to container as begin,length pairs... Thus 'to be {*} not {*} be' has three parts. The gap-length of a flexgram will always be its minimum length one.

const size_t PatternPointer::hash ( ) const

Compute a hash value for this pattern

bool PatternPointer::instanceof ( const Pattern & skipgram ) const

const bool PatternPointer::isflexgram ( ) const

inline

bool PatternPointer::isgap ( int i ) const

const bool PatternPointer::isskipgram ( ) const

inline

const size_t PatternPointer::n ( ) const

int PatternPointer::ngrams	(	std::vector< PatternPointer > &	container,
		const int	n
	)		const

int PatternPointer::ngrams	(	std::vector< std::pair< PatternPointer, int >> &	container,
		const int	n
	)		const

PatternPointer::operator Pattern ( )

inline

bool PatternPointer::operator!= ( const PatternPointer & other ) const

inline

bool PatternPointer::operator!= ( const Pattern & other ) const

inline

PatternPointer & PatternPointer::operator++ ( )

Return a new patternpointer one token to the right, maintaining the same token length and same skip configuration (if any). Note that this will cause segmentation faults if the new PatternPointer exceeds the original data!!! It's up to the caller to check this!

bool PatternPointer::operator< ( const PatternPointer & other ) const

inline

PatternPointer& PatternPointer::operator= ( const PatternPointer & other )

inline

bool PatternPointer::operator== ( const PatternPointer & other ) const

bool PatternPointer::operator== ( const Pattern & other ) const

bool PatternPointer::out ( ) const

int PatternPointer::parts ( std::vector< PatternPointer > & container ) const

Finds all the parts of a skipgram, parts are the portions that are not skips and adds them to container... Thus 'to be {*} not {*} be' has three parts

int PatternPointer::parts ( std::vector< std::pair< int, int > > & container ) const

Finds all the parts of a skipgram, parts are the portions that are not skips and adds them to container as begin,length pairs... Thus 'to be {*} not {*} be' has three parts

Pattern PatternPointer::pattern ( ) const

inline

const size_t PatternPointer::size ( ) const

inline

const unsigned int PatternPointer::skipcount ( ) const

return the number of skips in this pattern

int PatternPointer::subngrams	(	std::vector< PatternPointer > &	container,
		int	minn = `1`,
		int	maxn = `9`
	)		const

int PatternPointer::subngrams	(	std::vector< std::pair< PatternPointer, int >> &	container,
		int	minn = `1`,
		int	maxn = `9`
	)		const

PatternPointer PatternPointer::toflexgram ( ) const

std::string PatternPointer::tostring ( const ClassDecoder & classdecoder ) const

void PatternPointer::write	(	std::ostream *	out,
		const unsigned char *	corpusstart = `NULL`
	)		const

Write Pattern to output stream (in binary form)

Parameters

out	The output stream

Member Data Documentation

uint32_t PatternPointer::bytes

Pointer to Pattern data

unsigned char* PatternPointer::data

uint32_t PatternPointer::mask

const int PatternPointer::patterntype = PATTERNPOINTER

static

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

include/pattern.h
src/pattern.cpp

Public Member Functions

Public Attributes

Static Public Attributes

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation