From 7b945968e02a7156e7d01af3896a473e6787e9c9 Mon Sep 17 00:00:00 2001
From: Jan Travnicek <Jan.Travnicek@fit.cvut.cz>
Date: Tue, 6 Dec 2016 10:24:33 +0100
Subject: [PATCH] template some arbology algorithms
---
.../src/arbology/exact/BoyerMooreHorspool.cpp | 114 ---------
.../src/arbology/exact/BoyerMooreHorspool.h | 125 ++++++++-
...neUsingBadCharacterShiftAndBorderArray.cpp | 106 --------
...ZoneUsingBadCharacterShiftAndBorderArray.h | 128 +++++++++-
.../exact/ReversedBoyerMooreHorspool.cpp | 219 ----------------
.../exact/ReversedBoyerMooreHorspool.h | 239 +++++++++++++++++-
.../properties/ExactSubtreeRepeatsNaive.cpp | 93 +++++++
.../properties/ExactSubtreeRepeatsNaive.h | 118 +--------
8 files changed, 576 insertions(+), 566 deletions(-)
diff --git a/alib2algo/src/arbology/exact/BoyerMooreHorspool.cpp b/alib2algo/src/arbology/exact/BoyerMooreHorspool.cpp
index 2df5634ce4..d31d1e892b 100644
--- a/alib2algo/src/arbology/exact/BoyerMooreHorspool.cpp
+++ b/alib2algo/src/arbology/exact/BoyerMooreHorspool.cpp
@@ -6,17 +6,6 @@
*/
#include "BoyerMooreHorspool.h"
-#include <tree/properties/BadCharacterShiftTable.h>
-#include <tree/properties/SubtreeJumpTable.h>
-#include <tree/properties/ExactSubtreeRepeatsNaive.h>
-
-#include <tree/Tree.h>
-#include <tree/ranked/PrefixRankedBarTree.h>
-#include <tree/ranked/PrefixRankedBarPattern.h>
-#include <tree/ranked/PrefixRankedBarNonlinearPattern.h>
-#include <alphabet/RankedSymbol.h>
-
-#include <map>
namespace arbology {
@@ -26,111 +15,8 @@ std::set < unsigned > BoyerMooreHorspool::match ( const tree::Tree & subject, co
return dispatch ( subject.getData ( ), pattern.getData ( ) );
}
-std::set < unsigned > BoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarTree < > & pattern ) {
- return match ( subject, tree::PrefixRankedBarPattern < > ( pattern ) );
-}
-
auto BoyerMooreHorspoolPrefixRankedBarTreePrefixRankedBarTree = BoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarTree < > > ( BoyerMooreHorspool::match );
-
-std::set < unsigned > BoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bcs = tree::properties::BadCharacterShiftTable::bcs ( pattern ); //NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
-
- // index to the subject
- unsigned i = 0;
-
- // main loop of the algorithm over all possible indexes where the pattern can start
- while ( i + pattern.getContent ( ).size ( ) <= subject.getContent ( ).size ( ) ) {
-
- // index to the pattern
- int j = pattern.getContent ( ).size ( ) - 1;
-
- // offset to the subject
- int offset = i + j;
-
- while ( ( j >= 0 ) && ( offset >= 0 ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
- // match of symbol
- offset = offset - 1;
- j = j - 1;
- } else if ( ( pattern.getContent ( )[j] == pattern.getVariablesBar ( ) ) && ( pattern.getBars ( ).count ( subject.getContent ( )[offset] )) ) {
- // match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- j = j - 2;
- } else {
- break;
- }
- }
-
- // match was found
- if ( j == -1 ) occ.insert ( offset );
-
- // shift heuristics
- i += bcs[subject.getContent ( )[i + pattern.getContent ( ).size ( ) - 1]];
- }
-
- return occ;
-}
-
auto BoyerMooreHorspoolPrefixRankedBarTreePrefixRankedBarPattern = BoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarPattern < > > ( BoyerMooreHorspool::match );
-
-std::set < unsigned > BoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarNonlinearPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bcs = tree::properties::BadCharacterShiftTable::bcs ( pattern ); //NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
- std::map < std::ranked_symbol < >, alphabet::Symbol > variablesSetting;
-
- tree::PrefixRankedBarTree < > repeats = tree::properties::ExactSubtreeRepeatsNaive::repeats ( subject );
-
- // index to the subject
- unsigned i = 0;
-
- // main loop of the algorithm over all possible indexes where the pattern can start
- while ( i + pattern.getContent ( ).size ( ) <= subject.getContent ( ).size ( ) ) {
- // clear the current state of variable to subtree repeat
- variablesSetting.clear();
-
- // index to the pattern
- int j = pattern.getContent ( ).size ( ) - 1;
-
- // offset to the subject
- int offset = i + j;
-
- while ( ( j >= 0 ) && ( offset >= 0 ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
- // match of symbol
- offset = offset - 1;
- j = j - 1;
- } else if ( ( pattern.getContent ( )[j] == pattern.getVariablesBar ( ) ) && ( pattern.getBars ( ).count ( subject.getContent ( )[offset] )) ) {
- // else match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- j = j - 2;
-
- // check nonlinear variable
- if ( pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j + 1 ] ) ) {
- auto setting = variablesSetting.find ( pattern.getContent ( )[ j + 1 ] );
-
- if ( setting != variablesSetting.end ( ) && repeats.getContent ( )[ offset + 1 ].getSymbol ( ) != setting->second )
- break;
-
- variablesSetting.insert ( std::make_pair ( pattern.getContent ( )[ j + 1 ], repeats.getContent( )[ offset + 1 ].getSymbol ( ) ) );
- }
- } else {
- break;
- }
- }
-
- // match was found
- if ( j == -1 ) occ.insert ( offset );
-
- // shift heuristics
- i += bcs[subject.getContent ( )[i + pattern.getContent ( ).size ( ) - 1]];
- }
-
- return occ;
-}
-
auto BoyerMooreHorspoolPrefixRankedBarTreePrefixRankedBarNonlinearPattern = BoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarNonlinearPattern < > > ( BoyerMooreHorspool::match );
} /* namespace exact */
diff --git a/alib2algo/src/arbology/exact/BoyerMooreHorspool.h b/alib2algo/src/arbology/exact/BoyerMooreHorspool.h
index be6136aead..67ad272db2 100644
--- a/alib2algo/src/arbology/exact/BoyerMooreHorspool.h
+++ b/alib2algo/src/arbology/exact/BoyerMooreHorspool.h
@@ -9,9 +9,21 @@
#define _ARBOLOGY_BOYER_MOORE_HORSPOOL_H_
#include <set>
+#include <map>
+
#include <core/multipleDispatch.hpp>
#include <tree/TreeFeatures.h>
+#include <tree/properties/BadCharacterShiftTable.h>
+#include <tree/properties/SubtreeJumpTable.h>
+#include <tree/properties/ExactSubtreeRepeatsNaive.h>
+
+#include <tree/Tree.h>
+#include <tree/ranked/PrefixRankedBarTree.h>
+#include <tree/ranked/PrefixRankedBarPattern.h>
+#include <tree/ranked/PrefixRankedBarNonlinearPattern.h>
+#include <alphabet/RankedSymbol.h>
+
namespace arbology {
namespace exact {
@@ -28,12 +40,119 @@ public:
*/
static std::set < unsigned > match ( const tree::Tree & subject, const tree::Tree & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarTree < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarNonlinearPattern < > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarTree < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarNonlinearPattern < SymbolType, RankType > & pattern );
};
+template < class SymbolType, class RankType >
+std::set < unsigned > BoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarTree < SymbolType, RankType > & pattern ) {
+ return match ( subject, tree::PrefixRankedBarPattern < SymbolType, RankType > ( pattern ) );
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > BoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bcs = tree::properties::BadCharacterShiftTable::bcs ( pattern ); //NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+
+ // index to the subject
+ unsigned i = 0;
+
+ // main loop of the algorithm over all possible indexes where the pattern can start
+ while ( i + pattern.getContent ( ).size ( ) <= subject.getContent ( ).size ( ) ) {
+
+ // index to the pattern
+ int j = pattern.getContent ( ).size ( ) - 1;
+
+ // offset to the subject
+ int offset = i + j;
+
+ while ( ( j >= 0 ) && ( offset >= 0 ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
+ // match of symbol
+ offset = offset - 1;
+ j = j - 1;
+ } else if ( ( pattern.getContent ( )[j] == pattern.getVariablesBar ( ) ) && ( pattern.getBars ( ).count ( subject.getContent ( )[offset] )) ) {
+ // match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ j = j - 2;
+ } else {
+ break;
+ }
+ }
+
+ // match was found
+ if ( j == -1 ) occ.insert ( offset );
+
+ // shift heuristics
+ i += bcs[subject.getContent ( )[i + pattern.getContent ( ).size ( ) - 1]];
+ }
+
+ return occ;
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > BoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarNonlinearPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bcs = tree::properties::BadCharacterShiftTable::bcs ( pattern ); //NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+ std::map < std::ranked_symbol < SymbolType, RankType >, SymbolType > variablesSetting;
+
+ tree::PrefixRankedBarTree < > repeats = tree::properties::ExactSubtreeRepeatsNaive::repeats ( subject );
+
+ // index to the subject
+ unsigned i = 0;
+
+ // main loop of the algorithm over all possible indexes where the pattern can start
+ while ( i + pattern.getContent ( ).size ( ) <= subject.getContent ( ).size ( ) ) {
+ // clear the current state of variable to subtree repeat
+ variablesSetting.clear();
+
+ // index to the pattern
+ int j = pattern.getContent ( ).size ( ) - 1;
+
+ // offset to the subject
+ int offset = i + j;
+
+ while ( ( j >= 0 ) && ( offset >= 0 ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
+ // match of symbol
+ offset = offset - 1;
+ j = j - 1;
+ } else if ( ( pattern.getContent ( )[j] == pattern.getVariablesBar ( ) ) && ( pattern.getBars ( ).count ( subject.getContent ( )[offset] )) ) {
+ // else match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ j = j - 2;
+
+ // check nonlinear variable
+ if ( pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j + 1 ] ) ) {
+ auto setting = variablesSetting.find ( pattern.getContent ( )[ j + 1 ] );
+
+ if ( setting != variablesSetting.end ( ) && repeats.getContent ( )[ offset + 1 ].getSymbol ( ) != setting->second )
+ break;
+
+ variablesSetting.insert ( std::make_pair ( pattern.getContent ( )[ j + 1 ], repeats.getContent( )[ offset + 1 ].getSymbol ( ) ) );
+ }
+ } else {
+ break;
+ }
+ }
+
+ // match was found
+ if ( j == -1 ) occ.insert ( offset );
+
+ // shift heuristics
+ i += bcs[subject.getContent ( )[i + pattern.getContent ( ).size ( ) - 1]];
+ }
+
+ return occ;
+}
+
} /* namespace exact */
} /* namespace arbology */
diff --git a/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.cpp b/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.cpp
index 436381fbdf..8ef1994b84 100644
--- a/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.cpp
+++ b/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.cpp
@@ -6,18 +6,6 @@
*/
#include "DeadZoneUsingBadCharacterShiftAndBorderArray.h"
-#include <tree/properties/ReversedBadCharacterShiftTable.h>
-#include <tree/properties/BorderArrayNaive.h>
-#include <tree/properties/SubtreeJumpTable.h>
-
-#include <tree/Tree.h>
-#include <tree/ranked/PrefixRankedBarTree.h>
-#include <tree/ranked/PrefixRankedBarPattern.h>
-#include <tree/ranked/PrefixRankedTree.h>
-#include <tree/ranked/PrefixRankedPattern.h>
-#include <alphabet/RankedSymbol.h>
-
-#include <map>
namespace arbology {
@@ -27,103 +15,9 @@ std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( cons
return dispatch ( subject.getData ( ), pattern.getData ( ) );
}
-std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarTree < > & pattern ) {
- return match ( subject, tree::PrefixRankedBarPattern < > ( pattern ) );
-}
-
auto DeadZoneUsingBadCharacterShiftAndBorderArrayPrefixRankedBarTreePrefixRankedBarTree = DeadZoneUsingBadCharacterShiftAndBorderArray::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarTree < > > ( DeadZoneUsingBadCharacterShiftAndBorderArray::match );
-
-std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bbcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < size_t > fba = tree::properties::BorderArrayNaive::ba ( pattern );
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
-
- match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, 0, subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1 );
- return occ;
-}
-
-void DeadZoneUsingBadCharacterShiftAndBorderArray::match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high ) {
- if ( low >= high ) return;
-
- int i = ( low + high ) / 2;
-
- // index to the pattern
- unsigned j = 0;
-
- // offset to the subject
- unsigned offset = i;
-
- while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
- // match of symbol
- offset++;
- j++;
- } else if ( ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) ) && ( ! pattern.getBars ( ).count ( subject.getContent ( )[offset] ) ) ) {
- // match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- j += 2;
- } else {
- break;
- }
- }
-
- // match was found
- if ( j >= pattern.getContent ( ).size ( ) ) occ.insert ( i );
-
- match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, low, i - bbcs[subject.getContent ( )[i]] + 1 );
- match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, i + j - fba[j], high );
-}
-
auto DeadZoneUsingBadCharacterShiftAndBorderArrayPrefixRankedBarTreePrefixRankedBarPattern = DeadZoneUsingBadCharacterShiftAndBorderArray::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarPattern < > > ( DeadZoneUsingBadCharacterShiftAndBorderArray::match );
-
-std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedTree < > & pattern ) {
- return match ( subject, tree::PrefixRankedPattern < > ( pattern ) );
-}
-
auto DeadZoneUsingBadCharacterShiftAndBorderArrayPrefixRankedTreePrefixRankedTree = DeadZoneUsingBadCharacterShiftAndBorderArray::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedTree < >, tree::PrefixRankedTree < > > ( DeadZoneUsingBadCharacterShiftAndBorderArray::match );
-
-std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bbcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < size_t > fba = tree::properties::BorderArrayNaive::ba ( pattern );
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
-
- match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, 0, subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1 );
- return occ;
-}
-
-void DeadZoneUsingBadCharacterShiftAndBorderArray::match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedPattern < > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high ) {
- if ( low >= high ) return;
-
- int i = ( low + high ) / 2;
-
- // index to the pattern
- unsigned j = 0;
-
- // offset to the subject
- unsigned offset = i;
-
- while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] )
- // match of symbol
- offset++;
- else if ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) )
- // match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- else
- break;
-
- j++;
- }
-
- // match was found
- if ( j >= pattern.getContent ( ).size ( ) ) occ.insert ( i );
-
- match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, low, i - bbcs[subject.getContent ( )[i]] + 1 );
- match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, i + j - fba[j], high );
-}
-
auto DeadZoneUsingBadCharacterShiftAndBorderArrayPrefixRankedTreePrefixRankedPattern = DeadZoneUsingBadCharacterShiftAndBorderArray::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedTree < >, tree::PrefixRankedPattern < > > ( DeadZoneUsingBadCharacterShiftAndBorderArray::match );
} /* namespace exact */
diff --git a/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.h b/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.h
index 16c2498691..51f1a64787 100644
--- a/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.h
+++ b/alib2algo/src/arbology/exact/DeadZoneUsingBadCharacterShiftAndBorderArray.h
@@ -14,6 +14,19 @@
#include <tree/TreeFeatures.h>
#include <alphabet/RankedSymbol.h>
+#include <tree/properties/ReversedBadCharacterShiftTable.h>
+#include <tree/properties/BorderArrayNaive.h>
+#include <tree/properties/SubtreeJumpTable.h>
+
+#include <tree/Tree.h>
+#include <tree/ranked/PrefixRankedBarTree.h>
+#include <tree/ranked/PrefixRankedBarPattern.h>
+#include <tree/ranked/PrefixRankedTree.h>
+#include <tree/ranked/PrefixRankedPattern.h>
+#include <alphabet/RankedSymbol.h>
+
+#include <map>
+
namespace arbology {
namespace exact {
@@ -29,16 +42,119 @@ public:
*/
static std::set < unsigned > match ( const tree::Tree & subject, const tree::Tree & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarTree < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern );
- static void match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarTree < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static void match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < SymbolType, RankType >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high );
- static std::set < unsigned > match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedTree < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedPattern < > & pattern );
- static void match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedPattern < > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedTree < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedPattern < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static void match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedPattern < SymbolType, RankType > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < SymbolType, RankType >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high );
};
+template < class SymbolType, class RankType >
+std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarTree < SymbolType, RankType > & pattern ) {
+ return match ( subject, tree::PrefixRankedBarPattern < SymbolType, RankType > ( pattern ) );
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bbcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < size_t > fba = tree::properties::BorderArrayNaive::ba ( pattern );
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+
+ match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, 0, subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1 );
+ return occ;
+}
+
+template < class SymbolType, class RankType >
+void DeadZoneUsingBadCharacterShiftAndBorderArray::match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < SymbolType, RankType >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high ) {
+ if ( low >= high ) return;
+
+ int i = ( low + high ) / 2;
+
+ // index to the pattern
+ unsigned j = 0;
+
+ // offset to the subject
+ unsigned offset = i;
+
+ while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
+ // match of symbol
+ offset++;
+ j++;
+ } else if ( ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) ) && ( ! pattern.getBars ( ).count ( subject.getContent ( )[offset] ) ) ) {
+ // match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ j += 2;
+ } else {
+ break;
+ }
+ }
+
+ // match was found
+ if ( j >= pattern.getContent ( ).size ( ) ) occ.insert ( i );
+
+ match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, low, i - bbcs[subject.getContent ( )[i]] + 1 );
+ match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, i + j - fba[j], high );
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedTree < SymbolType, RankType > & pattern ) {
+ return match ( subject, tree::PrefixRankedPattern < SymbolType, RankType > ( pattern ) );
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > DeadZoneUsingBadCharacterShiftAndBorderArray::match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bbcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < size_t > fba = tree::properties::BorderArrayNaive::ba ( pattern );
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+
+ match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, 0, subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1 );
+ return occ;
+}
+
+template < class SymbolType, class RankType >
+void DeadZoneUsingBadCharacterShiftAndBorderArray::match_rec ( std::set < unsigned > & occ, const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedPattern < SymbolType, RankType > & pattern, std::vector < size_t > & fba, std::map < std::ranked_symbol < SymbolType, RankType >, size_t > & bbcs, std::vector < int > & subjectSubtreeJumpTable, int low, int high ) {
+ if ( low >= high ) return;
+
+ int i = ( low + high ) / 2;
+
+ // index to the pattern
+ unsigned j = 0;
+
+ // offset to the subject
+ unsigned offset = i;
+
+ while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] )
+ // match of symbol
+ offset++;
+ else if ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) )
+ // match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ else
+ break;
+
+ j++;
+ }
+
+ // match was found
+ if ( j >= pattern.getContent ( ).size ( ) ) occ.insert ( i );
+
+ match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, low, i - bbcs[subject.getContent ( )[i]] + 1 );
+ match_rec ( occ, subject, pattern, fba, bbcs, subjectSubtreeJumpTable, i + j - fba[j], high );
+}
+
} /* namespace exact */
} /* namespace arbology */
diff --git a/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.cpp b/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.cpp
index 135aaede2d..8535cf8561 100644
--- a/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.cpp
+++ b/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.cpp
@@ -6,20 +6,6 @@
*/
#include "ReversedBoyerMooreHorspool.h"
-#include <tree/properties/ReversedBadCharacterShiftTable.h>
-#include <tree/properties/SubtreeJumpTable.h>
-#include <tree/properties/ExactSubtreeRepeatsNaive.h>
-
-#include <tree/Tree.h>
-#include <tree/ranked/PrefixRankedTree.h>
-#include <tree/ranked/PrefixRankedBarTree.h>
-#include <tree/ranked/PrefixRankedPattern.h>
-#include <tree/ranked/PrefixRankedBarPattern.h>
-#include <tree/ranked/PrefixRankedNonlinearPattern.h>
-#include <tree/ranked/PrefixRankedBarNonlinearPattern.h>
-#include <alphabet/RankedSymbol.h>
-
-#include <map>
namespace arbology {
@@ -29,216 +15,11 @@ std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::Tree & sub
return dispatch ( subject.getData ( ), pattern.getData ( ) );
}
-std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarTree < > & pattern ) {
- return match ( subject, tree::PrefixRankedBarPattern < > ( pattern ) );
-}
-
auto ReversedBoyerMooreHorspoolPrefixRankedBarTreePrefixRankedBarTree = ReversedBoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarTree < > > ( ReversedBoyerMooreHorspool::match );
-
-std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
-
- // index to the subject
- int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
-
- // main loop of the algorithm over all possible indexes where the pattern can start
- while ( i >= 0 ) {
-
- // index to the pattern
- unsigned j = 0;
-
- // offset to the subject
- unsigned offset = i;
-
- while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
- // match of symbol
- offset = offset + 1;
- j = j + 1;
- } else if ( ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) ) && ( ! pattern.getBars ( ).count ( subject.getContent ( )[offset] ) ) ) {
- // match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- j = j + 2;
- } else {
- break;
- }
- }
-
- // match was found
- if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
-
- // shift heuristics
- i -= bcs[subject.getContent ( )[i]];
- }
-
- return occ;
-}
-
auto ReversedBoyerMooreHorspoolPrefixRankedBarTreePrefixRankedBarPattern = ReversedBoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarPattern < > > ( ReversedBoyerMooreHorspool::match );
-
-std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarNonlinearPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
- std::map < std::ranked_symbol < >, alphabet::Symbol > variablesSetting;
-
- tree::PrefixRankedBarTree < > repeats = tree::properties::ExactSubtreeRepeatsNaive::repeats ( subject );
-
- // index to the subject
- int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
-
- // main loop of the algorithm over all possible indexes where the pattern can start
- while ( i >= 0 ) {
- // clear the current state of variable to subtree repeat
- variablesSetting.clear();
-
- // index to the pattern
- unsigned j = 0;
-
- // offset to the subject
- unsigned offset = i;
-
- while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
- // match of symbol
- offset = offset + 1;
- j = j + 1;
- } else if ( ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) || pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[j] ) ) && ( ! pattern.getBars ( ).count ( subject.getContent ( )[offset] ) ) ) {
- // check nonlinear variable
- if ( pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j ] ) ) {
- auto setting = variablesSetting.find ( pattern.getContent ( )[ j ] );
-
- if ( setting != variablesSetting.end ( ) && repeats.getContent ( )[ offset ].getSymbol ( ) != setting->second )
- break;
-
- variablesSetting.insert ( std::make_pair ( pattern.getContent ( )[ j ], repeats.getContent( )[ offset ].getSymbol ( ) ) );
- }
-
- // match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- j = j + 2;
- } else {
- break;
- }
- }
-
- // match was found
- if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
-
- // shift heuristics
- i -= bcs[subject.getContent ( )[i]];
- }
-
- return occ;
-}
-
auto ReversedBoyerMooreHorspoolPrefixRankedBarTreePrefixRankedBarNonlinearPattern = ReversedBoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedBarTree < >, tree::PrefixRankedBarNonlinearPattern < > > ( ReversedBoyerMooreHorspool::match );
-
-std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedTree < > & pattern ) {
- return match ( subject, tree::PrefixRankedPattern < > ( pattern ) );
-}
-
auto ReversedBoyerMooreHorspoolPrefixRankedBarTreePrefixRankedTree = ReversedBoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedTree < >, tree::PrefixRankedTree < > > ( ReversedBoyerMooreHorspool::match );
-
-std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
-
- // index to the subject
- int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
-
- // main loop of the algorithm over all possible indexes where the pattern can start
- while ( i >= 0 ) {
-
- // index to the pattern
- unsigned j = 0;
-
- // offset to the subject
- unsigned offset = i;
-
- while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] )
- // match of symbol
- offset = offset + 1;
- else if ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) )
- // match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- else
- break;
-
- j = j + 1;
- }
-
- // match was found
- if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
-
- // shift heristics
- i -= bcs[subject.getContent ( )[i]];
- }
-
- return occ;
-}
-
auto ReversedBoyerMooreHorspoolPrefixRankedBarTreePrefixRankedPattern = ReversedBoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedTree < >, tree::PrefixRankedPattern < > > ( ReversedBoyerMooreHorspool::match );
-
-std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedNonlinearPattern < > & pattern ) {
- std::set < unsigned > occ;
- std::map < std::ranked_symbol < >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
- std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
- std::map < std::ranked_symbol < >, alphabet::Symbol > variablesSetting;
-
- tree::PrefixRankedTree < > repeats = tree::properties::ExactSubtreeRepeatsNaive::repeats ( subject );
-
- // index to the subject
- int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
-
- // main loop of the algorithm over all possible indexes where the pattern can start
- while ( i >= 0 ) {
- // clear the current state of variable to subtree repeat
- variablesSetting.clear();
-
- // index to the pattern
- unsigned j = 0;
-
- // offset to the subject
- unsigned offset = i;
-
- while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
- if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] )
- // match of symbol
- offset = offset + 1;
- else if ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) || pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j ] ) ) {
- // check nonlinear variable
- if ( pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j ] ) ) {
- auto setting = variablesSetting.find ( pattern.getContent ( )[ j ] );
-
- if ( setting != variablesSetting.end ( ) && repeats.getContent ( )[ offset ].getSymbol ( ) != setting->second )
- break;
-
- variablesSetting.insert ( std::make_pair ( pattern.getContent ( )[ j ], repeats.getContent( )[ offset ].getSymbol ( ) ) );
- }
-
- // match of variable with subtree
- offset = subjectSubtreeJumpTable[offset];
- } else
- break;
-
- j = j + 1;
- }
-
- // match was found
- if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
-
- // shift heristics
- i -= bcs[subject.getContent ( )[i]];
- }
-
- return occ;
-}
-
auto ReversedBoyerMooreHorspoolPrefixRankedBarTreePrefixRankedNonlinearPattern = ReversedBoyerMooreHorspool::RegistratorWrapper < std::set < unsigned >, tree::PrefixRankedTree < >, tree::PrefixRankedNonlinearPattern < > > ( ReversedBoyerMooreHorspool::match );
} /* namespace exact */
diff --git a/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.h b/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.h
index 2d2f0d3c6e..35c3492793 100644
--- a/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.h
+++ b/alib2algo/src/arbology/exact/ReversedBoyerMooreHorspool.h
@@ -12,6 +12,21 @@
#include <core/multipleDispatch.hpp>
#include <tree/TreeFeatures.h>
+#include <tree/properties/ReversedBadCharacterShiftTable.h>
+#include <tree/properties/SubtreeJumpTable.h>
+#include <tree/properties/ExactSubtreeRepeatsNaive.h>
+
+#include <tree/Tree.h>
+#include <tree/ranked/PrefixRankedTree.h>
+#include <tree/ranked/PrefixRankedBarTree.h>
+#include <tree/ranked/PrefixRankedPattern.h>
+#include <tree/ranked/PrefixRankedBarPattern.h>
+#include <tree/ranked/PrefixRankedNonlinearPattern.h>
+#include <tree/ranked/PrefixRankedBarNonlinearPattern.h>
+#include <alphabet/RankedSymbol.h>
+
+#include <map>
+
namespace arbology {
namespace exact {
@@ -28,15 +43,227 @@ public:
*/
static std::set < unsigned > match ( const tree::Tree & subject, const tree::Tree & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarTree < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarPattern < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedBarTree < > & subject, const tree::PrefixRankedBarNonlinearPattern < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedTree < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedPattern < > & pattern );
- static std::set < unsigned > match ( const tree::PrefixRankedTree < > & subject, const tree::PrefixRankedNonlinearPattern < > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarTree < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarNonlinearPattern < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedTree < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedPattern < SymbolType, RankType > & pattern );
+ template < class SymbolType, class RankType >
+ static std::set < unsigned > match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedNonlinearPattern < SymbolType, RankType > & pattern );
};
+template < class SymbolType, class RankType >
+std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarTree < SymbolType, RankType > & pattern ) {
+ return match ( subject, tree::PrefixRankedBarPattern < SymbolType, RankType > ( pattern ) );
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+
+ // index to the subject
+ int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
+
+ // main loop of the algorithm over all possible indexes where the pattern can start
+ while ( i >= 0 ) {
+
+ // index to the pattern
+ unsigned j = 0;
+
+ // offset to the subject
+ unsigned offset = i;
+
+ while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
+ // match of symbol
+ offset = offset + 1;
+ j = j + 1;
+ } else if ( ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) ) && ( ! pattern.getBars ( ).count ( subject.getContent ( )[offset] ) ) ) {
+ // match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ j = j + 2;
+ } else {
+ break;
+ }
+ }
+
+ // match was found
+ if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
+
+ // shift heuristics
+ i -= bcs[subject.getContent ( )[i]];
+ }
+
+ return occ;
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedBarTree < SymbolType, RankType > & subject, const tree::PrefixRankedBarNonlinearPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+ std::map < std::ranked_symbol < SymbolType, RankType >, SymbolType > variablesSetting;
+
+ tree::PrefixRankedBarTree < > repeats = tree::properties::ExactSubtreeRepeatsNaive::repeats ( subject );
+
+ // index to the subject
+ int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
+
+ // main loop of the algorithm over all possible indexes where the pattern can start
+ while ( i >= 0 ) {
+ // clear the current state of variable to subtree repeat
+ variablesSetting.clear();
+
+ // index to the pattern
+ unsigned j = 0;
+
+ // offset to the subject
+ unsigned offset = i;
+
+ while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] ) {
+ // match of symbol
+ offset = offset + 1;
+ j = j + 1;
+ } else if ( ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) || pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[j] ) ) && ( ! pattern.getBars ( ).count ( subject.getContent ( )[offset] ) ) ) {
+ // check nonlinear variable
+ if ( pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j ] ) ) {
+ auto setting = variablesSetting.find ( pattern.getContent ( )[ j ] );
+
+ if ( setting != variablesSetting.end ( ) && repeats.getContent ( )[ offset ].getSymbol ( ) != setting->second )
+ break;
+
+ variablesSetting.insert ( std::make_pair ( pattern.getContent ( )[ j ], repeats.getContent( )[ offset ].getSymbol ( ) ) );
+ }
+
+ // match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ j = j + 2;
+ } else {
+ break;
+ }
+ }
+
+ // match was found
+ if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
+
+ // shift heuristics
+ i -= bcs[subject.getContent ( )[i]];
+ }
+
+ return occ;
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedTree < SymbolType, RankType > & pattern ) {
+ return match ( subject, tree::PrefixRankedPattern < SymbolType, RankType > ( pattern ) );
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+
+ // index to the subject
+ int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
+
+ // main loop of the algorithm over all possible indexes where the pattern can start
+ while ( i >= 0 ) {
+
+ // index to the pattern
+ unsigned j = 0;
+
+ // offset to the subject
+ unsigned offset = i;
+
+ while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] )
+ // match of symbol
+ offset = offset + 1;
+ else if ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) )
+ // match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ else
+ break;
+
+ j = j + 1;
+ }
+
+ // match was found
+ if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
+
+ // shift heristics
+ i -= bcs[subject.getContent ( )[i]];
+ }
+
+ return occ;
+}
+
+template < class SymbolType, class RankType >
+std::set < unsigned > ReversedBoyerMooreHorspool::match ( const tree::PrefixRankedTree < SymbolType, RankType > & subject, const tree::PrefixRankedNonlinearPattern < SymbolType, RankType > & pattern ) {
+ std::set < unsigned > occ;
+ std::map < std::ranked_symbol < SymbolType, RankType >, size_t > bcs = tree::properties::ReversedBadCharacterShiftTable::bcs ( pattern ); // NOTE: the subjects alphabet must be a subset or equal to the pattern
+ std::vector < int > subjectSubtreeJumpTable = tree::properties::SubtreeJumpTable::compute ( subject );
+ std::map < std::ranked_symbol < SymbolType, RankType >, SymbolType > variablesSetting;
+
+ tree::PrefixRankedTree < > repeats = tree::properties::ExactSubtreeRepeatsNaive::repeats ( subject );
+
+ // index to the subject
+ int i = ( int ) subject.getContent ( ).size ( ) - pattern.getContent ( ).size ( ) + 1;
+
+ // main loop of the algorithm over all possible indexes where the pattern can start
+ while ( i >= 0 ) {
+ // clear the current state of variable to subtree repeat
+ variablesSetting.clear();
+
+ // index to the pattern
+ unsigned j = 0;
+
+ // offset to the subject
+ unsigned offset = i;
+
+ while ( ( j < pattern.getContent ( ).size ( ) ) && ( offset < subject.getContent ( ).size ( ) ) ) {
+ if ( subject.getContent ( )[offset] == pattern.getContent ( )[j] )
+ // match of symbol
+ offset = offset + 1;
+ else if ( pattern.getContent ( )[j] == pattern.getSubtreeWildcard ( ) || pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j ] ) ) {
+ // check nonlinear variable
+ if ( pattern.getNonlinearVariables ( ).count ( pattern.getContent ( )[ j ] ) ) {
+ auto setting = variablesSetting.find ( pattern.getContent ( )[ j ] );
+
+ if ( setting != variablesSetting.end ( ) && repeats.getContent ( )[ offset ].getSymbol ( ) != setting->second )
+ break;
+
+ variablesSetting.insert ( std::make_pair ( pattern.getContent ( )[ j ], repeats.getContent( )[ offset ].getSymbol ( ) ) );
+ }
+
+ // match of variable with subtree
+ offset = subjectSubtreeJumpTable[offset];
+ } else
+ break;
+
+ j = j + 1;
+ }
+
+ // match was found
+ if ( j == pattern.getContent ( ).size ( ) ) occ.insert ( i );
+
+ // shift heristics
+ i -= bcs[subject.getContent ( )[i]];
+ }
+
+ return occ;
+}
+
} /* namespace exact */
} /* namespace arbology */
diff --git a/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.cpp b/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.cpp
index 51ff2af299..9c14e08469 100644
--- a/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.cpp
+++ b/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.cpp
@@ -6,6 +6,13 @@
*/
#include "ExactSubtreeRepeatsNaive.h"
+#include "SubtreeJumpTable.h"
+
+#include <tree/ranked/RankedTree.h>
+#include <tree/ranked/PrefixRankedTree.h>
+#include <tree/ranked/PrefixRankedBarTree.h>
+#include <tree/Tree.h>
+#include <global/GlobalData.h>
namespace tree {
@@ -15,8 +22,94 @@ tree::Tree ExactSubtreeRepeatsNaive::repeats ( const tree::Tree & tree ) {
return dispatch ( tree.getData ( ) );
}
+std::tree < std::ranked_symbol < > > ExactSubtreeRepeatsNaive::repeats ( const std::tree < std::ranked_symbol < > > & node, std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > & data, int & minId ) {
+ std::vector < std::tree < std::ranked_symbol < > > > children;
+ std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > > childRepeatsKey ( node.getData ( ), std::vector < std::ranked_symbol < > > ( ) );
+
+ for ( const std::tree < std::ranked_symbol < > > & child : node.getChildren() ) {
+ children.push_back ( repeats ( child, data, minId ) );
+ childRepeatsKey.second.push_back ( children.back ( ).getData ( ) );
+ }
+
+ int & uniqueRepeatId = data[childRepeatsKey];
+
+ if ( uniqueRepeatId == 0 ) uniqueRepeatId = minId++;
+
+ return std::tree < std::ranked_symbol < > > ( std::ranked_symbol < > ( alphabet::symbolFrom ( uniqueRepeatId ), node.getData ( ).getRank ( ) ), std::move ( children ) );
+}
+
+tree::RankedTree < > ExactSubtreeRepeatsNaive::repeats ( const tree::RankedTree < > & tree ) {
+ int minId = 1;
+ std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > data;
+
+ return tree::RankedTree < > ( repeats ( tree.getContent ( ), data, minId ) );
+}
+
auto ExactRepeatsNaiveRankedTree = ExactSubtreeRepeatsNaive::RegistratorWrapper < tree::RankedTree < >, tree::RankedTree < > > ( ExactSubtreeRepeatsNaive::repeats );
+
+std::ranked_symbol < > ExactSubtreeRepeatsNaive::repeatsPrefixRanked ( const std::vector < std::ranked_symbol < > > & symbols, std::vector < std::ranked_symbol < > > & res, std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > & data, int & minId, int & index ) {
+ int begin = index;
+ std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > > childRepeatsKey ( symbols[begin], std::vector < std::ranked_symbol < > > ( ) );
+
+ res.push_back ( std::ranked_symbol < > ( alphabet::symbolFrom ( 0 ), symbols[begin].getRank ( ) ) );
+
+ index++;
+
+ for ( unsigned i = 0; i < ( unsigned ) symbols[begin].getRank ( ); ++i )
+ childRepeatsKey.second.push_back ( repeatsPrefixRanked ( symbols, res, data, minId, index ) );
+
+ int & uniqueRepeatId = data[childRepeatsKey];
+
+ if ( uniqueRepeatId == 0 ) uniqueRepeatId = minId++;
+
+ res[begin] = std::ranked_symbol < > ( alphabet::symbolFrom ( uniqueRepeatId ), symbols[begin].getRank ( ) );
+ return res[begin];
+}
+
+tree::PrefixRankedTree < > ExactSubtreeRepeatsNaive::repeats ( const tree::PrefixRankedTree < > & tree ) {
+ int minId = 1;
+ int index = 0;
+ std::vector < std::ranked_symbol < > > res;
+ std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > data;
+
+ repeatsPrefixRanked ( tree.getContent ( ), res, data, minId, index );
+ return tree::PrefixRankedTree < > ( res );
+}
+
auto ExactRepeatsNaivePrefixRankedTree = ExactSubtreeRepeatsNaive::RegistratorWrapper < tree::PrefixRankedTree < >, tree::PrefixRankedTree < > > ( ExactSubtreeRepeatsNaive::repeats );
+
+std::ranked_symbol < > ExactSubtreeRepeatsNaive::repeatsPrefixRankedBar ( const std::vector < std::ranked_symbol < > > & symbols, std::vector < std::ranked_symbol < > > & res, std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > & data, int & minId, int & index ) {
+ int begin = index;
+ std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > > childRepeatsKey ( symbols[begin], std::vector < std::ranked_symbol < > > ( ) );
+
+ res.push_back ( std::ranked_symbol < > ( alphabet::symbolFrom ( 0 ), symbols[begin].getRank ( ) ) );
+
+ index++;
+
+ for ( unsigned i = 0; i < ( unsigned ) symbols[begin].getRank ( ); ++i )
+ childRepeatsKey.second.push_back ( repeatsPrefixRankedBar ( symbols, res, data, minId, index ) );
+
+ int & uniqueRepeatId = data[childRepeatsKey];
+
+ if ( uniqueRepeatId == 0 ) uniqueRepeatId = minId++;
+
+ res[begin] = std::ranked_symbol < > ( alphabet::symbolFrom ( uniqueRepeatId ), symbols[begin].getRank ( ) );
+ res.push_back ( symbols[index] );
+ index++;
+
+ return res[begin];
+}
+
+tree::PrefixRankedBarTree < > ExactSubtreeRepeatsNaive::repeats ( const tree::PrefixRankedBarTree < > & tree ) {
+ int minId = 1;
+ int index = 0;
+ std::vector < std::ranked_symbol < > > res;
+ std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > data;
+
+ repeatsPrefixRankedBar ( tree.getContent ( ), res, data, minId, index );
+ return tree::PrefixRankedBarTree < > ( tree.getBars ( ), res );
+}
+
auto ExactRepeatsNaivePrefixRankedBarTree = ExactSubtreeRepeatsNaive::RegistratorWrapper < tree::PrefixRankedBarTree < >, tree::PrefixRankedBarTree < > > ( ExactSubtreeRepeatsNaive::repeats );
} /* namespace properties */
diff --git a/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.h b/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.h
index 3f81f6e3da..2378926916 100644
--- a/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.h
+++ b/alib2algo/src/tree/properties/ExactSubtreeRepeatsNaive.h
@@ -17,14 +17,6 @@
#include <tree>
#include <alphabet/RankedSymbol.h>
-#include "SubtreeJumpTable.h"
-
-#include <tree/ranked/RankedTree.h>
-#include <tree/ranked/PrefixRankedTree.h>
-#include <tree/ranked/PrefixRankedBarTree.h>
-#include <tree/Tree.h>
-#include <global/GlobalData.h>
-
namespace tree {
namespace properties {
@@ -33,14 +25,9 @@ namespace properties {
* Simple computation of subtree repeats
*/
class ExactSubtreeRepeatsNaive : public std::SingleDispatch < ExactSubtreeRepeatsNaive, tree::Tree, const tree::TreeBase & > {
- template < class SymbolType, class RankType >
- static std::tree < std::ranked_symbol < SymbolType, RankType > > repeats ( const std::tree < std::ranked_symbol < SymbolType, RankType > > & node, std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > & data, int & minId );
-
- template < class SymbolType, class RankType >
- static std::ranked_symbol < SymbolType, RankType > repeatsPrefixRanked ( const std::vector < std::ranked_symbol < SymbolType, RankType > > & symbols, std::vector < std::ranked_symbol < SymbolType, RankType > > & res, std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > & data, int & minId, int & index );
-
- template < class SymbolType, class RankType >
- static std::ranked_symbol < SymbolType, RankType > repeatsPrefixRankedBar ( const std::vector < std::ranked_symbol < SymbolType, RankType > > & symbols, std::vector < std::ranked_symbol < SymbolType, RankType > > & res, std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > & data, int & minId, int & index );
+ static std::tree < std::ranked_symbol < > > repeats ( const std::tree < std::ranked_symbol < > > & node, std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > & data, int & minId );
+ static std::ranked_symbol < > repeatsPrefixRanked ( const std::vector < std::ranked_symbol < > > & symbols, std::vector < std::ranked_symbol < > > & res, std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > & data, int & minId, int & index );
+ static std::ranked_symbol < > repeatsPrefixRankedBar ( const std::vector < std::ranked_symbol < > > & symbols, std::vector < std::ranked_symbol < > > & res, std::map < std::pair < std::ranked_symbol < >, std::vector < std::ranked_symbol < > > >, int > & data, int & minId, int & index );
public:
/**
@@ -53,105 +40,12 @@ public:
* Compute a same shaped tree with nodes containing unique subtree ids.
* @return Tree of repeats
*/
- template < class SymbolType, class RankType >
- static tree::RankedTree < SymbolType, RankType > repeats ( const tree::RankedTree < SymbolType, RankType > & tree );
- template < class SymbolType, class RankType >
- static tree::PrefixRankedTree < SymbolType, RankType > repeats ( const tree::PrefixRankedTree < SymbolType, RankType > & tree );
- template < class SymbolType, class RankType >
- static tree::PrefixRankedBarTree < SymbolType, RankType > repeats ( const tree::PrefixRankedBarTree < SymbolType, RankType > & tree );
+ static tree::RankedTree < > repeats ( const tree::RankedTree < > & tree );
+ static tree::PrefixRankedTree < > repeats ( const tree::PrefixRankedTree < > & tree );
+ static tree::PrefixRankedBarTree < > repeats ( const tree::PrefixRankedBarTree < > & tree );
};
-template < class SymbolType, class RankType >
-std::tree < std::ranked_symbol < SymbolType, RankType > > ExactSubtreeRepeatsNaive::repeats ( const std::tree < std::ranked_symbol < SymbolType, RankType > > & node, std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > & data, int & minId ) {
- std::vector < std::tree < std::ranked_symbol < SymbolType, RankType > > > children;
- std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > > childRepeatsKey ( node.getData ( ), std::vector < std::ranked_symbol < SymbolType, RankType > > ( ) );
-
- for ( const std::tree < std::ranked_symbol < SymbolType, RankType > > & child : node.getChildren() ) {
- children.push_back ( repeats ( child, data, minId ) );
- childRepeatsKey.second.push_back ( children.back ( ).getData ( ) );
- }
-
- int & uniqueRepeatId = data[childRepeatsKey];
-
- if ( uniqueRepeatId == 0 ) uniqueRepeatId = minId++;
-
- return std::tree < std::ranked_symbol < SymbolType, RankType > > ( std::ranked_symbol < SymbolType, RankType > ( alphabet::symbolFrom ( uniqueRepeatId ), node.getData ( ).getRank ( ) ), std::move ( children ) );
-}
-
-template < class SymbolType, class RankType >
-tree::RankedTree < SymbolType, RankType > ExactSubtreeRepeatsNaive::repeats ( const tree::RankedTree < SymbolType, RankType > & tree ) {
- int minId = 1;
- std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > data;
-
- return tree::RankedTree < SymbolType, RankType > ( repeats ( tree.getContent ( ), data, minId ) );
-}
-
-template < class SymbolType, class RankType >
-std::ranked_symbol < SymbolType, RankType > ExactSubtreeRepeatsNaive::repeatsPrefixRanked ( const std::vector < std::ranked_symbol < SymbolType, RankType > > & symbols, std::vector < std::ranked_symbol < SymbolType, RankType > > & res, std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > & data, int & minId, int & index ) {
- int begin = index;
- std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > > childRepeatsKey ( symbols[begin], std::vector < std::ranked_symbol < SymbolType, RankType > > ( ) );
-
- res.push_back ( std::ranked_symbol < SymbolType, RankType > ( alphabet::symbolFrom ( 0 ), symbols[begin].getRank ( ) ) );
-
- index++;
-
- for ( unsigned i = 0; i < ( unsigned ) symbols[begin].getRank ( ); ++i )
- childRepeatsKey.second.push_back ( repeatsPrefixRanked ( symbols, res, data, minId, index ) );
-
- int & uniqueRepeatId = data[childRepeatsKey];
-
- if ( uniqueRepeatId == 0 ) uniqueRepeatId = minId++;
-
- res[begin] = std::ranked_symbol < SymbolType, RankType > ( alphabet::symbolFrom ( uniqueRepeatId ), symbols[begin].getRank ( ) );
- return res[begin];
-}
-
-template < class SymbolType, class RankType >
-tree::PrefixRankedTree < SymbolType, RankType > ExactSubtreeRepeatsNaive::repeats ( const tree::PrefixRankedTree < SymbolType, RankType > & tree ) {
- int minId = 1;
- int index = 0;
- std::vector < std::ranked_symbol < SymbolType, RankType > > res;
- std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > data;
-
- repeatsPrefixRanked ( tree.getContent ( ), res, data, minId, index );
- return tree::PrefixRankedTree < SymbolType, RankType > ( res );
-}
-
-template < class SymbolType, class RankType >
-std::ranked_symbol < SymbolType, RankType > ExactSubtreeRepeatsNaive::repeatsPrefixRankedBar ( const std::vector < std::ranked_symbol < SymbolType, RankType > > & symbols, std::vector < std::ranked_symbol < SymbolType, RankType > > & res, std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > & data, int & minId, int & index ) {
- int begin = index;
- std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > > childRepeatsKey ( symbols[begin], std::vector < std::ranked_symbol < SymbolType, RankType > > ( ) );
-
- res.push_back ( std::ranked_symbol < SymbolType, RankType > ( alphabet::symbolFrom ( 0 ), symbols[begin].getRank ( ) ) );
-
- index++;
-
- for ( unsigned i = 0; i < ( unsigned ) symbols[begin].getRank ( ); ++i )
- childRepeatsKey.second.push_back ( repeatsPrefixRankedBar ( symbols, res, data, minId, index ) );
-
- int & uniqueRepeatId = data[childRepeatsKey];
-
- if ( uniqueRepeatId == 0 ) uniqueRepeatId = minId++;
-
- res[begin] = std::ranked_symbol < SymbolType, RankType > ( alphabet::symbolFrom ( uniqueRepeatId ), symbols[begin].getRank ( ) );
- res.push_back ( symbols[index] );
- index++;
-
- return res[begin];
-}
-
-template < class SymbolType, class RankType >
-tree::PrefixRankedBarTree < SymbolType, RankType > ExactSubtreeRepeatsNaive::repeats ( const tree::PrefixRankedBarTree < SymbolType, RankType > & tree ) {
- int minId = 1;
- int index = 0;
- std::vector < std::ranked_symbol < SymbolType, RankType > > res;
- std::map < std::pair < std::ranked_symbol < SymbolType, RankType >, std::vector < std::ranked_symbol < SymbolType, RankType > > >, int > data;
-
- repeatsPrefixRankedBar ( tree.getContent ( ), res, data, minId, index );
- return tree::PrefixRankedBarTree < SymbolType, RankType > ( tree.getBars ( ), res );
-}
-
} /* namespace properties */
} /* namespace tree */
--
GitLab