From 9a95b5d24b2b2f5111e1862875d4136964a59548 Mon Sep 17 00:00:00 2001 From: Adrian Kummerländer Date: Sat, 19 Oct 2013 20:11:56 +0200 Subject: Improvement: Replaced "priority" with TokenType * Made implementation more expressive by replacing the integer priority with an strictly typed enum called TokenType ** Made removal of character comparisons from tree construction and lexer possible * As a side effect distinct numbers had to be assigned to each token ** Operators of same priority do not have identical numbers anymore --- src/tree.cc | 10 ++++------ src/utils.cc | 53 ++++++++++++++++++++++++++++------------------------- src/utils.h | 13 ++++++++++++- 3 files changed, 44 insertions(+), 32 deletions(-) diff --git a/src/tree.cc b/src/tree.cc index 1dcfca6..0f5c2d8 100644 --- a/src/tree.cc +++ b/src/tree.cc @@ -111,21 +111,19 @@ Node* Tree::buildTree(std::string term) { std::vector tmpLexer; std::vector lexerOutput = lexer(term); - int8_t priority; - for ( auto termIter = lexerOutput.begin(); termIter != lexerOutput.end(); termIter++ ) { - std::string& currTerm = (*termIter); - priority = getPriority(currTerm[0]); + const std::string& currTerm = (*termIter); + const TokenType token = getTokenType(currTerm[0]); - if ( priority != -1 && (*termIter).size() == 1 ) { + if ( token != TokenType::VALUE_NUMBER && (*termIter).size() == 1 ) { if ( !operatorStack.empty() ) { OperatorNode* lastNode( static_cast(topNodeFrom(operatorStack)) ); - if ( getPriority(lastNode->getFunction()) < priority ) { + if ( token > getTokenType(lastNode->getFunction()) ) { operatorStack.push( this->addOperator(nullptr, currTerm[0]) ); diff --git a/src/utils.cc b/src/utils.cc index 7457ae2..112bc20 100644 --- a/src/utils.cc +++ b/src/utils.cc @@ -5,26 +5,26 @@ namespace SimpleParser { -int8_t getPriority(char tmp) { +TokenType getTokenType(char tmp) { switch ( tmp ) { case '-': - return 10; + return TokenType::OPERATOR_PLUS; case '+': - return 10; + return TokenType::OPERATOR_MINUS; case '/': - return 20; + return TokenType::OPERATOR_DIVIDE; case '*': - return 20; + return TokenType::OPERATOR_MULTIPLY; case '^': - return 30; + return TokenType::OPERATOR_POWER; case '(': - return 90; + return TokenType::PARENTHESES_OPEN; case ')': - return 90; + return TokenType::PARENTHESES_CLOSE; case ',': - return -1; + return TokenType::VALUE_NUMBER; default: - return -1; + return TokenType::VALUE_NUMBER; } } @@ -32,31 +32,33 @@ std::vector lexer(std::string term) { std::string tmp; std::string tmpNum; std::vector output; + TokenType token; + TokenType lastToken; - int8_t priority = 0; - int8_t lastPriority = 0; - uint32_t level = 0; + uint32_t level = 0; for ( auto termIter = term.begin(); termIter != term.end(); termIter++ ) { - priority = getPriority(*termIter); + token = getTokenType(*termIter); - if ( priority == -1 || ( termIter == term.begin() && - priority == 10 ) ) { + if ( token == TokenType::VALUE_NUMBER || + ( token == TokenType::OPERATOR_MINUS && + termIter == term.begin() ) ) { if ( level > 0 ) { tmp += *termIter; } else { tmpNum += *termIter; } } else { - if ( lastPriority == -1 && level == 0 ) { + if ( lastToken == TokenType::VALUE_NUMBER && + level == 0 ) { output.push_back(tmpNum); tmpNum.clear(); } - switch ( *termIter ) { - case '(': { + switch ( token ) { + case TokenType::PARENTHESES_OPEN: { if ( level > 0 ) { tmp += *termIter; } @@ -65,7 +67,7 @@ std::vector lexer(std::string term) { break; } - case ')': { + case TokenType::PARENTHESES_CLOSE: { level--; if ( level == 0 ) { @@ -92,20 +94,21 @@ std::vector lexer(std::string term) { } } - lastPriority = priority; + lastToken = token; } - if ( lastPriority == -1 ) { + if ( lastToken == TokenType::VALUE_NUMBER ) { output.push_back(tmpNum); - } else if ( lastPriority != 90 ) { + } else if ( lastToken != TokenType::PARENTHESES_CLOSE ) { throw operator_exception(); } - if (level != 0) { + if ( level != 0 ) { throw parenthese_exception(); } - if ( lastPriority == 90 && output.size() == 1 ) { + if ( lastToken == TokenType::PARENTHESES_CLOSE && + output.size() == 1 ) { output = lexer(output[0]); } diff --git a/src/utils.h b/src/utils.h index f71c835..6e43605 100644 --- a/src/utils.h +++ b/src/utils.h @@ -8,7 +8,18 @@ namespace SimpleParser { -int8_t getPriority(char); +enum class TokenType : int8_t { + OPERATOR_PLUS = 10, + OPERATOR_MINUS = 11, + OPERATOR_DIVIDE = 20, + OPERATOR_MULTIPLY = 21, + OPERATOR_POWER = 30, + PARENTHESES_OPEN = 90, + PARENTHESES_CLOSE = 91, + VALUE_NUMBER = -1, +}; + +TokenType getTokenType(char); std::vector lexer(std::string); } -- cgit v1.2.3