Created
July 18, 2013 20:54
-
-
Save vittorioromeo/6032999 to your computer and use it in GitHub Desktop.
Temp tokenizer
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
enum class TokenType | |
{ | |
Whitespace, // Sequence of blanks, newlines and tabs | |
Keyword, // Hardcoded words | |
Identifier, // String of letters + digits starting with a letter | |
Number, // String of digits | |
ParenthesisRoundOpen, | |
ParenthesisRoundClose, | |
ParenthesisSquareOpen, | |
ParenthesisSquareClose, | |
ParenthesisCurlyOpen, | |
ParenthesisCurlyClose, | |
ParenthesisAngleOpen, | |
ParenthesisAngleClose, | |
Semicolon, | |
Operator | |
}; | |
string getTokenStr(TokenType type) | |
{ | |
switch(type) | |
{ | |
case TokenType::Whitespace: return "TokenType::Whitespace"; | |
case TokenType::Keyword: return "TokenType::Keyword"; | |
case TokenType::Identifier: return "TokenType::Identifier"; | |
case TokenType::Number: return "TokenType::Number"; | |
case TokenType::ParenthesisRoundOpen: return "TokenType::ParenthesisRoundOpen"; | |
case TokenType::ParenthesisRoundClose: return "TokenType::ParenthesisRoundClose"; | |
case TokenType::ParenthesisSquareOpen: return "TokenType::ParenthesisSquareOpen"; | |
case TokenType::ParenthesisSquareClose: return "TokenType::ParenthesisSquareClose"; | |
case TokenType::ParenthesisCurlyOpen: return "TokenType::ParenthesisCurlyOpen"; | |
case TokenType::ParenthesisCurlyClose: return "TokenType::ParenthesisCurlyClose"; | |
case TokenType::ParenthesisAngleOpen: return "TokenType::ParenthesisAngleOpen"; | |
case TokenType::ParenthesisAngleClose: return "TokenType::ParenthesisAngleClose"; | |
case TokenType::Semicolon: return "TokenType::Semicolon"; | |
case TokenType::Operator: return "TokenType::Operator"; | |
} | |
} | |
int main() | |
{ | |
using MyTokenizer = Tokenizer<TokenType>; | |
MyTokenizer tokenizer; | |
vector<string> keywords { "if", "else", "for", "while", "struct", "class", "enum" }; | |
unordered_map<string, int> operators { {"*", 0}, {"/", 0}, {"%", 0}, {"+", 1}, {"-", 1}, {"<", 2}, {">", 2}, {"<=", 2}, {">=", 2}, {"==", 3}, {"!=", 3}, }; | |
tokenizer.createRuleWord | |
( | |
[](const string&) { return TokenType::Whitespace; }, | |
[](char mChar) { return isspace(mChar); }, | |
[](char mChar) { return isspace(mChar); } | |
); | |
tokenizer.createRuleWord | |
( | |
[=](const string& mToken) { return contains(keywords, mToken) ? TokenType::Keyword : TokenType::Identifier; }, | |
[](char mChar) { return isalpha(mChar); }, | |
[](char mChar) { return isalnum(mChar); } | |
); | |
tokenizer.createRuleWord | |
( | |
[](const string&) { return TokenType::Number; }, | |
[](char mChar) { return isdigit(mChar); }, | |
[](char mChar) { return isdigit(mChar); } | |
); | |
tokenizer.createRuleWord | |
( | |
[](const string&) { return TokenType::Operator; }, | |
[=](char mChar) { for(const auto& o : operators) if(o.first[0] == mChar) return true; return false; }, | |
[=](char mChar) { for(const auto& o : operators) if(contains(o.first, mChar)) return true; return false; } | |
); | |
tokenizer.createRuleChar(TokenType::ParenthesisRoundOpen, '('); | |
tokenizer.createRuleChar(TokenType::ParenthesisRoundClose, ')'); | |
tokenizer.createRuleChar(TokenType::ParenthesisSquareOpen, '['); | |
tokenizer.createRuleChar(TokenType::ParenthesisSquareClose, ']'); | |
tokenizer.createRuleChar(TokenType::ParenthesisCurlyOpen, '{'); | |
tokenizer.createRuleChar(TokenType::ParenthesisCurlyClose, '}'); | |
tokenizer.createRuleChar(TokenType::ParenthesisAngleOpen, '<'); | |
tokenizer.createRuleChar(TokenType::ParenthesisAngleClose, '>'); | |
tokenizer.createRuleChar(TokenType::Semicolon, ';'); | |
tokenizer.setSource("if(BananaPhone == Blah) { mmh(12345); } else { templateFailsToTokenize<TypeFFF>(hi); }"); | |
tokenizer.process(); | |
for(const auto& t : tokenizer.getTokens()) | |
{ | |
log("\t\t\t\t" + t.getStr(), getTokenStr(t.getType())); | |
} | |
return 0; | |
} |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#ifndef SSVU_TOKENIZER | |
#define SSVU_TOKENIZER | |
#include <string> | |
#include <vector> | |
#include <unordered_map> | |
#include <functional> | |
#include "SSVUtils/Utils/UtilsContainers.h" | |
#include "SSVUtils/Log/Log.h" | |
#include "SSVUtils/Global/Typedefs.h" | |
namespace ssvu | |
{ | |
template<typename T> class Token | |
{ | |
private: | |
T type; | |
std::string str; | |
public: | |
Token(T mType, const std::string& mStr) : type{mType}, str{mStr} { } | |
inline const T& getType() const { return type; } | |
inline const std::string& getStr() const { return str; } | |
}; | |
template<typename T> class Tokenizer | |
{ | |
public: | |
struct RuleWord | |
{ | |
std::function<T(const std::string&)> getType; | |
std::function<bool(char)> isValidStart, isValid; | |
RuleWord(std::function<T(const std::string&)> mGetType, std::function<bool(char)> mIsValidStart, std::function<bool(char)> mIsValid) : getType(mGetType), isValidStart(mIsValidStart), isValid(mIsValid) { } | |
}; | |
struct RuleChar | |
{ | |
T type; | |
char tokenChar; | |
RuleChar(T mType, char mTokenChar) : type{mType}, tokenChar{mTokenChar} { } | |
}; | |
private: | |
std::vector<RuleChar> ruleChars; | |
std::vector<RuleWord> ruleWords; | |
std::vector<Token<T>> tokens; | |
std::string source; | |
std::string::iterator itr, tokenStart; | |
bool finished{false}; | |
void advance() { ++itr; if(itr >= std::end(source)) finished = true; } | |
void emitToken(T mTokenType) { tokens.push_back({mTokenType, std::string(tokenStart, itr)}); } | |
public: | |
inline void setSource(const std::string& mSource) { source = mSource; } | |
inline const std::vector<Token<T>>& getTokens() const { return tokens; } | |
template<typename... TArgs> inline void createRuleChar(TArgs&&... mArgs) { ruleChars.emplace_back(std::forward<TArgs>(mArgs)...); } | |
template<typename... TArgs> inline void createRuleWord(TArgs&&... mArgs) { ruleWords.emplace_back(std::forward<TArgs>(mArgs)...); } | |
void process() | |
{ | |
itr = std::begin(source); | |
tokenStart = itr; | |
while(!finished) | |
{ | |
tokenStart = itr; | |
bool found{false}; | |
for(auto& r : ruleWords) | |
{ | |
if(r.isValidStart(*itr)) | |
{ | |
found = true; | |
while(r.isValid(*itr)) advance(); | |
emitToken(r.getType(std::string(tokenStart, itr))); | |
goto end; | |
} | |
} | |
for(auto& r : ruleChars) | |
{ | |
found = true; | |
if(r.tokenChar != *itr) continue; | |
advance(); emitToken(r.type); | |
goto end; | |
} | |
end: | |
if(!found) finished = true; | |
} | |
} | |
}; | |
} | |
#endif |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment