CFG Reader(LR(0))
A. Tenth Edition
This is tenth edition of my CFG Reader which stands for Context-Free-Grammar Reader. In this edition,
I implement the LR(0) table by constructing a DFA.
The original grammar from teacher:
M -> program i ; Dl B
Dl -> Dv Ml
Ml -> Ml Mo | e
Mo -> module i ( Vl ) Dv B
Dv -> variables Vl | e
Vl -> Vl V | V
V -> Il : T ;
T -> integer Ad | char Ad
Il -> i | Il , i
L -> i Ar
Ad -> e | array [ n ]
B -> begin Sl end ;
Sl -> S | S Sl
S -> L := E ; | if C then S else S | loop Sl end ; | exit ; | i ( Lp ) ; | B |
read Ln ; | write Lo ; | e ;
Lp -> e | Ln
Ln -> L { , L }
Lo -> Lr { , Lr }
Lr -> i Ar | n | c
Ar -> e | [ E ]
E -> F { Oa F }
Oa -> + | -
F -> R { Om R }
Om -> * | /
R -> L | n | ( E ) | c
C -> E Or E
Or -> = | < | > | <= | >= | !=
The revised grammar by my class
Grammar's optimize method:
M ==> program i ; Dl B
Dl ==> Dv Ml
B ==> begin Sl end ;
Dv ==> variables Vl | e
Ml ==> e Ml0
Mo ==> module i ( Vl ) Dv B
Vl ==> V Vl0
V ==> Il : T ;
Il ==> i Il0
T ==> integer Ad | char Ad
Ad ==> e | array [ n ]
L ==> i Ar
Ar ==> e | [ E ]
Sl ==> S Sl0
S ==> i S0 | if C then S else S | loop Sl end ; | exit ; | begin Sl end ; |
read Ln ; | write
Lo ; | e ;
E ==> F M0
C ==> F M0 Or E
Lp ==> e | Ln
Ln ==> i Ar M1
Lo ==> Lr M2
Lr ==> i Ar | n | c
F ==> R M3
Oa ==> + | -
R ==> i Ar | n | ( E ) | c
Om ==> * | /
Or ==> = | < | > | <= | >= | !=
M0 ==> + F M0 | e | - F M0
M1 ==> , L M1 | e
M2 ==> , Lr M2 | e
M3 ==> * R M3 | e | / R M3
Ml0 ==> module i ( Vl ) Dv B Ml0 | e
Vl0 ==> i Il0 : T ; Vl0 | e
Il0 ==> , i Il0 | e
Sl0 ==> e | Sl
S0 ==> Ar := E ; | ( Lp ) ;
START ==> M $
The First and Follow set of all non-terminals:
name: M isNull: false isTerminal: false first: [0]=program follow: [0]=$ name: Dl isNull: true isTerminal: false first: [0]=e,[1]=variables,[2]=module follow: [0]=begin name: B isNull: false isTerminal: false first: [0]=begin follow: [0]=$,[1]=module,[2]=begin name: Dv isNull: true isTerminal: false first: [0]=variables,[1]=e follow: [0]=module,[1]=begin name: Ml isNull: true isTerminal: false first: [0]=e,[1]=module follow: [0]=begin name: Mo isNull: false isTerminal: false first: [0]=module follow: name: Vl isNull: false isTerminal: false first: [0]=i follow: [0]=module,[1]=begin,[2]=) name: V isNull: false isTerminal: false first: [0]=i follow: [0]=i,[1]=module,[2]=begin,[3]=) name: Il isNull: false isTerminal: false first: [0]=i follow: [0]=: name: T isNull: false isTerminal: false first: [0]=integer,[1]=char follow: [0]=; name: Ad isNull: true isTerminal: false first: [0]=e,[1]=array follow: [0]=; name: L isNull: false isTerminal: false first: [0]=i follow: [0]=,,[1]=;,[2]=) name: Ar isNull: true isTerminal: false first: [0]=e,[1]=[ follow: [0]=,,[1]=:=,[2]=;,[3]=),[4]=*,[5]=/,[6]=+,[7]=-,[8]=],[9]==,[10]=<,[11]=>,[12]=<=,[13]=> =,[14]=!=,[15]=then name: Sl isNull: false isTerminal: false first: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=; follow: [0]=end name: S isNull: false isTerminal: false first: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=; follow: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=;,[8]=end,[9]=else name: E isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=],[1]=),[2]=;,[3]=then name: C isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=then name: Lp isNull: true isTerminal: false first: [0]=e,[1]=i follow: [0]=) name: Ln isNull: false isTerminal: false first: [0]=i follow: [0]=;,[1]=) name: Lo isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=c follow: [0]=; name: Lr isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=c follow: [0]=,,[1]=; name: F isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=+,[1]=-,[2]=],[3]=),[4]=;,[5]==,[6]=<,[7]=>,[8]=<=,[9]=>=,[10]=!=,[11]=then name: Oa isNull: false isTerminal: false first: [0]=+,[1]=- follow: name: R isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=*,[1]=/,[2]=+,[3]=-,[4]=],[5]=),[6]=;,[7]==,[8]=<,[9]=>,[10]=<=,[11]=>=,[12]=!=,[13]= then name: Om isNull: false isTerminal: false first: [0]=*,[1]=/ follow: name: Or isNull: false isTerminal: false first: [0]==,[1]=<,[2]=>,[3]=<=,[4]=>=,[5]=!= follow: [0]=i,[1]=n,[2]=(,[3]=c name: M0 isNull: true isTerminal: false first: [0]=+,[1]=e,[2]=- follow: [0]=],[1]=),[2]=;,[3]==,[4]=<,[5]=>,[6]=<=,[7]=>=,[8]=!=,[9]=then name: M1 isNull: true isTerminal: false first: [0]=,,[1]=e follow: [0]=;,[1]=) name: M2 isNull: true isTerminal: false first: [0]=,,[1]=e follow: [0]=; name: M3 isNull: true isTerminal: false first: [0]=*,[1]=e,[2]=/ follow: [0]=+,[1]=-,[2]=],[3]=),[4]=;,[5]==,[6]=<,[7]=>,[8]=<=,[9]=>=,[10]=!=,[11]=then name: Ml0 isNull: true isTerminal: false first: [0]=module,[1]=e follow: [0]=begin name: Vl0 isNull: true isTerminal: false first: [0]=i,[1]=e follow: [0]=module,[1]=begin,[2]=) name: Il0 isNull: true isTerminal: false first: [0]=,,[1]=e follow: [0]=: name: Sl0 isNull: true isTerminal: false first: [0]=e,[1]=i,[2]=if,[3]=loop,[4]=exit,[5]=begin,[6]=read,[7]=write,[8]=; follow: [0]=end name: S0 isNull: false isTerminal: false first: [0]=[,[1]=:=,[2]=( follow: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=;,[8]=end,[9]=else name: START isNull: false isTerminal: false first: [0]=program follow: Press any key to continue
The idea is to follow to standard algorithm in text book. The debugging is a really hard job, considering the
error may arise from different part of program: grammar, scanner, parser or the source program itself!
E.Further improvement
F.File listing
1. CFGReader.h
2. CFGReader.cpp
3. Grammar.h
4. Grammar.cpp
5. parser.h
6. parser.cpp
7. scanner.h
8. scanner.cpp
9. errorno.h
10. errorNo.cpp
11. initialize.cpp
12. main.cpp (main)
file name: CFGReader.h
#include "Grammar.h"
class CFGReader
{
private:
char buffer[BufferLength];
FILE* stream;
void newRule(const char* str);
void readRule();
void addRule(const char* str);
int addToken(const char* str, bool isTerminal=true);
void printRule(int index);
public:
void readFromFile(const char* fileName);
void optimize();
void calculateLookAhead();
void print();
};
file name: CFGReader.cpp
#include <iostream>
#include "CFGReader.h"
using namespace std;
Grammar grammar;
const char* deliStr=" |\n";
//this would only remove the immediate left recursion
//and should I expand it to remove all kinds of left-recursion?
void CFGReader::readFromFile(const char* fileName)
{
if ((stream=fopen(fileName, "r"))==NULL)
{
cout<<"cannot open file "<<fileName<<endl;
}
else
{
readRule();
}
}
void CFGReader::readRule()
{
char* ptr=buffer;
char* hold;
int count=0;
while (!feof(stream))
{
count=0;
fgets(buffer, BufferLength-1, stream);
ptr=buffer;
while ((ptr=strtok(ptr, " \n"))!=NULL)
{
if (strcmp(ptr, "|")!=0)
{
//test the first two token to see if old rule continues
if (count==0)
{
hold=ptr;
}
if (count==1)
{
if (strcmp("->", ptr)==0)
{
/*
curIndex=addToken(hold, false);//the index of cur token
grammar[curIndex]->terminal=false;
newRule();
*/
newRule(hold);
}
else
{
addRule(hold);
addRule(ptr);
}
}
if (count>=2)//always add
{
addRule(ptr);
}
count++;
}
else
{
//this is an alternative production rule
newRule(NULL);
}
ptr=NULL;
}
}
}
void CFGReader::newRule(const char* str)
{
grammar.newRule(str);
}
void CFGReader::optimize()
{
grammar.optimize();
}
void CFGReader::calculateLookAhead()
{
grammar.calculateLookAhead();
}
void CFGReader::addRule(const char* str)
{
grammar.addRule(str);
}
int CFGReader::addToken(const char* str, bool isTerminal)
{
return grammar.addToken(str, isTerminal);
}
void CFGReader::printRule(int index)
{
grammar.printRule(index);
}
void CFGReader::print()
{
grammar.print();
//grammar.printToken(true);
grammar.buildTable(false);
grammar.printTable(false);
//grammar.printAllRules();
}
file name: Grammar.h
#ifndef GRAMMAR_H
#define GRAMMAR_H
#include <bitset>
#include <iostream>
using namespace std;
const int BufferLength=256;
const int MaxTokenCount=100;
const int MaxRHSCount=40;
const int MaxRuleCount=200;
const int MaxGrammarTokenLength=10;
const int MaxProduction=10;
const int TokenTypeCount=38;
const int MaxStateCount=200;
const int MaxItemCount=500;
//a macro to ease the job
//#define BitSet (bitset<MaxItemCount>)
struct GrammarToken
{
//bool isMeta;
bool terminal;
bool isNull;
char* name;
int production[MaxProduction];//pointer to the production it gives
int count;
int firstCount;
int followCount;
int follow[MaxTokenCount];
int first[MaxTokenCount];
};
struct Item
{
int varIndex;
int rulePos;
int dotPos;
};
bool operator == (Item& first, Item& second);
/*
Item& operator= (Item& first, Item& second)
{
first.dotPos=second.dotPos;
first.rulePos=second.rulePos;
first.varIndex=second.varIndex;
return first;
}
*/
class BitSet: public bitset<MaxItemCount>
{
private:
int current;
public:
BitSet& operator=(const BitSet& theSet);
int next();
BitSet();
void restart(){current=-1;}
};
class Grammar
{
//to allow Parser to access Grammar
friend class Parser;
private:
int LRTable[MaxStateCount][MaxTokenCount];
int terminalCount;
int nonTermCount;
GrammarToken* token[MaxTokenCount];
int tokenCount;
int curIndex;//the index of current token
int curPos;//the position at production rule
//MaxRuleCount>=MaxVariableCount!!!
//the last one in each rule is reserved for the length
int production[MaxRuleCount][MaxRHSCount+1];
int prodIndex;//pointing to current production, initialized to -1
int checkRecursion(int tIndex, int curToken);
void grammarError(int theVar, int theToken, int rule1, int rule2);
void addFirstIntoTable(int theVariable, int theFirst, int theRule);
void addFollowIntoTable(int theVariable, int theRule);
void addBeginEnd();
void addEpsilonForToken(int tIndex);
bool addIntoFirst(int target, int source);
bool addFirst(int target, int source);
void shiftRule(int ruleIndex, bool Left2Right, int offset);
void addAtEnd(int ruleIndex, int toAdd);
void addRuleForToken(int tIndex, int ruleIndex);
int copyRule(int source, int target);//return the position of -1
void replaceRule(int curToken, int ruleIndex);
int findMetaSymbol(int& begin, int& end);
void initialize();
void doReplaceMetaSymbol(int ruleIndex, int begin, int end);
void removeRuleFromToken(int tIndex, int ruleIndex);
void checkEpsilon(int ruleIndex);
void removeImmediate(int tIndex, int ruleIndex);
void doAddRule(int tIndex);
void commonLeftFactor();
int findCommonFactor(int tokenIndex, int ruleIndex);
void removeLeftRecursion();
void doCommonFactor(int tokenIndex, int ruleIndex, int index);
int forgeToken(int index);//create a new variable
void epsilonRule(int ruleIndex);
bool isRedundant(int tIndex);
void replaceMetaSymbol();
void calculateFirst();
void calculateFollow();
void calculateNull();
bool Null(int tIndex);
bool addFollow(int target, int source);
bool addIntoFollow(int target, int source);
bool addFollowIntoFollow(int target, int source);
void removeRedundant();
void removeToken(int tIndex);
void LRoptimize();
void LLoptimize();
void buildLLTable();
void buildLRTable();
int ruleLen(int ruleIndex);
void calculateProperty();
int shift2Mask(int input);
int reduce2Mask(int input);
bool* expandFinished;
int mask2State(int input);
bool isReduce(int input);
bool isShift(int input);
//items-related methods
void initializeDFA();
void uninitializeDFA();
void constructDFA();
//void addItems(int itemIndex, bitset<MaxStateCount>* theSet);
int addItem(Item& theItem);
int itemCount;
int stateCount;
Item**items;
//int addState(bitset
void addSonItem(int tIndex, BitSet& theSet);
int createState(const BitSet& theSet);
BitSet** sets;
int createItem(int theVar, int theNo, int position);
void addClosure(int itemIndex, BitSet& theSet);
void expandState(int stateIndex);
int findNextTokenOfItem(int itemIndex);
void doExpandState(int itemIndex, int tIndex, BitSet& theSet);
void add2Table(int stateIndex, int targetState, int tIndex);//this is shift
void add2Table(int stateIndex, int ruleIndex);//this is reduce
public:
Grammar();
void buildTable(bool forLLGrammar=true);
void optimize(bool forLLGrammar=true);
void printRule(int index);
void printRule(int tIndex, int rulePos);
void print();
void printTable(bool isLL=true);
void printAllRules();
void printToken(bool onlyVar=false);
int varCount();
int termCount();
void addRule(const char* str);
void newRule(const char* str);//this is an internal method, not suitable for outside
GrammarToken* operator[](int index);
int addToken(const char* str, bool isTerminal=true);
GrammarToken* createToken(const char* theName, bool isTerminal);
int tokenIndex(const char* str);
void calculateLookAhead();
};
#endif
file name: Grammar.cpp
#include <iostream>
#include <iomanip>
#include "errorNo.h"
#include "Grammar.h"
using namespace std;
/*
#define REDUCEMASK 0xd10000000
#define SHIFTMASK 0xd20000000
#define STATEMASK 0xd01111111
#define SHIFTREDUCEMASK 0xd30000000
*/
const int REDUCEMASK = 0xd10000000;
const int SHIFTMASK = 0xd20000000;
const int STATEMASK = 0xd01111111;
const int SHIFTREDUCEMASK = 0xd30000000;
const char* emptyStr="e";
const char* optionBegin="{";
const char* optionEnd="}";
const char* startStr="START";
const char* endStr="$";
int startSymbolIndex=-1;
int stackBottomIndex=-1;
int beginIndex=-1;
int endIndex=-1;
int emptyIndex=-1;
int epsilonIndex=-1;
int table[MaxTokenCount][MaxTokenCount];
char* terminalStr[TokenTypeCount]=
{
//GENERAL TYPE 5
"i", "n", "c", "COMMENT", "ERROR",
//THE FOLLOWING ARE SYMBOL TYPE 18
"(", ")", ";", "+", "-", "*",
"/", ":=", "<", ">", "=", "<=",
">=", "!=", "[", "]", ",",
":",
//THE FOLLOWING ARE RESERVED TYPE 15
"begin", "end", "program", "variables","integer", "array", "char",
"module", "if", "then", "else", "loop", "exit", "read", "write"
};
//these are "hash-table-like" tables
//this is exactly the opposite of below
int token2type[MaxTokenCount];
//this is to translate tokenType in scanner to token-index of grammar
int type2token[MaxTokenCount];
int matchTokens(const char* str)
{
for (int i=0; i<TokenTypeCount; i++)
{
if (strcmp(terminalStr[i], str)==0)
{
return i;
}
}
return -1;
}
void Grammar::calculateProperty()
{
terminalCount=nonTermCount=0;
for (int i=0; i<tokenCount; i++)
{
if (token[i]->terminal)
{
terminalCount++;
continue;
}
for (int j=0; j<token[i]->count; j++)
{
int len;
len=ruleLen(token[i]->production[j]);
production[token[i]->production[j]][MaxRHSCount]=len;
}
nonTermCount++;
}
}
void Grammar::addEpsilonForToken(int tIndex)
{
if (epsilonIndex==-1)
{
epsilonRule(++prodIndex);
addRuleForToken(tIndex, prodIndex);
}
else
{
addRuleForToken(tIndex, epsilonIndex);
}
}
void Grammar::printTable(bool isLL)
{
/*
cout<<" ";
for (int i=0; i<tokenCount; i++)
{
cout<<"| "<<i;
}
cout<<endl;
*/
if (isLL)
{
for (int r=0; r<tokenCount; r++)
{
if (token[r]->terminal)
{
continue;
}
cout<<token[r]->name<<":";
for (int c=0; c<tokenCount; c++)
{
if (table[r][c]!=-1)
{
cout<<token[c]->name<<"="<<table[r][c]<<",";
}
}
cout<<endl;
}
}
else
{
cout<<"the following is LR table<<\n";
for (int col=0; col<tokenCount; col++)
{
cout<<token[col]->name<<",";
}
cout<<endl;
for (int r=0; r<stateCount; r++)
{
for (int c=0; c<tokenCount; c++)
{
int result=LRTable[r][c];
if (isReduce(result))
{
cout<<"r";
cout<<mask2State(result);
}
else
{
if (isShift(result))
{
cout<<"s";
cout<<mask2State(result);
}
else
{
cout<<result;
}
}
cout<<",";
}
cout<<endl;
}
}
}
void Grammar::buildLRTable()
{
constructDFA();
}
//initialized at "initialize()"
void Grammar::buildLLTable()
{
int typeIndex;
for (int r=0; r<tokenCount; r++)
{
if (token[r]->terminal)
{
typeIndex=matchTokens(token[r]->name);
if (typeIndex!=-1)
{
type2token[typeIndex]=r;
token2type[r]=typeIndex;
}
}
//initialize
/*
for (int c=0; c<MaxTokenCount; c++)
{
table[r][c]=-1;
}
*/
}
for (int i=0; i<tokenCount; i++)
{
if (token[i]->terminal)
{
continue;
}
for (int j=0; j<token[i]->count; j++)
{
int k=0, theRule=token[i]->production[j];
int theToken=production[theRule][k];
while (theToken!=-1)
{
addFirstIntoTable(i, theToken, theRule);
if (!token[theToken]->isNull)
{
break;
}
k++;
theToken=production[theRule][k];
}
if (theToken==-1)
{
addFollowIntoTable(i, theRule);
}
}
}
}
void Grammar::buildTable(bool forLLGrammar)
{
if (forLLGrammar)
{
buildLLTable();
}
else
{
buildLRTable();
}
}
/*
nCount=tCount=0;
for (int r=0; r<MaxTokenCount; r++)
{
if (r<tokenCount)
{
if (token[r]->terminal)
{
typeIndex=matchTokens(token[r]->name);
if (typeIndex!=-1)
{
type2token[typeIndex]=r;
token2type[r]=typeIndex;
}
//I am making a invertable table!!!
tArray[r]=tCount++;
}
else
{
//it is an invertable table!!!
nArray[r]=nCount++;
}
}
for (int c=0; c<MaxTokenCount; c++)
{
table[r][c]=-1;
}
}
for (int i=0; i<tCount; i++)
{
for (int j=0; j<token[i]->count; j++)
{
int k=0, theRule=token[i]->production[j];
int theToken=production[theRule][k];
while (theToken!=-1)
{
addFirstIntoTable(theToken, j);
if (!token[theToken]->isNull)
{
break;
}
k++;
theToken=production[theRule][k];
}
if (theToken==-1)
{
addFollowIntoTable(j);
}
}
}
}
*/
void Grammar::grammarError(int theVar, int theToken, int rule1, int rule2)
{
cout<<"error! the conflict of grammar at ";
cout<<token[theVar]->name<<" for token of "
<<token[theToken]->name
<<" between rules of \n";
printRule(rule1);
cout<<" and ";
printRule(rule2);
cout<<endl;
}
void Grammar::addFollowIntoTable(int theVariable, int theRule)
{
int temp;
for (int i=0; i<token[theVariable]->followCount; i++)
{
temp=token[theVariable]->follow[i];
if (table[theVariable][temp]!=theRule)
{
if (table[theVariable][temp]!=-1)
{
grammarError(theVariable, temp, theRule, table[theVariable][temp]);
}
table[theVariable][temp]=theRule;
}
}
}
void Grammar::addFirstIntoTable(int theVariable, int theFirst, int theRule)
{
for (int i=0; i<token[theFirst]->firstCount; i++)
{
if (table[theVariable][token[theFirst]->first[i]]!=theRule)
{
if (table[theVariable][token[theFirst]->first[i]]!=-1)
{
grammarError(theVariable, token[theFirst]->first[i], theRule,
table[theVariable][token[theFirst]->first[i]]);
}
table[theVariable][token[theFirst]->first[i]]=theRule;
}
}
}
int Grammar::varCount()
{
int result=0;
for (int i=0; i<tokenCount; i++)
{
if (!token[i]->terminal)
{
result++;
}
}
return result;
}
int Grammar::termCount()
{
int result=0;
for (int i=0; i<tokenCount; i++)
{
if (token[i]->terminal)
{
result++;
}
}
return result;
}
//leave the rule unremoved!!! as I have no way to do it!!!
void Grammar::removeToken(int tIndex)
{
delete[]token[tIndex]->name;
delete token[tIndex];
tokenCount--;
for (int i=tIndex; i<tokenCount; i++)
{
token[i]=token[i+1];
}
}
bool Grammar::isRedundant(int tIndex)
{
int k, theRule, theToken;
for (int i=0; i<tokenCount; i++)
{
for (int j=0; j<token[i]->count; j++)
{
k=0;
theRule=token[i]->production[j];
theToken=production[theRule][k];
while (theToken!=-1)
{
if (theToken==tIndex)
{
return false;
}
k++;
theToken=production[theRule][k];
}
}
}
return true;
}
//what is redundant? except the start variable
//the non-terminal never appears in other rules!
void Grammar::removeRedundant()
{
int tIndex=1;
bool findNew=false;
while (tIndex<tokenCount)
{
findNew=false;
if (!token[tIndex]->terminal)
{
if (isRedundant(tIndex))
{
removeToken(tIndex);
findNew=true;
}
}
if (findNew)
{
tIndex=1;
}
else
{
tIndex++;
}
}
}
void Grammar::printAllRules()
{
/*
cout<<"\nnow print all rules\n";
for (int i=0; i<=prodIndex; i++)
{
cout<<"rule index "<<i<<": ";
printRule(i);
cout<<"\n";
}
*/
int sum=0;
for (int i=0; i<tokenCount; i++)
{
if (!token[i]->terminal)
{
sum+=token[i]->count;
}
}
cout<<" total rules is:"<<prodIndex+1;
cout<<"\n and the sum is "<<sum<<endl;
}
void Grammar::addBeginEnd()
{
int begin=addToken(startStr, false);
int end=addToken(endStr, true);
prodIndex++;
production[prodIndex][0]=0;
production[prodIndex][1]=end;
production[prodIndex][2]=-1;
addRuleForToken(begin, prodIndex);
startSymbolIndex=begin;
stackBottomIndex=end;
}
/*
void Grammar::calculateProperty()
{
calculateNull();
calculateFirst();
calculateFollow();
}
*/
void Grammar::printToken(bool onlyVar)
{
for (int i=0; i<tokenCount; i++)
{
if (onlyVar)
{
if (token[i]->terminal)
{
continue;
}
}
cout<<"name: "<<token[i]->name<<endl;
cout<<" isNull: "<<(token[i]->isNull?"true":"false")<<endl;
cout<<" isTerminal: "<<(token[i]->terminal?"true":"false")<<endl;
cout<<" first: ";
for (int j=0; j<token[i]->firstCount; j++)
{
if (j!=0)
{
cout<<",";
}
cout<<"["<<j<<"]="<<token[token[i]->first[j]]->name;
}
cout<<"\n follow: ";
for (j=0; j<token[i]->followCount; j++)
{
if (j!=0)
{
cout<<",";
}
cout<<"["<<j<<"]="<<token[token[i]->follow[j]]->name;
}
cout<<endl;
}
}
//must use while loop to discover new set!!!!!
void Grammar::calculateFirst()
{
int i;
bool addNew;
do
{
i=0;
addNew=false;
while (i<tokenCount)
{
//the terminal contains itself
if (token[i]->terminal)
{
//addFirst don't judge if it is nullable!!
addFirst(i, i);
//token[i]->first[token[i]->firstCount++]=i;
}
else
{
//for all its rules
for (int j=0; j<token[i]->count; j++)
{
int theToken, k=0;
theToken=production[token[i]->production[j]][k];
//for each token in each rule
do
{
//add non epsilon set
if (addIntoFirst(i, theToken))
{
addNew=true;
}
//if it is not null, means it is end
if (!token[theToken]->isNull)
{
break;
}
//if it is nullable, continue
k++;
theToken=production[token[i]->production[j]][k];
}while (theToken!=-1);
//it means all token in this rule is nullable, so
if (theToken==-1)
{
//it is nullable
//addEpsilonIntoFirst(i);
addFirst(i, emptyIndex);
}
}
}
i++;
}
}while (addNew);
}
bool Grammar::addFollowIntoFollow(int target, int source)
{
bool addNew=false;
for (int i=0; i<token[source]->followCount; i++)
{
if (addFollow(target, token[source]->follow[i]))
{
addNew=true;
}
}
return addNew;
}
bool Grammar::addIntoFollow(int target, int source)
{
bool addNew=false;
if (source==-1)
{
return false;
}
for (int i=0; i<token[source]->firstCount; i++)
{
//add non-epsilon
if (!token[token[source]->first[i]]->isNull)
{
if (addFollow(target, token[source]->first[i]))
{
addNew=true;
}
}
}
return addNew;
}
void Grammar::calculateFollow()
{
int i;
bool addNew, started;
//token[startSymbolIndex]->follow[0]=stackBottomIndex;
//token[startSymbolIndex]->followCount=1;
do
{
i=0;
addNew=false;
while (i<tokenCount)
{
//the terminal contains itself
if (!token[i]->terminal)
{
for (int tIndex=0; tIndex<tokenCount; tIndex++)
{
//for all its rules
if (token[tIndex]->terminal)
{
continue;
}
//for each its rule
for (int j=0; j<token[tIndex]->count; j++)
{
int theToken, k=0, theRule=token[tIndex]->production[j];
started=false;
theToken=production[theRule][k];
//for each token in each rule
do
{
//the token appears here
if (started)
{
if (addIntoFollow(i, theToken))
{
addNew=true;
}
if (!token[theToken]->isNull)
{
break;
}
}
if (theToken==i)
{
started=true;
//add non epsilon set, including -1 situation!!!
}
//if it is not null, means it is end
//if it is nullable, continue
k++;
theToken=production[theRule][k];
}while (theToken!=-1);
//it means all token in this rule is nullable, so
if (started&&theToken==-1)
{
//it is nullable
//add current variable Follow(j) into Follow(i);
if (addFollowIntoFollow(i, tIndex))
{
addNew=true;
}
}
}
}
}
i++;
}
}while (addNew);
}
void Grammar::addRuleForToken(int tIndex, int ruleIndex)
{
token[tIndex]->production[token[tIndex]->count++]=ruleIndex;
}
void Grammar::shiftRule(int ruleIndex, bool left2Right, int offset)
{
int end=0;
/*
while (production[ruleIndex][end]!=-1)
{
end++;
}
*/
end=ruleLen(ruleIndex);
if (left2Right)
{
for (int i=end; i>=0; i--)
{
production[ruleIndex][i+offset]=production[ruleIndex][i];
}
}
else
{
for (int i=0; i<=end-offset; i++)
{
production[ruleIndex][i]=production[ruleIndex][i+offset];
}
checkEpsilon(ruleIndex);
}
}
void Grammar::checkEpsilon(int ruleIndex)
{
if (production[ruleIndex][0]==-1)
{
epsilonRule(ruleIndex);
}
}
bool Grammar::addFollow(int target, int source)
{
//check if it is already there
for (int i=0; i<token[target]->followCount; i++)
{
if (token[target]->follow[i]==source)
{
return false;
}
}
//add at end
token[target]->follow[token[target]->followCount++]=source;
return true;
}
bool Grammar::addFirst(int target, int source)
{
//check if it is already there
for (int i=0; i<token[target]->firstCount; i++)
{
if (token[target]->first[i]==source)
{
return false;
}
}
//add at end
token[target]->first[token[target]->firstCount++]=source;
return true;
}
//add non epsilon into it.
bool Grammar::addIntoFirst(int target, int source)
{
bool addNew=false;
if (token[source]->terminal)
{
if (!token[source]->isNull)
{
return addFirst(target, source);
}
else
{
return false;
}
}
for (int i=0; i<token[source]->firstCount; i++)
{
//add non epsilon
if (!token[token[source]->first[i]]->isNull)
{
if (addFirst(target, token[source]->first[i]))
{
addNew=true;
}
}
}
return addNew;
}
bool Grammar::Null(int tIndex)
{
if (token[tIndex]->terminal)
{
return token[tIndex]->isNull;
}
for (int i=0; i<token[tIndex]->count; i++)
{
int j=0, theToken;
theToken=production[token[tIndex]->production[i]][j];
do
{
if (theToken==tIndex||!Null(theToken))
{
break;
}
j++;
theToken=production[token[tIndex]->production[i]][j];
}while (theToken!=-1);
if (theToken==-1)
{
return true;
}
}
return false;
}
void Grammar::calculateNull()
{
for (int i=0; i<tokenCount; i++)
{
token[i]->isNull=Null(i);
}
}
int Grammar::findMetaSymbol(int& begin, int& end)
{
int theRule, theToken, k;
begin=end=-1;
for (int i=0; i<tokenCount; i++)
{
for (int j=0; j<token[i]->count; j++)
{
k=0;
theRule=token[i]->production[j];
theToken=production[theRule][k];
while (theToken!=-1)
{
if (theToken==beginIndex)
{
begin=k;
}
if (theToken==endIndex)
{
end=k;
return theRule;
}
k++;
theToken=production[theRule][k];
}
}
}
return -1;
}
void Grammar::doReplaceMetaSymbol(int ruleIndex, int begin, int end)
{
int newTokenIndex=forgeToken(0), i=0;
addRuleForToken(newTokenIndex, ++prodIndex);
//token[newTokenIndex]->production[token[newTokenIndex]->count++]=++prodIndex;
//token[newTokenIndex]->terminal=false;
copyRule(ruleIndex, prodIndex);
//shrink
while (production[ruleIndex][i+end+1]!=-1)
{
production[ruleIndex][begin+i]=production[ruleIndex][i+end+1];
i++;
}
production[ruleIndex][begin+i]=-1;
addAtEnd(ruleIndex, newTokenIndex);
/*
production[ruleIndex][begin]=newTokenIndex;
production[ruleIndex][begin+1]=-1;
*/
shiftRule(prodIndex, false, begin+1);
production[prodIndex][end-begin-1]=newTokenIndex;
production[prodIndex][end-begin]=-1;
addEpsilonForToken(newTokenIndex);
/*
if (epsilonIndex==-1)
{
eRuleIndex=++prodIndex;
epsilonRule(eRuleIndex);
}
else
{
eRuleIndex=epsilonIndex;
}
addRuleForToken(newTokenIndex, eRuleIndex);
*/
}
void Grammar::replaceMetaSymbol()
{
int begin, end, ruleIndex;
while ((ruleIndex=findMetaSymbol(begin, end))!=-1)
{
doReplaceMetaSymbol(ruleIndex, begin, end);
}
/*
for (int i=0; i<tokenCount; i++)
{
//if (token[i]->isMeta)
if (i==beginIndex||i==endIndex)
{
removeToken(i);
}
}
*/
}
void Grammar::addAtEnd(int ruleIndex, int toAdd)
{
int end;
end=ruleLen(ruleIndex);
production[ruleIndex][end++]=toAdd;
production[ruleIndex][end]=-1;
}
//left-recursion: A -> A a | b | c
//change to: A -> b A' | c A'
// A' -> a A' | empty
void Grammar::removeImmediate(int tIndex, int ruleIndex)
{
int newIndex=forgeToken(tIndex);
int holdRuleIndex=token[tIndex]->production[ruleIndex];
//sequence matters!
//change to: A -> b A'
for (int i=0; i<token[tIndex]->count; i++)
{
if (i!=ruleIndex)
{
addAtEnd(token[tIndex]->production[i], newIndex);
}
}
//shift
removeRuleFromToken(tIndex, ruleIndex);
addRuleForToken(newIndex, holdRuleIndex);
//token[newIndex]->production[token[newIndex]->count++]=holdRuleIndex;
shiftRule(holdRuleIndex, false, 1);
addAtEnd(holdRuleIndex, newIndex);
//add epsilon rule for new variable
addEpsilonForToken(newIndex);
/*
epsilonRule(++prodIndex);
addRuleForToken(newIndex, prodIndex);
*/
}
int Grammar::forgeToken(int index)
{
char str[MaxGrammarTokenLength+2], ch;
int len=strlen(token[index]->name);
int temp=0, i=0;
strcpy(str, token[index]->name);
ch=str[len-1];//get last char of name
if (ch>='0'&&ch<'9')
{
str[len-1]=ch+i+1;
}
else
{
str[len]='0'+i;
str[len+1]='\0';
}
//I won't bother to check repetitation of name
while (tokenIndex(str)!=-1)
{
i++;
if (ch>='0'&&ch<'9')
{
str[len-1]=ch+i+1;
}
else
{
str[len]='0'+i;
str[len+1]='\0';
}
}
return addToken(str, false);//it is non-terminal for sure
}
int Grammar::copyRule(int source, int target)
{
int i=0;
while (production[source][i]!=-1)
{
production[target][i]=production[source][i];
i++;
}
production[target][i]=-1;
return i;
}
void Grammar::doAddRule(int tIndex)
{
token[tIndex]->production[token[tIndex]->count++]=++prodIndex;
}
void Grammar::addRule(const char* str)
{
int index;
index=addToken(str);
production[prodIndex][curPos++]=index;
production[prodIndex][curPos]=-1;//set end
}
//if the token is already there, it return the index
//otherwise, it add new node in token array
int Grammar::addToken(const char* str, bool isTerminal)
{
int index;
index=tokenIndex(str);
if (index==-1)
{
index=tokenCount;
}
else
{
return index;
}
token[index]=createToken(str, isTerminal);
if (strcmp(str, optionBegin)==0)
{
beginIndex=index;
}
if (strcmp(str, optionEnd)==0)
{
endIndex=index;
}
tokenCount++;
if (strcmp(str, emptyStr)==0)
{
token[index]->isNull=true;
emptyIndex=index;
}
return index;
}
void Grammar::newRule(const char* str)
{
if (str!=NULL)
{
curIndex=addToken(str, false);
}
//add one pointer
token[curIndex]->production[token[curIndex]->count++]=++prodIndex;
token[curIndex]->terminal=false;
curPos=0;//reset to restart;
}
GrammarToken* Grammar::createToken(const char* theName, bool isTerminal)
{
GrammarToken* ptr=new GrammarToken;
ptr->name=new char[strlen(theName)+1];
strcpy(ptr->name, theName);
ptr->terminal=isTerminal;
ptr->count=ptr->firstCount=ptr->followCount=0;
ptr->isNull=false;
return ptr;
}
int Grammar::tokenIndex(const char* str)
{
for (int i=0; i<tokenCount; i++)
{
if (strcmp(str, token[i]->name)==0)
{
return i;
}
}
return -1;
}
int Grammar::checkRecursion(int tIndex, int curToken)
{
for (int i=0; i<token[curToken]->count; i++)
{
//token[tIndex]->production[i]=ruleIndex
//production[ruleIndex][0] is the first left-recursion one
if (production[token[curToken]->production[i]][0]<=curToken)
{
return i;
}
}
return -1;
}
void Grammar::printRule(int index)
{
int nodeIndex=0;
//cout<<" ~"<<index<<"~ ";
while (production[index][nodeIndex]!=-1)
{
cout<<token[production[index][nodeIndex]]->name<<" ";
nodeIndex++;
}
}
void Grammar::printRule(int tIndex, int rulePos)
{
cout<<token[tIndex]->name<<" ==> ";
printRule(token[tIndex]->production[rulePos]);
}
void Grammar::initialize()
{
tokenCount=curIndex=curPos=0;
prodIndex=-1;//in order to ++ blindly
for (int i=0; i<MaxStateCount; i++)
{
for (int j=0; j<MaxTokenCount; j++)
{
LRTable[i][j]=-1;
}
}
for (i=0; i<MaxTokenCount; i++)
{
for (int j=0; j<MaxTokenCount; j++)
{
table[i][j]=-1;
}
}
}
Grammar::Grammar()
{
initialize();
}
void Grammar::removeLeftRecursion()
{
int tIndex=0, curToken=0;
bool newChange=false;
while (tIndex<tokenCount)
{
if (!token[tIndex]->terminal)
{
for (int i=0; i<token[tIndex]->count; i++)
{
curToken=production[token[tIndex]->production[i]][0];
if (curToken<=tIndex&&!token[curToken]->terminal)
{
if (curToken!=tIndex)
{
replaceRule(tIndex, i);
}
else
{
removeImmediate(tIndex, i);
}
newChange=true;
}
}
}
//whenever there is some new findings, restart
if (!newChange)
{
tIndex++;
}
else
{
tIndex=0;
newChange=false;
}
}
}
void Grammar::replaceRule(int tIndex, int ruleIndex)
{
int pos, j, targetEnd, sourceEnd, curRule;
curRule=token[tIndex]->production[ruleIndex];
int curToken=production[curRule][0];
for (int i=token[curToken]->count-1; i>=0; i--)
{
if (i>0)
{
doAddRule(tIndex);
pos=copyRule(token[curToken]->production[i], prodIndex);
j=0;
while (production[curRule][j+1]!=-1)
{
production[prodIndex][pos+j]=production[curRule][j+1];
j++;
}
production[prodIndex][pos+j]=-1;
//addRuleForToken(curToken, prodIndex);
}
else
{
targetEnd=sourceEnd=0;
//curRule=token[tIndex]->production[ruleIndex];
while (true)
{
if (production[token[curToken]->production[0]][sourceEnd]==-1&&
production[curRule][targetEnd]==-1)
{
break;
}
if (production[token[curToken]->production[0]][sourceEnd]!=-1)
{
sourceEnd++;
}
if (production[curRule][targetEnd]!=-1)
{
targetEnd++;
}
}
j=targetEnd+sourceEnd-1;
while (j>=0)
{
if (j>=sourceEnd)
{
production[curRule][j]=production[curRule][j-sourceEnd+1];
}
else
{
production[curRule][j]=production[token[curToken]->production[0]][j];
}
j--;
}
}
}
}
void Grammar::calculateLookAhead()
{
replaceMetaSymbol();
calculateNull();
cout<<setw(20)<<"The Rule "<<"Look-Ahead "<<endl;
for (int i=0; i<tokenCount; i++)
{
if (token[i]->terminal)
{
continue;
}
for (int j=0; j<token[i]->count; j++)
{
int theRule, theToken, k=0;
theRule=token[i]->production[j];
cout<<setw(5);
cout<<token[i]->name<<" -> ";
cout<<setiosflags(ios::left);
printRule(theRule);
cout<<" ";
theToken=production[theRule][k];
while (theToken!=-1)
{
if (k!=0)
{
cout<<"+";
}
cout<<"First("<<token[theToken]->name<<")";
if (!token[theToken]->isNull)
{
break;
}
k++;
theToken=production[theRule][k];
}
if (theToken==-1)
{
if (k!=0)
{
cout<<"+";
}
cout<<"Follow("<<token[i]->name<<")";
}
cout<<"\n";
}
}
}
void Grammar::LLoptimize()
{
removeLeftRecursion();
commonLeftFactor();
}
void Grammar::LRoptimize()
{
calculateProperty();
}
//the return value is the count, not the index of rule
int Grammar::ruleLen(int ruleIndex)
{
int end=0;
while (production[ruleIndex][end]!=-1)
{
end++;
}
return end;
}
//optimize sequence is first common-factor then remove left recursion
//therefore I don't have to check if for same variable if there will be
//more than one left-recursion
void Grammar::optimize(bool forLLGrammar)
{
replaceMetaSymbol();
addBeginEnd();
if (forLLGrammar)
{
LLoptimize();
}
else
{
LRoptimize();
}
//removeRedundant();
calculateNull();
calculateFirst();
calculateFollow();
buildTable(forLLGrammar);
}
int Grammar::findCommonFactor(int tIndex, int ruleIndex)
{
for (int i=ruleIndex+1; i<token[tIndex]->count; i++)
{
//if the two rule has the same first token
if (production[token[tIndex]->production[ruleIndex]][0]==
production[token[tIndex]->production[i]][0])
{
/*
//calculate if there is epsilon
if (emptyIndex==-1)
{
emptyIndex=tokenIndex(emptyStr);
}
//if it is not epsilon
if (production[token[tIndex]->production[i]][0]!=emptyIndex)
{
return i;
}
*/
return i;
}
}
return -1;
}
void Grammar::epsilonRule(int ruleIndex)
{
production[ruleIndex][0]=addToken(emptyStr);
production[ruleIndex][1]=-1;
}
//sample: x -> Aa
// x -> Ab
//changed to: x -> Ax' //this is to change the old rule
// x' -> b //this is to change the old rule
// x' -> a //this is the new-added rule
void Grammar::doCommonFactor(int tIndex, int ruleIndex, int index)
{
int newTokenIndex=forgeToken(tIndex);//create a new variable
//move the second and after part to the new rule of new variable
//doing: x' -> a
//curPos=0;
//prepare to add one new rule
addRuleForToken(newTokenIndex, ++prodIndex);
//token[newTokenIndex]->production[token[newTokenIndex]->count++]=++prodIndex;
token[newTokenIndex]->terminal=false;
copyRule(token[tIndex]->production[ruleIndex], prodIndex);
shiftRule(prodIndex, false, 1);
/*
do
{
//do copying
production[prodIndex][curPos]=
production[token[tIndex]->production[ruleIndex]][curPos+1];
curPos++;
//even the -1 at end is copied
}while (production[token[tIndex]->production[ruleIndex]][curPos]!=-1);
*/
//in order to show an empty string, in case the string is "epsilon"
/*
if (curPos==1)
{
epsilonRule(prodIndex);
}
*/
//replace x -> Aa with x -> Ax'
production[token[tIndex]->production[ruleIndex]][1]=newTokenIndex;
production[token[tIndex]->production[ruleIndex]][2]=-1;//end
//doing: x' -> b
//curPos=0;
//prepare to add one new rule
//pointing new token to where old rule lies
addRuleForToken(newTokenIndex, token[tIndex]->production[index]);
/*
token[newTokenIndex]->production[token[newTokenIndex]->count++]=
token[tIndex]->production[index];
*/
shiftRule(token[tIndex]->production[index], false, 1);
removeRuleFromToken(tIndex, index);
}
void Grammar::removeRuleFromToken(int tIndex, int ruleIndex)
{
token[tIndex]->count--;
for (int i=ruleIndex; i<token[tIndex]->count; i++)
{
token[tIndex]->production[i]=token[tIndex]->production[i+1];
}
}
void Grammar::commonLeftFactor()
{
int index=-1, tIndex=0, ruleIndex=0;
bool flag;
//whenever a newrule is done, restart!
while (tIndex<tokenCount)
{
ruleIndex=0;
flag=false;
while (ruleIndex<token[tIndex]->count)
{
index=findCommonFactor(tIndex, ruleIndex);
if (index!=-1)
{
doCommonFactor(tIndex, ruleIndex, index);
//restart
flag=true;
break;
}
else
{
ruleIndex++;
}
}
if (flag)
{
tIndex=0;
}
else
{
tIndex++;
}
}
}
GrammarToken* Grammar::operator[](int index)
{
if (index>=0&&index<tokenCount)
{
return token[index];
}
else
{
return NULL;
}
}
void Grammar::print()
{
for (int i=0; i<tokenCount; i++)
{
if (!token[i]->terminal)
{
cout<<token[i]->name<<" ==> ";
for (int j=0; j<token[i]->count; j++)
{
printRule(token[i]->production[j]);
if (j!=token[i]->count-1)
{
cout<<" | ";
}
}
cout<<"\n";
}
}
}
void Grammar::initializeDFA()
{
items=new Item*[MaxItemCount];
for (int i=0; i<MaxItemCount; i++)
{
items[i]=new Item;
items[i]->rulePos=items[i]->dotPos=items[i]->varIndex=-1;
}
sets= new BitSet*[MaxStateCount];
expandFinished=new bool[MaxStateCount];
for (i=0; i<MaxStateCount; i++)
{
sets[i]=new BitSet;
sets[i]->reset();
expandFinished[i]=false;
}
itemCount=stateCount=0;
//initialize LRTable
for (i=0; i<MaxStateCount; i++)
{
for (int j=0; j<tokenCount; j++)
{
LRTable[i][j]=-1;
}
}
}
int Grammar::addItem(Item& theItem)
{
for (int i=0; i<itemCount; i++)
{
if (*(items[i])== theItem)
{
return i;
}
}
items[i]=new Item;
*(items[i])=theItem;
itemCount++;
return i;
}
void Grammar::addSonItem(int tIndex, BitSet& theSet)
{
if (!token[tIndex]->terminal)
{
for (int i=0; i<token[tIndex]->count; i++)
{
Item temp;
int itemIndex;//easy reading
temp.varIndex=tIndex;
temp.rulePos=i;
temp.dotPos=0;
itemIndex=addItem(temp);
theSet.set(itemIndex);
//recursive
addSonItem(production[token[tIndex]->production[i]][0], theSet);
}
}
}
int Grammar::createItem(int theVar, int theNo, int position)
{
Item temp;
temp.varIndex=theVar;
temp.rulePos=theNo;
temp.dotPos=position;
return addItem(temp);
}
void Grammar::constructDFA()
{
int startIndex, itemIndex, stateIndex;
BitSet theSet;
//should initialize everything!!!
initializeDFA();
startIndex=createItem(startSymbolIndex, 0, 0);
theSet.set(startIndex);
addClosure(startIndex, theSet);
stateIndex=itemIndex=0;
stateIndex= createState(theSet);
expandState(stateIndex);//recursion inside
//for all choices, do createState(startStart, tansition token) (
}
void Grammar::addClosure(int itemIndex, BitSet& theSet)
{
int theNextVar=findNextTokenOfItem(itemIndex);
theSet.set(itemIndex);//add itself first!!!
/*
theSet.set(itemIndex);
tIndex=items[itemIndex]->varIndex;//easy for reading
theNo=items[itemIndex]->rulePos;
thePos=items[itemIndex]->dotPos;
//maybe I should write small function to wrap this kind of indexing!!!!
theVar=production[token[tIndex]->production[theNo]][thePos];
*/
if (theNextVar!=-1)
{
addSonItem(theNextVar, theSet);//add son!!!
}
}
int Grammar::createState(const BitSet& theSet)
{
for (int i=0; i<stateCount; i++)
{
if (*(sets[i])==theSet)
{
return i;
}
}
sets[i]=new BitSet;
*sets[i] = theSet;
stateCount++;
return i;
}
//indicating if it is reduceable by returning true
void Grammar::doExpandState(int itemIndex, int tIndex, BitSet& theSet)
{
int theNewItemIndex;
//int theNextVar=findNextTokenOfItem(itemIndex);
theNewItemIndex=createItem(items[itemIndex]->varIndex, items[itemIndex]->rulePos,
items[itemIndex]->dotPos+1);
addClosure(theNewItemIndex, theSet);
// return findNextTokenOfItem(theNewItemIndex)==-1;//true=reduceable
}
void Grammar::expandState(int stateIndex)
{
int itemIndex, targetState, tIndex, reduceIndex, shiftIndex;
BitSet theSet;//a new temp
bool findShift=false, findReduce=false;
//this will prevent looping expanding!!!???
expandFinished[stateIndex]=true;
//to see if it is to reduce
sets[stateIndex]->restart();
while ((itemIndex=sets[stateIndex]->next())!=-1)
{
if ((tIndex=findNextTokenOfItem(itemIndex))==-1)
{
findReduce=true;
int ruleIndex, rulePos, varIndex;
//only for easy reading
varIndex=items[itemIndex]->varIndex;
rulePos=items[itemIndex]->rulePos;
ruleIndex =token[varIndex]->production[rulePos];
add2Table(stateIndex, ruleIndex);
reduceIndex=itemIndex;
}
else
{
shiftIndex=itemIndex;
findShift=true;
}
}
//how to do in LR(1)????
//I am currently working on LR(0). I try not to think about it
if (findReduce&&findShift)
{
//in LR(1), this is not necessarily a conflict, because you need to
//check the first and follow set? I forgot it now. :)
errorHandle(ShiftReduceConflict);
//temparily
int temp1, temp2;
temp1=items[shiftIndex]->varIndex;
temp2=items[shiftIndex]->rulePos;
cout<<"\nconflict at shift:\n";
printRule(temp1, temp2);
temp1=items[reduceIndex]->varIndex;
temp2=items[reduceIndex]->rulePos;
cout<<"\nconflict with reduce:\n";
printRule(temp1, temp2);
cout<<endl;
return;//currently ignore LR(1) and should be improved when LR(1)
}
for (int i=0; i<tokenCount; i++)
{
sets[stateIndex]->restart();
//for all items, search if there is token to match to advance the dot
while ((itemIndex=sets[stateIndex]->next())!=-1)
{
if (findNextTokenOfItem(itemIndex)== i)//
{
doExpandState(itemIndex, i, theSet);//should finish addClosure
}
}
//only expandState after you find one "transition"
if (theSet.count()>0)//if you do find some
{
targetState=createState(theSet);
//here to write the LRTable
add2Table(stateIndex, targetState, i);//3 params means to reduce
findShift=true;
theSet.reset();//prepare for next token
if (!expandFinished[targetState])
{
expandState(targetState);
}
}//no one is set
}
/*
sets[stateIndex]->restart();
while ((itemIndex=sets[stateIndex]->next())!=-1)
{
if ((tIndex=findNextTokenOfItem(itemIndex))==-1)
{
findReduce=true;
int ruleIndex, rulePos, varIndex;
//only for easy reading
varIndex=items[itemIndex]->varIndex;
rulePos=items[itemIndex]->rulePos;
ruleIndex =token[varIndex]->production[rulePos];
add2Table(stateIndex, ruleIndex);
}
}
//conflict
if (findReduce&&findShift)
{
errorHandle(ShiftReduceConflict);
}
*/
}
//this is for reduce
void Grammar::add2Table(int stateIndex, int ruleIndex)
{
int result= reduce2Mask(ruleIndex);
//currently this is for LR(0)
for (int i=0; i<tokenCount; i++)
{
if (token[i]->terminal)
{
LRTable[stateIndex][i]=result;
}
}
}
//this is for shift
void Grammar::add2Table(int stateIndex, int targetIndex, int tIndex)
{
int result = shift2Mask(targetIndex);
LRTable[stateIndex][tIndex]=result;
}
int Grammar::findNextTokenOfItem(int itemIndex)
{
int tIndex, theNo, thePos, theVar;
tIndex=items[itemIndex]->varIndex;//easy for reading
theNo=items[itemIndex]->rulePos;
thePos=items[itemIndex]->dotPos;
//maybe I should write small function to wrap this kind of indexing!!!!
theVar=production[token[tIndex]->production[theNo]][thePos];
return theVar;
}
bool Grammar::isReduce(int input)
{
return (SHIFTREDUCEMASK&input)==REDUCEMASK;
}
bool Grammar::isShift(int input)
{
return (SHIFTREDUCEMASK&input)==SHIFTMASK;
}
int Grammar::mask2State(int input)
{
return STATEMASK&input;
}
//these are trivial functions and will later be changed to be inline
int Grammar::reduce2Mask(int input)
{
return REDUCEMASK|input;
}
int Grammar::shift2Mask(int input)
{
return SHIFTMASK|input;
}
BitSet::BitSet()
{
current=-1;
reset();
}
BitSet& BitSet::operator =(const BitSet& theSet)
{
reset();
this->operator |=(theSet);
return *this;
}
int BitSet::next()
{
for (int i=current+1; i<bitset_size; i++)
{
if (at(i))
{
current=i;
return i;
}
}
current=-1;
return -1;
}
bool operator == (Item& first, Item& second)
{
return (first.dotPos==second.dotPos)&&(first.rulePos==second.rulePos)&&
(first.varIndex==second.varIndex);
}
file name: parser.h
#include "grammar.h"
const int MaxStackLength=50;
class Parser
{
private:
int stack[MaxStackLength];
bool push(int num);
bool pop(int& num);
int top;
void initialize();
bool pushToken(int tIndex, int theToken);
void prepare();
void pushRule(int theRule);
public:
Parser();
void parseFile(const char* fileName);
};
file name: parser.cpp
#include <iostream>
#include "parser.h"
#include "scanner.h"
#include "errorNo.h"
using namespace std;
Scanner scanner;
extern Grammar grammar;
const char* defaultListFile="nickListFile.txt";
extern int startSymbolIndex;
extern int stackBottomIndex;
extern int table[MaxTokenCount][MaxTokenCount];
extern int token2type[MaxTokenCount];
extern int type2token[MaxTokenCount];
extern int emptyIndex;
bool Parser::pushToken(int tIndex, int theToken)
{
if (table[tIndex][theToken]!=-1)
{
cout<<grammar.token[tIndex]->name<<" => ";
grammar.printRule(table[tIndex][theToken]);
cout<<endl;
pushRule(table[tIndex][theToken]);
return true;
}
return false;
}
void Parser::pushRule(int theRule)
{
int len=0;
while (grammar.production[theRule][len]!=-1)
{
len++;
}
for (int i=len-1; i>=0; i--)
{
if (!push(grammar.production[theRule][i]))
{
errorHandle(StackOverFlow);
}
}
}
void Parser::parseFile(const char*fileName)
{
int theToken, theVar;
bool canPop=true;
scanner.readFromFile(fileName, defaultListFile);
prepare();
while (scanner.nextToken())
{
if (!canPop)
{
errorHandle(UnexpectedEmptyStack);
}
if (scanner.token.type==COMMENTTYPE)
{
continue;
}
theToken=type2token[scanner.token.type];
while ((canPop=pop(theVar))==true)
{
if (grammar.token[theVar]->terminal)
{
if (theToken==theVar)//match
{
break;
}
else
{
if (theVar==emptyIndex)
{
continue;
}
errorHandle(IllegalGrammarToken);
}
}
else
{
if (!pushToken(theVar, theToken))
{
errorHandle(IllegalGrammarToken);
}
}
}
}
if (pop(theVar))
{
if (theVar!=stackBottomIndex)
{
errorHandle(NotEmptyStack);
}
/*
if (theVar!=stackBottomIndex)
{
errorHandle(NotEmptyStack);
}
*/
}
}
Parser::Parser()
{
initialize();
}
void Parser::initialize()
{
top=0;
}
bool Parser::pop(int& num)
{
if (top==0)
{
return false;
}
num=stack[--top];
return true;
}
bool Parser::push(int num)
{
if (top==MaxStackLength-1)
{
return false;
}
stack[top++]=num;
return true;
}
void Parser::prepare()
{
top=0;
pushRule(grammar.token[startSymbolIndex]->production[0]);
}
file name: scanner.h
///////////////////////////////////////////////////////////////////////////////////////////
//Program: SLang Scanner
//Author: Qingzhe Huang
//Date: Jan. 18, 2004
//FileName: scanner.h
//Features:
// 1. I want to improve efficiency of scanning, so I used table-driven method.
// 2. I used enum to represent character of all ASCII---CharType---where "space, tab,
// end of line, end of file are all considered to be White Space.
// 3. All legal token is represented by enum TokenType.
// 4. I defined a huge amount of TokenState which is basically the state of a DFA. As
// I don't want to search reserved keyword with linear search or whatever, I have
// many states for the reserved words.
// 5. I deliberately make the sequence of first 38 TokenState elements exactly same as
// all that of TokenType, so that each final state of DFA has a 1-1 correspondence with
// type of token.
// 6. I defined a struct of Token which may be used in future parser.
// 7. I defined an errorNo variable to represent various errors. And a series error string
// for displaying information.
// 8. When class Scanner is created, it will initialize the big "state-charType" table.
// 9. When readFromFile is called, it will first read one char in advance.
// 10. When an error is encountered, the caller of Scanner should understand that no further
// char is read in. So, stop calling "nextToken()". This is a bit controvercial, and I
// plan to change it in next version.
////////////////////////////////////////////////////////////////////////////////////////////
/*////////////////////////////////////////////////////////////////////////////
Program: SLang Scanner
Author: Qingzhe Huang
Date: Jan. 21, 2004
FileName: scanner.h
Features:
1. I restructured the struct Token, to make it a union field in order to store int
value for number.
2. I restructured the function "nextToken()" in order to give out correct line no.
when error occurs.
*////////////////////////////////////////////////////////////////////////////////
#ifndef SCANNER_H
#define SCANNER_H
#include <iostream>
using namespace std;
extern enum ErrorCode;
const int TokenStateCount=138;
const int CharTypeCount=72;
const int MaxTokenLength=255;
const int MaxNumberLength=12;
enum CharType
{
//all small letters 26
SMALLA,SMALLB,SMALLC,SMALLD,SMALLE,SMALLF,SMALLG,SMALLH,SMALLI,SMALLJ,SMALLK,SMALLL,
SMALLM,SMALLN,SMALLO,SMALLP,SMALLQ,SMALLR,SMALLS,SMALLT,SMALLU,SMALLV,SMALLW,SMALLX,
SMALLY,SMALLZ,
//all big letters 26
BIGA,BIGB,BIGC,BIGD,BIGE,BIGF,BIGG,BIGH,BIGI,BIGJ,BIGK,BIGL,BIGM,BIGN,BIGO,BIGP,BIGQ,
BIGR,BIGS,BIGT,BIGU,BIGV,BIGW,BIGX,BIGY,BIGZ,
//all digit 1
DIGIT,
//all symbols 16
QUOTE, OPENPAR, CLOSEPAR, SEMICOLON,PLUS, MINUS, TIMES, SLASH, COLON,
EQUAL,SMALLER,GREATER,EXCLAIM,OPENBRACKET, CLOSEBRACKET,COMMA,
//space, tab, end of line are regarded as whitespace, 1
WHITESPACE,
//UNDERSCORE IS A SPECIAL SYMBOL 1
UNDERSCORE,
//all other ASCII is regarded as illegal 1
ILLEGAL
};
//TOTAL 38, JUST 1-1 WITH THE FIRST 38 OF TOKENSTATE
enum TokenType
{
//GENERAL TYPE 5
IDTYPE, NUMBERTYPE, CHARCONSTTYPE, COMMENTTYPE, ERRORTYPE,
//THE FOLLOWING ARE SYMBOL TYPE 18
OPENPARTYPE, CLOSEPARTYPE, SEMICOLONTYPE, PLUSTYPE, MINUSTYPE, TIMESTYPE,
SLASHTYPE, ASSIGNMENTTYPE, SMALLERTYPE, GREATERTYPE, EQUALTYPE, SMALLEREQUALTYPE,
GREATEREQUALTYPE, NOTEQUALTYPE, OPENBRACKETTYPE, CLOSEBRACKETTYPE, COMMATYPE,
COLONTYPE,
//THE FOLLOWING ARE RESERVED TYPE 15
BEGINTYPE, ENDTYPE, PROGRAMTYPE, VARIABLESTYPE,INTEGERTYPE, ARRAYTYPE, CHARTYPE,
MODULETYPE, IFTYPE, THENTYPE, ELSETYPE, LOOPTYPE, EXITTYPE, READTYPE, WRITETYPE
};
//total 138 states
enum TokenState
{
//THE FINAL STATE 38, in order to easy initialize "finalState", I put them in beginning
//5 generals
IDEND, NUMBEREND, CONSTCHAREND, COMMENTEND, ERROR,
//18 symbols
OPENPAREND, CLOSEPAREND, SEMICOLONEND, PLUSEND, MINUSEND, TIMESEND,
SLASHEND, ASSIGNMENTEND, SMALLEREND, GREATEREND, EQUALEND, SMALLEREQUALEND,
GREATEREQUALEND, NOTEQUALEND, OPENBRACKETEND, CLOSEBRACKETEND, COMMAEND,
COLONEND,
//15 reserved
BEGINEND, ENDEND, PROGRAMEND, VARIABLESEND, INTEGEREND, ARRAYEND, CHAREND,
MODULEEND, IFEND, THENEND, ELSEEND, LOOPEND, EXITEND, READEND, WRITEEND,
//THE FOLLOWING ARE ALL NON-FINAL STATES
//THE very FIRST CHAR 1
READY,
//THE FOLLOWING ARE ALL RESERVED STATE
//the first char 12
ARRAY1, BEGIN1, CHAR1, E1, I1, LOOP1, MODULE1, PROGRAM1, READ1, THEN1, VARIABLES1,
WRITE1,
//THE SECOND CHAR 15
ARRAY2, BEGIN2, CHAR2, ELSE2, END2, EXIT2, IF2, INTEGER2, LOOP2, MODULE2, PROGRAM2,
READ2, THEN2, VARIABLES2, WRITE2,
//THE THIRD CHAR 14
ARRAY3, BEGIN3, CHAR3, ELSE3, END3, EXIT3, INTEGER3, LOOP3, MODULE3, PROGRAM3, READ3,
THEN3, VARIABLES3, WRITE3,
//THE FOURTH CHAR 13
ARRAY4, BEGIN4, CHAR4, ELSE4, EXIT4, INTEGER4, LOOP4, MODULE4, PROGRAM4, READ4, THEN4,
VARIABLES4, WRITE4,
//THE FIFTH CHAR 7
ARRAY5, BEGIN5, INTEGER5, MODULE5, PROGRAM5, VARIABLES5, WRITE5,
//THE SIXTH CHAR 4
INTEGER6, MODULE6, PROGRAM6, VARIABLES6,
//THE SEVENTH CHAR 3
INTEGER7, PROGRAM7, VARIABLES7,
//THE EIGHTH CHAR 1
VARIABLES8,
//THE NINETH CHAR 1
VARIABLES9,
//THESE ARE NON-RESERVED
//THESE ARE GENERAL 9
IDBEGIN, IDUNDERSCORE, NUMBERBEGIN, CONSTCHARQUOTEBEGIN, CONSTCHARBEGIN, COMMENTSTARBEGIN,
COMMENTBEGIN, COMMENTSTAREND, COMMENTSLASHBEGIN,
//the SINGLE symbols 16
QUOTEBEGIN, OPENPARBEGIN, CLOSEPARBEGIN, SEMICOLONBEGIN,
PLUSBEGIN, MINUSBEGIN, TIMESBEGIN, SLASHBEGIN, COLONBEGIN, SMALLERBEGIN, GREATERBEGIN,
EQUALBEGIN, EXCLAIMBEGIN, OPENBRACKETBEGIN, CLOSEBRACKETBEGIN, COMMABEGIN,
//MULTI SYMBOL 4
ASSIGNMENTBEGIN, SMALLEREQUALBEGIN,
GREATEREQUALBEGIN, NOTEQUALBEGIN
};
//extern ErrorCode errorNo;
extern void errorHandle(ErrorCode errorNo);
struct Token
{
TokenType type;
union
{
char name[MaxTokenLength+1];
int number;
};
};
class Scanner
{
private:
int tokenCount;
unsigned char ch;
void printLineNo();
FILE* stream;
bool nextChar();
void initialize();
bool resume();
public:
Scanner();
static Token token;
bool readFromFile(const char* fileName, const char* listFileName);
const char* getToken(){return token.name;}
bool nextToken();
void report();
};
void initialTokenState();
#endif
file name: scanner.cpp
/*////////////////////////////////////////////////////////////////////////////
Program: SLang Scanner
Author: Qingzhe Huang
Date: Jan. 21, 2004
FileName: scanner.cpp
Features:
1. As Dr. Optrany said, the number should be stored as int or double whatever.
2. I restructured the function "nextToken()" in order to give out correct line no.
when error occurs.
*////////////////////////////////////////////////////////////////////////////////
#include <iostream>
#include <fstream>
#include "scanner.h"
#include "errorNo.h"
using namespace std;
ofstream fList;
//this will determine how many errors of maximum the scanner will tolerant
const int MaxErrortolerant=10;
//as integer usually have max 12 digit roughly
int errorCount=0;
int lineCount=1;
//static memeber
Token Scanner::token;
const int TokenTypeCount=38;
//extern void errorHandle(int errorNo);
extern char* errorStr[ErrorCount];
//this is purely for displaying purpose
char* tokenTypeStr[TokenTypeCount]=
{
//GENERAL TYPE 5
"ID", "NUMBER", "CHARACTER CONSTANT", "COMMENT", "ERROR",
//THE FOLLOWING ARE SYMBOL TYPE 18
"(", ")", ";", "+", "-", "*",
"/", ":=", "<", ">", "=", "<=",
">=", "!=", "[", "]", ",",
":",
//THE FOLLOWING ARE RESERVED TYPE 15
"begin", "end", "program", "variables","integer", "array", "char",
"module", "if", "then", "else", "loop", "exit", "read", "write"
};
CharType charType[256];
TokenState tokenState[TokenStateCount][CharTypeCount];
//when error occurs, no message is immediately output, it is
//postponed to next time, because when '\n' is read in,
//lineCount is not incremented until token is decided.
//Therefore, when a token is ended with '\n', the line no is not updated
//until next round. So, we can keep the correct line no. for each token
bool Scanner::nextToken()
{
TokenState state=READY;
int digitCount=0;
int value=0;
int count=0;//to count the length of token
char* ptr=token.name;
bool isComment=false;
do
{
//map ch to CharType reducing 256 ASCII to 73 CharTypes
//the table for "state" and "CharType is 138x73, each entry is a
//index for state.
state=tokenState[state][charType[ch]];
if (state==NUMBERBEGIN)
{
digitCount++;
if (digitCount>=MaxNumberLength)
{
errorHandle(ExceedNumberLimit);
return resume();
}
//to accumulate the value
value*=10;
value+=ch-'0';
}
//because I put all final state in the first 38 positions
if (state<38)
{
//This is a dirty trick! Because I make the "TokenType" 1-1 with
//TokenState for the 38 finals.
*ptr='\0';
token.type=(TokenType)(state);
if (state==ERROR)
{
errorHandle(IllegalToken);
//printLineNo();
return resume();
}
tokenCount++;
if (state==NUMBEREND)
{
token.number=value;
}
//printLineNo();
return true;
}
if (state==COMMENTBEGIN)
{
isComment=true;
}
if (count>=MaxTokenLength)
{
errorHandle(TokenTooLong);
token.type=ERRORTYPE;
//printLineNo();
return false;
}
//cout<<ch;
if (!isComment&&state!=READY)
{
*ptr=ch;
ptr++;
count++;
}
//it is only at end to update line no.
printLineNo();
}while (nextChar());
state=tokenState[state][charType[ch]];
//at this point, it is either in ready state, or error state
if (state==ERROR)
{
token.type=(TokenType)(state);
errorHandle(UnexpectedReachEOF);
}
//but in all case it means end of file, so return false
return false;
}
bool Scanner::resume()
{
//the scanner will try to continue if error number is within 10
if (errorCount==MaxErrortolerant)
{
return false;
}
return nextChar();
}
void Scanner::report()
{
fList<<"\ntotal number of tokens is "<<tokenCount;
fList<<"\ntotal number of errors is "<<errorCount;
}
void Scanner::printLineNo()
{
if (ch=='\n')
{
fList<<++lineCount<<" ";
}
}
Scanner::Scanner()
{
initialize();
}
void Scanner::initialize()
{
errorCount=0;
lineCount=1;
tokenCount=0;
initialTokenState();
}
bool Scanner::readFromFile(const char* fileName, const char* listFileName)
{
if ((stream=fopen(fileName, "r"))==NULL)
{
errorHandle(CannotOpenFile);
return false;
}
else
{
fList.open(listFileName);
fList<<lineCount<<" ";
//this is to prevent the empty file situation in which
//you cannot even read one single char because my scanner need to read
//one char ahead
if (!nextChar())
{
errorHandle(FileEmptyError);
return false;
}
}
return true;
}
bool Scanner::nextChar()
{
ch=fgetc(stream);
fList<<ch;
return ch!=255;
}
file name: errorno.h
#ifndef ERRORNO_H #define ERRORNO_H extern char* errorStr[]; void errorHandle(int errorNo); const int ErrorCount=10; const int ScannerErrorCount=6; const int ParserErrorCount=4; //errors of scanner = 6 #define IllegalToken 0 #define TokenTooLong 1 #define UnexpectedReachEOF 2 #define FileEmptyError 3 #define CannotOpenFile 4 #define ExceedNumberLimit 5 //error of parser =4 #define UnexpectedEmptyStack 6 #define IllegalGrammarToken 7 #define NotEmptyStack 8 #define StackOverFlow 9 //error of LR(0) #define ShiftReduceConflict 10 #endif
file name: errorno.cpp
#include <iostream>
#include <fstream>
#include "scanner.h"
#include "errorNo.h"
using namespace std;
char* errorStr[ErrorCount]=
{
//these are scanner errors:
"IllegalToken",
"TokenTooLong",
"UnexpectedReachEOF",
"FileEmptyError",
"CannotOpenFile",
"ExceedNumberLimit",
//these are parser errors
"UnexpectedEmptyStack",
"IllegalGrammarToken",
"NotEmptyStack",
"StackOverFlow"
};
extern int errorCount;
extern int lineCount;
extern ofstream fList;
//this is going to be improved in future as parser need to
//call it, too. so, more parameter should be added?
//No! the error no. itself specifies the error and it is
//error handler to try to find necessary info to display.
void errorHandle(int errorNo)
{
if (errorNo<ScannerErrorCount)
{
errorCount++;
//the illegal token may be for various reason and I only suggest
//a possible nearby place to spot the error occurs.
fList<<"\nerror of "<<errorStr[errorNo]<<" occurred at line "
<<lineCount<<" near token "<<Scanner::token.name<<endl;
}
else
{
if (errorNo<ScannerErrorCount+ParserErrorCount)
{
errorCount++;
fList<<"\nerror of "<<errorStr[errorNo]<<" occurred at line "
<<lineCount<<" near token "<<Scanner::token.name<<endl;
exit(errorNo);
}
}
}
file name: initialize.cpp
/*////////////////////////////////////////////////////////////////////////////
Program: SLang Scanner
Author: Qingzhe Huang
Date: Jan. 18, 2004
FileName: initialize.cpp
Features:
1. This is purely mechnical job, you know to initialize a huge state table:
138x72 is really a boring, routine job.
2. For EOF, I want "ch" to be able to be an index in "CharType" array, so, it
cannot be -1, but 255 for "unsigned char" which is declared in class Scanner.
*////////////////////////////////////////////////////////////////////////////////
#include "scanner.h"
extern enum CharType;
extern enum TokenState;
extern CharType charType[256];
extern TokenState tokenState[TokenStateCount][CharTypeCount];
void finalSymbolToken(TokenState state, TokenState endState);
void finalReservedToken(TokenState state, TokenState endState);
void initialCharType();
void setFinalTokenState();
void initialReserved(TokenState state);
void setRange(TokenState state, CharType start, CharType end, TokenState target);
void setState(TokenState state, TokenState targetState);
void setDefaultState();
void setDefaultState()
{
//all states are by default error
for (int i=0; i<TokenStateCount; i++)
{
setState((TokenState)i, ERROR);
}
//the default for all letters are IDBEGIN
setRange(READY, SMALLA, BIGZ, IDBEGIN);
//THIS IS another dirty trick, since I put all reserved states together
//so you can initialize them together.
for (i=ARRAY1; i<=VARIABLES9; i++)
{
initialReserved((TokenState)i);
}
setFinalTokenState();
}
void setFinalTokenState()
{
//FOR ID
finalReservedToken(IDBEGIN, IDEND);
//for number
finalReservedToken(NUMBERBEGIN, NUMBEREND);
//THESE FOR RESERVED WORDS
finalReservedToken(ARRAY5, ARRAYEND);
finalReservedToken(BEGIN5, BEGINEND);
finalReservedToken(CHAR4, CHAREND);
finalReservedToken(ELSE4, ELSEEND);
finalReservedToken(END3, ENDEND);
finalReservedToken(EXIT4, EXITEND);
finalReservedToken(IF2, IFEND);
finalReservedToken(INTEGER7, INTEGEREND);
finalReservedToken(LOOP4, LOOPEND);
finalReservedToken(MODULE6, MODULEEND);
finalReservedToken(PROGRAM7, PROGRAMEND);
finalReservedToken(READ4, READEND);
finalReservedToken(THEN4, THENEND);
finalReservedToken(VARIABLES9, VARIABLESEND);
finalReservedToken(WRITE5, WRITEEND);
//THESE FOR SYMBOLS
finalSymbolToken(OPENPARBEGIN, OPENPAREND);
finalSymbolToken(CLOSEPARBEGIN, CLOSEPAREND);
finalSymbolToken(SEMICOLONBEGIN, SEMICOLONEND);
finalSymbolToken(PLUSBEGIN, PLUSEND);
finalSymbolToken(MINUSBEGIN, MINUSEND);
finalSymbolToken(TIMESBEGIN, TIMESEND);
finalSymbolToken(SLASHBEGIN, SLASHEND);
finalSymbolToken(ASSIGNMENTBEGIN, ASSIGNMENTEND);
finalSymbolToken(SMALLERBEGIN, SMALLEREND);
finalSymbolToken(GREATERBEGIN, GREATEREND);
finalSymbolToken(EQUALBEGIN, EQUALEND);
finalSymbolToken(SMALLEREQUALBEGIN, SMALLEREQUALEND);
finalSymbolToken(GREATEREQUALBEGIN, GREATEREQUALEND);
finalSymbolToken(NOTEQUALBEGIN, NOTEQUALEND);
finalSymbolToken(OPENBRACKETBEGIN, OPENBRACKETEND);
finalSymbolToken(CLOSEBRACKETBEGIN, CLOSEBRACKETEND);
finalSymbolToken(COMMABEGIN, COMMAEND);
finalSymbolToken(COLONBEGIN, COLONEND);
//COMMENT
finalSymbolToken(COMMENTSLASHBEGIN, COMMENTEND);
//CONSTCHAR
finalSymbolToken(CONSTCHARQUOTEBEGIN, CONSTCHAREND);
}
void initialTokenState()
{
//initialize all charType
initialCharType();
//default is always error
setDefaultState();
//loop
tokenState[READY][WHITESPACE]=READY;
//number
tokenState[READY][DIGIT]=NUMBERBEGIN;
tokenState[NUMBERBEGIN][DIGIT]=NUMBERBEGIN;//HOW LONG SHOULD NUMBER BE?
//ID
//setRange(READY, SMALLA, BIGZ, IDBEGIN); THIS IS IN DEFAULT
setRange(IDBEGIN, SMALLA, DIGIT, IDBEGIN);
tokenState[IDBEGIN][UNDERSCORE]=IDUNDERSCORE;
setRange(IDUNDERSCORE, SMALLA, DIGIT, IDBEGIN);
//reserved words
//ARRAY1, BEGIN1, CHAR1, E1, I1, LOOP1, MODULE1, PROGRAM1, READ1, THEN1, WRITE1,
//VARIABLES1,
tokenState[READY][SMALLA]=ARRAY1;
tokenState[READY][SMALLB]=BEGIN1;
tokenState[READY][SMALLC]=CHAR1;
tokenState[READY][SMALLE]=E1;
tokenState[READY][SMALLI]=I1;
tokenState[READY][SMALLL]=LOOP1;
tokenState[READY][SMALLM]=MODULE1;
tokenState[READY][SMALLP]=PROGRAM1;
tokenState[READY][SMALLR]=READ1;
tokenState[READY][SMALLT]=THEN1;
tokenState[READY][SMALLV]=VARIABLES1;
tokenState[READY][SMALLW]=WRITE1;
/* RESERVED WORDS
ARRAY2 */
tokenState[ARRAY1][SMALLR]=ARRAY2;
//BEGIN2
tokenState[BEGIN1][SMALLE]=BEGIN2;
//CHAR2
tokenState[CHAR1][SMALLH]=CHAR2;
//ELSE2,
tokenState[E1][SMALLL]=ELSE2;
//EXIT2
tokenState[E1][SMALLX]=EXIT2;
//END2
tokenState[E1][SMALLN]=END2;
//IF2
tokenState[I1][SMALLF]=IF2;
//INTEGER2
tokenState[I1][SMALLN]=INTEGER2;
//LOOP2
tokenState[LOOP1][SMALLO]=LOOP2;
//MODULE2
tokenState[MODULE1][SMALLO]=MODULE2;
//PROGRAM2
tokenState[PROGRAM1][SMALLR]=PROGRAM2;
//READ2
tokenState[READ1][SMALLE]=READ2;
//THEN2
tokenState[THEN1][SMALLH]=THEN2;
//VARIABLES2
tokenState[VARIABLES1][SMALLA]=VARIABLES2;
//WRITE2
tokenState[WRITE1][SMALLR]=WRITE2;
/* RESERVED WORDS
ARRAY3 */
tokenState[ARRAY2][SMALLR]=ARRAY3;
//BEGIN2
tokenState[BEGIN2][SMALLG]=BEGIN3;
//CHAR2
tokenState[CHAR2][SMALLA]=CHAR3;
//ELSE2,
tokenState[ELSE2][SMALLS]=ELSE3;
//END2
tokenState[END2][SMALLD]=END3;
//EXIT2
tokenState[EXIT2][SMALLI]=EXIT3;
//INTEGER2
tokenState[INTEGER2][SMALLT]=INTEGER3;
//LOOP2
tokenState[LOOP2][SMALLO]=LOOP3;
//MODULE2
tokenState[MODULE2][SMALLD]=MODULE3;
//PROGRAM2
tokenState[PROGRAM2][SMALLO]=PROGRAM3;
//READ2
tokenState[READ2][SMALLA]=READ3;
//THEN2
tokenState[THEN2][SMALLE]=THEN3;
//VARIABLES2
tokenState[VARIABLES2][SMALLR]=VARIABLES3;
//WRITE2
tokenState[WRITE2][SMALLI]=WRITE3;
/* RESERVED WORDS
ARRAY3 */
tokenState[ARRAY3][SMALLA]=ARRAY4;
//BEGIN2
tokenState[BEGIN3][SMALLI]=BEGIN4;
//CHAR2
tokenState[CHAR3][SMALLR]=CHAR4;
//ELSE2,
tokenState[ELSE3][SMALLE]=ELSE4;
//EXIT2
tokenState[EXIT3][SMALLT]=EXIT4;
//INTEGER2
tokenState[INTEGER3][SMALLE]=INTEGER4;
//LOOP2
tokenState[LOOP3][SMALLP]=LOOP4;
//MODULE2
tokenState[MODULE3][SMALLU]=MODULE4;
//PROGRAM2
tokenState[PROGRAM3][SMALLG]=PROGRAM4;
//READ2
tokenState[READ3][SMALLD]=READ4;
//THEN2
tokenState[THEN3][SMALLN]=THEN4;
//VARIABLES2
tokenState[VARIABLES3][SMALLI]=VARIABLES4;
//WRITE2
tokenState[WRITE3][SMALLT]=WRITE4;
/* RESERVED WORDS
ARRAY */
tokenState[ARRAY4][SMALLY]=ARRAY5;
//BEGIN2
tokenState[BEGIN4][SMALLN]=BEGIN5;
//INTEGER2
tokenState[INTEGER4][SMALLG]=INTEGER5;
//MODULE2
tokenState[MODULE4][SMALLL]=MODULE5;
//PROGRAM2
tokenState[PROGRAM4][SMALLR]=PROGRAM5;
//VARIABLES2
tokenState[VARIABLES4][SMALLA]=VARIABLES5;
//WRITE2
tokenState[WRITE4][SMALLE]=WRITE5;
// RESERVED WORDS*/
//INTEGER2
tokenState[INTEGER5][SMALLE]=INTEGER6;
//MODULE2
tokenState[MODULE5][SMALLE]=MODULE6;
//PROGRAM2
tokenState[PROGRAM5][SMALLA]=PROGRAM6;
//VARIABLES2
tokenState[VARIABLES5][SMALLB]=VARIABLES6;
// RESERVED WORDS*/
//INTEGER2
tokenState[INTEGER6][SMALLR]=INTEGER7;
//PROGRAM2
tokenState[PROGRAM6][SMALLM]=PROGRAM7;
//VARIABLES2
tokenState[VARIABLES6][SMALLL]=VARIABLES7;
// RESERVED WORDS*/
//VARIABLES2
tokenState[VARIABLES7][SMALLE]=VARIABLES8;
//VARIABLES2
tokenState[VARIABLES8][SMALLS]=VARIABLES9;
/*
CONSTCHAR, UNDERSCOREBEGIN, ASSIGNMENTBEGIN, SMALLEREQUALBEGIN,
GREATEREQUALBEGIN, NOTEQUAL, COMMENTBEGIN, IDUNDERSCORE,*/
//now is the symbols
//QUOTEBEGIN, OPENPARBEGIN, CLOSEPARBEGIN, SEMICOLONBEGIN,
//PLUSBEGIN, MINUSBEGIN, TIMESBEGIN, SLASHBEGIN, COLONBEGIN, SMALLERBEGIN, GREATERBEGIN,
//EQUALBEGIN, EXCLAIMBEGIN, OPENBRACKETBEGIN, CLOSEBRACKETBEGIN, COMMABEGIN,
//'
tokenState[READY][QUOTE]=QUOTEBEGIN;
//(
tokenState[READY][OPENPAR]=OPENPARBEGIN;
//)
tokenState[READY][CLOSEPAR]=CLOSEPARBEGIN;
//;
tokenState[READY][SEMICOLON]=SEMICOLONBEGIN;
//+
tokenState[READY][PLUS]=PLUSBEGIN;
//-
tokenState[READY][MINUS]=MINUSBEGIN;
//*
tokenState[READY][TIMES]=TIMESBEGIN;
///
tokenState[READY][SLASH]=SLASHBEGIN;
//:
tokenState[READY][COLON]=COLONBEGIN;
//<
tokenState[READY][SMALLER]=SMALLERBEGIN;
//>
tokenState[READY][GREATER]=GREATERBEGIN;
//=
tokenState[READY][EQUAL]=EQUALBEGIN;
//!
tokenState[READY][EXCLAIM]=EXCLAIMBEGIN;
//[
tokenState[READY][OPENBRACKET]=OPENBRACKETBEGIN;
//]
tokenState[READY][CLOSEBRACKET]=CLOSEBRACKETBEGIN;
//,
tokenState[READY][COMMA]=COMMABEGIN;
//AFTER QUOTE IT CAN BE ANY CHARACTER, INCLUDING ILLEGAL CHAR
setRange(QUOTEBEGIN, SMALLA, ILLEGAL, CONSTCHARBEGIN);
//ANY OTHER STATE IS BY DEFAULT ERROR
tokenState[CONSTCHARBEGIN][QUOTE]=CONSTCHARQUOTEBEGIN;
//FOR /, DEFAULT IS SLASHEND, EXCEPT * WHICH IS COMMENTSTARBEGIN
tokenState[SLASHBEGIN][TIMES]= COMMENTSTARBEGIN;
//FOR :, DEFAULT IS COLONEND, EXCEPT FOR = WHICH IS ASSIGNMENTBEGIN
tokenState[COLONBEGIN][EQUAL]= ASSIGNMENTBEGIN;
//FOR <, DEFAULT IS SMALLEREND, EXCEPT FOR= WHICH IS SMALLEREQAULBEGIN
tokenState[SMALLERBEGIN][EQUAL]=SMALLEREQUALBEGIN;
//FOR >, DEFAULT IS GREATEREND, EXCEPT FOR= WHICH IS GREATEREQAULBEGIN
tokenState[GREATERBEGIN][EQUAL]= GREATEREQUALBEGIN;
tokenState[EXCLAIMBEGIN][EQUAL]= NOTEQUALBEGIN;
//WITHIN COMMENT IT IS A LOOP, EXCEPT FOR * WHICH IS POSSIBLE FOR END OF COMMENT
setRange(COMMENTSTARBEGIN, SMALLA, ILLEGAL, COMMENTBEGIN);
tokenState[COMMENTSTARBEGIN][TIMES]=COMMENTSTAREND;
setRange(COMMENTBEGIN, SMALLA, ILLEGAL, COMMENTBEGIN);
tokenState[COMMENTBEGIN][TIMES]=COMMENTSTAREND;
//FROM COMMENTSTARBEGIN, ALL IS BACK TO COMMENTBEGIN, EXCEPT / WHICH IS END OF COMMENT
setRange(COMMENTSTAREND, SMALLA, ILLEGAL, COMMENTBEGIN);
tokenState[COMMENTSTAREND][SLASH]=COMMENTSLASHBEGIN;
//
}
void initialReserved(TokenState state)
{
setRange(state, SMALLA, DIGIT, IDBEGIN);
finalReservedToken(state, IDEND);
tokenState[state][UNDERSCORE]=IDUNDERSCORE;//a_
}
void finalSymbolToken(TokenState state, TokenState endState)
{
for (int i=SMALLA; i<=WHITESPACE; i++)
{
tokenState[state][(CharType)i]=endState;
}
}
void finalReservedToken(TokenState state, TokenState endState)
{
//all non-letter, non-digit is regarded to be delimeter
for (int i=QUOTE; i<=WHITESPACE; i++)
{
tokenState[state][(CharType)i]=endState;
}
}
//the default charType is ILLEGAL
void initialCharType()
{
int chType;
//the default charType is ILLEGAL
for (int i=0; i<256; i++)
{
charType[i]=ILLEGAL;
}
//chType is SMALLA
chType=SMALLA;
for (i='a'; i<='z'; i++)
{
charType[i]=(CharType)(chType);
chType++;
}
//chType is now BIGA
chType=BIGA;//I don't want to rely on the trick.
for (i='A'; i<='Z'; i++)
{
charType[i]=(CharType)(chType);
chType++;
}
chType=DIGIT;
for (i='0'; i<='9'; i++)
{
charType[i]=(CharType)(chType);
}
/*
UNDERSCORE, QUOTE, OPENPAR, CLOSEPAR, SEMICOLON,PLUS, MINUS, TIMES, SLASH, COLON,
EQUAL,SMALLER,GREATER,EXCLAIM,OPENBRACKET, CLOSEBRACKET,COMMA,
SPACE,TAB, ENDLINE, ILLEGAL
*/
charType['_']=UNDERSCORE;
charType['\'']=QUOTE;
charType['(']=OPENPAR;
charType[')']=CLOSEPAR;
charType[';']=SEMICOLON;
charType['+']=PLUS;
charType['-']=MINUS;
charType['*']=TIMES;
charType['/']=SLASH;
charType[':']=COLON;
charType['=']=EQUAL;
charType['<']=SMALLER;
charType['>']=GREATER;
charType['!']=EXCLAIM;
charType['[']=OPENBRACKET;
charType[']']=CLOSEBRACKET;
charType[',']=COMMA;
charType[' ']=WHITESPACE;
charType['\t']=WHITESPACE;
charType[10]=WHITESPACE;
charType[13]=WHITESPACE;
//pls note, since I changed the type of "ch" to be "unsigned char"
//the EOF now is not -1, but 255
charType[255]=WHITESPACE;//IT IS A KIND OF DELIMETER
}
void setRange(TokenState state, CharType start, CharType end, TokenState target)
{
for (int i=start; i<=end; i++)
{
tokenState[state][i]=target;
}
}
void setState(TokenState state, TokenState targetState)
{
for (int i=0; i<CharTypeCount; i++)
{
tokenState[state][i]=targetState;
}
}
file name: main.cpp (main)
#include <iostream>
#include "CFGReader.h"
#include "Parser.h"
using namespace std;
int main( int argc, char *argv[ ])
{
if (!(argc==1||argc==3))
{
cout<<"usage: CRGReader grammarSourceFileName sourceFileToBeParsed";
exit(1);
}
CFGReader R;
Parser P;
if (argc==1)
{
R.readFromFile("c:\\ruleTest.txt");
}
else
{
R.readFromFile(argv[1]);
}
R.optimize();
//R.calculateLookAhead();
//R.print();
cout<<"\n now begin parsing...\n";
if (argc==1)
{
P.parseFile("c:\\sourceTest.txt");
}
else
{
P.parseFile(argv[2]);
}
return 0;
}
The input of grammar file is something like following:("c:\\ruleTest.txt")
M -> program i ; Dl B
Dl -> Dv Ml
Ml -> Ml Mo | e
Mo -> module i ( Vl ) Dv B
Dv -> variables Vl | e
Vl -> Vl V | V
V -> Il : T ;
T -> integer Ad | char Ad
Il -> i | Il , i
L -> i Ar
Ad -> e | array [ n ]
B -> begin Sl end ;
Sl -> S | S Sl
S -> L := E ; | if C then S else S | loop Sl end ; | exit ; | i ( Lp ) ; | B | read Ln ; | write Lo ; | e ;
Lp -> e | Ln
Ln -> L { , L }
Lo -> Lr { , Lr }
Lr -> i Ar | n | c
Ar -> e | [ E ]
E -> F { Oa F }
Oa -> + | -
F -> R { Om R }
Om -> * | /
R -> L | n | ( E ) | c
C -> E Or E
Or -> = | < | > | <= | >= | !=
The input of source code file is something like following:("c:\\sourceTest.txt")
program average; /* a sample program. It calculates the average of integers, */ /* the first value gives the number of integers to read*/ variables n: integer; ave, total: integer; module sum(n, res :integer;) /* sums up n integers from input, puts the result into res*/ variables i: integer; val: integer array[100]; begin i:=i; res :=0; loop read val[i]; res:= res+ val[i]; i := i+1; if i>n then exit; else ; end; /* loop */ end;/* module sum*/ begin /* main program*/ read n; sum(n, total); ave:=(((total/n))); write 'a', 'v','e','r','a','g','e','='; write ave; end; /* of program*/
Here is the look-ahead of output for original grammar:
Here is the look-ahead of output for modified grammar:
Here is the calculated First, Follow sets:
M ==> program i ; Dl B Dl ==> Dv Ml B ==> begin Sl end ; Dv ==> variables Vl | e Ml ==> e Ml0 Mo ==> module i ( Vl ) Dv B Vl ==> V Vl0 V ==> Il : T ; Il ==> i Il0 T ==> integer Ad | char Ad Ad ==> e | array [ n ] L ==> i Ar Ar ==> e | [ E ] Sl ==> S Sl0 S ==> i S0 | if C then S else S | loop Sl end ; | exit ; | begin Sl end ; | read Ln ; | write Lo ; | e ; E ==> F M0 C ==> F M0 Or E Lp ==> e | Ln Ln ==> i Ar M1 Lo ==> Lr M2 Lr ==> i Ar | n | c F ==> R M3 Oa ==> + | - R ==> i Ar | n | ( E ) | c Om ==> * | / Or ==> = | < | > | <= | >= | != M0 ==> + F M0 | e | - F M0 M1 ==> , L M1 | e M2 ==> , Lr M2 | e M3 ==> * R M3 | e | / R M3 Ml0 ==> module i ( Vl ) Dv B Ml0 | e Vl0 ==> i Il0 : T ; Vl0 | e Il0 ==> , i Il0 | e Sl0 ==> e | Sl S0 ==> Ar := E ; | ( Lp ) ; START ==> M $ name: M isNull: false isTerminal: false first: [0]=program follow: [0]=$ name: Dl isNull: true isTerminal: false first: [0]=e,[1]=variables,[2]=module follow: [0]=begin name: B isNull: false isTerminal: false first: [0]=begin follow: [0]=$,[1]=module,[2]=begin name: Dv isNull: true isTerminal: false first: [0]=variables,[1]=e follow: [0]=module,[1]=begin name: Ml isNull: true isTerminal: false first: [0]=e,[1]=module follow: [0]=begin name: Mo isNull: false isTerminal: false first: [0]=module follow: name: Vl isNull: false isTerminal: false first: [0]=i follow: [0]=module,[1]=begin,[2]=) name: V isNull: false isTerminal: false first: [0]=i follow: [0]=i,[1]=module,[2]=begin,[3]=) name: Il isNull: false isTerminal: false first: [0]=i follow: [0]=: name: T isNull: false isTerminal: false first: [0]=integer,[1]=char follow: [0]=; name: Ad isNull: true isTerminal: false first: [0]=e,[1]=array follow: [0]=; name: L isNull: false isTerminal: false first: [0]=i follow: [0]=,,[1]=;,[2]=) name: Ar isNull: true isTerminal: false first: [0]=e,[1]=[ follow: [0]=,,[1]=:=,[2]=;,[3]=),[4]=*,[5]=/,[6]=+,[7]=-,[8]=],[9]==,[10]=<,[11]=>,[12]=<=,[13]=> =,[14]=!=,[15]=then name: Sl isNull: false isTerminal: false first: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=; follow: [0]=end name: S isNull: false isTerminal: false first: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=; follow: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=;,[8]=end,[9]=else name: E isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=],[1]=),[2]=;,[3]=then name: C isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=then name: Lp isNull: true isTerminal: false first: [0]=e,[1]=i follow: [0]=) name: Ln isNull: false isTerminal: false first: [0]=i follow: [0]=;,[1]=) name: Lo isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=c follow: [0]=; name: Lr isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=c follow: [0]=,,[1]=; name: F isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=+,[1]=-,[2]=],[3]=),[4]=;,[5]==,[6]=<,[7]=>,[8]=<=,[9]=>=,[10]=!=,[11]=then name: Oa isNull: false isTerminal: false first: [0]=+,[1]=- follow: name: R isNull: false isTerminal: false first: [0]=i,[1]=n,[2]=(,[3]=c follow: [0]=*,[1]=/,[2]=+,[3]=-,[4]=],[5]=),[6]=;,[7]==,[8]=<,[9]=>,[10]=<=,[11]=>=,[12]=!=,[13]= then name: Om isNull: false isTerminal: false first: [0]=*,[1]=/ follow: name: Or isNull: false isTerminal: false first: [0]==,[1]=<,[2]=>,[3]=<=,[4]=>=,[5]=!= follow: [0]=i,[1]=n,[2]=(,[3]=c name: M0 isNull: true isTerminal: false first: [0]=+,[1]=e,[2]=- follow: [0]=],[1]=),[2]=;,[3]==,[4]=<,[5]=>,[6]=<=,[7]=>=,[8]=!=,[9]=then name: M1 isNull: true isTerminal: false first: [0]=,,[1]=e follow: [0]=;,[1]=) name: M2 isNull: true isTerminal: false first: [0]=,,[1]=e follow: [0]=; name: M3 isNull: true isTerminal: false first: [0]=*,[1]=e,[2]=/ follow: [0]=+,[1]=-,[2]=],[3]=),[4]=;,[5]==,[6]=<,[7]=>,[8]=<=,[9]=>=,[10]=!=,[11]=then name: Ml0 isNull: true isTerminal: false first: [0]=module,[1]=e follow: [0]=begin name: Vl0 isNull: true isTerminal: false first: [0]=i,[1]=e follow: [0]=module,[1]=begin,[2]=) name: Il0 isNull: true isTerminal: false first: [0]=,,[1]=e follow: [0]=: name: Sl0 isNull: true isTerminal: false first: [0]=e,[1]=i,[2]=if,[3]=loop,[4]=exit,[5]=begin,[6]=read,[7]=write,[8]=; follow: [0]=end name: S0 isNull: false isTerminal: false first: [0]=[,[1]=:=,[2]=( follow: [0]=i,[1]=if,[2]=loop,[3]=exit,[4]=begin,[5]=read,[6]=write,[7]=;,[8]=end,[9]=else name: START isNull: false isTerminal: false first: [0]=program follow: Press any key to continue
Here is the result: