自己动手写C语言编译器（1）

直接上代码（表达式树在此初具模型）

由于没有做前期的“词法分析”，

1.支持单个字符形式的变量

2.支持单个字符形式的操作符

3.支持优先级。

不支持

1.不支持多个字符的操作符和变量。

2. 不支持左右结合性

3.不支持函数

4. 不支持语句

5. 不支持关键字。

这些将陆续添加。

// MyCompiler.cpp : Defines the entry point for the console application.
//
#include "stdafx.h"
#include <fstream>
#include <list>
#include <map>
#include <string>
#include <iostream>
#include <vector>
//把Token转换为字符串
#define  ATokenToString(l, r) if(l == r)return #r;
using namespace std;
namespace tokens{
enum TOKEN
{
INVALID_TYPE,
WHITE_SPACE,
NAME,
NUMBER,
END,
PLUS,
MINUS,
MUL,
DIV,
PRINT,
ASSIGN,
LP,
RP,
};
static string MyTokenToString(TOKEN token)
{
ATokenToString(token, WHITE_SPACE)
ATokenToString(token, NAME)
ATokenToString(token, NUMBER)
ATokenToString(token, END)
ATokenToString(token, PLUS)
ATokenToString(token, MINUS)
ATokenToString(token, MUL)
ATokenToString(token, DIV)
ATokenToString(token, PRINT)
ATokenToString(token, ASSIGN)
ATokenToString(token, LP)
ATokenToString(token, RP)
return "";
}
}
class tokenizer
{
int mTokenTypes[256];
string mCurrentStrToken;
tokens::TOKEN mCurrentTokenType;
istream& mIfStream;
public:
tokenizer(istream& infile):mIfStream(infile)
{
memset(mTokenTypes, 0, 256);
mTokenTypes['+'] = tokens::PLUS;
mTokenTypes['-'] = tokens::MINUS;
mTokenTypes['*'] = tokens::MUL;
mTokenTypes['/'] = tokens::DIV;
mTokenTypes[';'] = tokens::PRINT;
mTokenTypes['='] = tokens::ASSIGN;
mTokenTypes['('] = tokens::LP;
mTokenTypes[')'] = tokens::RP;
mTokenTypes['\t'] = tokens::WHITE_SPACE;
mTokenTypes[' '] = tokens::WHITE_SPACE;
char ch = 0;
for (ch = 'a'; ch <= 'z'; ch ++)
{
mTokenTypes[ch] = tokens::NAME;
}
for(ch = 'A'; ch <= 'Z'; ch ++)
{
mTokenTypes[ch] = tokens::NAME;
}
for(ch = '0'; ch <= '9'; ch ++)
{
mTokenTypes[ch] = tokens::NUMBER;
}
mTokenTypes['.'] = tokens::NUMBER;
}
bool hasNext()
{
char temptoken = 0;
if(mIfStream.get(temptoken))
{
mIfStream.putback(temptoken);
return true;
}
return false;
}
const string getCurrent()
{
return mCurrentStrToken;
}
tokens::TOKEN getTokenType()
{
return mCurrentTokenType;
}
const string getNext()
{
string strValue;
char ch = 0;
mIfStream>>ch;
mCurrentTokenType = (tokens::TOKEN)mTokenTypes[ch];
mIfStream.putback(ch);
mIfStream>>strValue;
mCurrentStrToken = strValue;
return strValue;
}
};
class Node;
struct NodeVisitor
{
virtual void BeginVisitTree(Node* nod) = 0;
virtual void VisitNode(Node* nod) = 0;
virtual void EndVisitTree(Node* nod) = 0;
};
class Node
{
char m_Data;
NodeVisitor* m_pGetNode;
public:
Node* m_pLeftChild;
Node* m_pRightChild;
Node* m_pParent;
int m_priority_level;
public:
char getData()
{
return m_Data;
}
Node()
{
m_Data = 0;
m_pLeftChild = NULL;
m_pRightChild = NULL;
m_pParent = NULL;
//空的Node，优先级最低
m_priority_level = 0;
}
Node(char data)
{
m_Data = data;
m_pLeftChild = NULL;
m_pRightChild = NULL;
m_pParent = NULL;
m_priority_level = 0;
}
void set(Node* pL, Node* pR)
{
m_pLeftChild = pL;
m_pRightChild = pR;
}
void setLeft(Node* pLeft)
{
m_pLeftChild = pLeft;
}
void setRight(Node* pRight)
{
m_pRightChild = pRight;
}
void accept(NodeVisitor* pGetIt)
{
m_pGetNode = pGetIt;
}
void Postorder_Traversal(NodeVisitor* pVisiter)
{
if(m_pLeftChild != NULL)
m_pLeftChild->Postorder_Traversal(pVisiter);
if (m_pRightChild != NULL)
m_pRightChild->Postorder_Traversal(pVisiter);
pVisiter->VisitNode(this);
}
void Inorder_Traversal(NodeVisitor* pVisiter)
{
//Begin Tree.
pVisiter->BeginVisitTree(this);
if(m_pLeftChild != NULL)m_pLeftChild->Inorder_Traversal(pVisiter);
//Visit this node.
//Node should not be diffrent from tree.
pVisiter->VisitNode(this);
if (m_pRightChild != NULL)m_pRightChild->Inorder_Traversal(pVisiter);
//End Tree.
pVisiter->EndVisitTree(this);
}
};
class ExpressionVisiter: public NodeVisitor
{
public:
void BeginVisitTree(Node* nod)
{
if (nod->m_pLeftChild != NULL && nod->m_pRightChild != NULL)
{
printf("%c", '(');
}
}
void VisitNode(Node* nod)
{
printf("%c", nod->getData());
}
void EndVisitTree(Node* nod)
{
if (nod->m_pLeftChild != NULL && nod->m_pRightChild != NULL)
{
printf("%c", ')');
}
}
};
class Tree
{
public:
Tree()
{
pHead = new Node();
}
Node* pHead;
~Tree()
{
}
Node* getNext()
{
}
void Postorder_Traversal(NodeVisitor* pVisiter)
{
if(pHead->m_pLeftChild!=NULL)
pHead->m_pLeftChild->Postorder_Traversal(pVisiter);
}
void Inorder_Traversal(NodeVisitor* pVisiter)
{
if(pHead->m_pLeftChild!=NULL)
pHead->m_pLeftChild->Inorder_Traversal(pVisiter);
}
Node* getCurrent()
{
}
Node* getFirst()
{
}
bool bHasNext()
{
}
void setValue(Node* pNode)
{
pHead->m_pLeftChild = pNode;
}
};
//打印一个字符串的所有排列。
void printSequence(char* pChara)
{
int i = 0;
static int MAX_LEN = strlen(pChara);
if (pChara[0] == '\0')
{
for (int i = 0; i < MAX_LEN; i ++)
{
printf("%c", pChara[i + 1]);
}
printf("\n", "");
}
char* tempChar = pChara;
while (*tempChar != '\0' )
{
char* pMyString = new char[MAX_LEN + 1];
memcpy(pMyString, pChara, MAX_LEN + 1);
memcpy(pMyString + i, pMyString + i + 1, MAX_LEN - i);
pMyString[MAX_LEN] = *tempChar;
printSequence(pMyString);
tempChar ++;
i ++;
delete[] pMyString;
}
}
#ifdef TEST
int main(int argc, char* argv[])
{
ExpressionVisiter visister;
Tree* pTree = new Tree();
Node* pNode1 = new Node('-');
Node* pNode11 = new Node('+');
Node* pNode12 = new Node('/');
Node* pNode111 = new Node('a');
Node* pNode112 = new Node('*');
Node* pNode121 = new Node('e');
Node* pNode122 = new Node('f');
Node* pNode1121 = new Node('b');
Node* pNode1122 = new Node('-');
Node* pNode11221 = new Node('c');
Node* pNode11222 = new Node('d');
pNode1->set(pNode11, pNode12);
pNode11->set(pNode111, pNode112);
pNode12->set(pNode121, pNode122);
pNode112->set(pNode1121, pNode1122);
pNode1122->set(pNode11221, pNode11222);
pTree->setValue(pNode1);
pTree->Inorder_Traversal(&visister);
//printSequence("a");
system("PAUSE");
return 1;
}
#else
//字符的类型
class SH_CHARACTER_TYPE
{
public:
enum CHARACTER_TYPE
{
ILLIGAL_CHAR = 0x00000000,
DIGITAL_CHAR = 0x00000001,
NON_DIGITAL_CHAR = 0x00000002,
UNIVERSAL_CHAR = 0x00000004,
INTEGER_CONSTANT_CHAR = 0x00000008,
FLOATING_CONSTANT_CHAR = 0x00000010,
ENUMERATION_CONSTANT_CHAR = 0x00000020,
CHARACTER_CONSTANT_CHAR = 0x00000040,
STRING_LITERAL_CHAR =  0x00000080,
PUNCTUATOR_CHAR = 0x00000100,
EOL_CHAR = 0x00000200,
EOF_CHAR = 0x00000400,
WHITESPACE_CHAR = 0x00000800,
OPERATOR_CHAR = 0x00001000,
};
};
class sh_ctype
{
private:
int ctypeTable[256];
public:
void clear()
{
//清除
for (int i = 0; i < 256;  ++ i)
{
ctypeTable[i] = SH_CHARACTER_TYPE::ILLIGAL_CHAR;
}
}
sh_ctype()
{
clear();
setCtype('a', 'z', SH_CHARACTER_TYPE::NON_DIGITAL_CHAR);
setCtype('A', 'Z', SH_CHARACTER_TYPE::NON_DIGITAL_CHAR);
setCtype('0', '9', SH_CHARACTER_TYPE::DIGITAL_CHAR);
setCtype('0', '9', SH_CHARACTER_TYPE::INTEGER_CONSTANT_CHAR);
setCtype('x', SH_CHARACTER_TYPE::INTEGER_CONSTANT_CHAR);
setCtype('X', SH_CHARACTER_TYPE::INTEGER_CONSTANT_CHAR);
setCtype('u', SH_CHARACTER_TYPE::INTEGER_CONSTANT_CHAR);
setCtype('U', SH_CHARACTER_TYPE::INTEGER_CONSTANT_CHAR);
setCtype('l', SH_CHARACTER_TYPE::INTEGER_CONSTANT_CHAR);
setCtype('L', SH_CHARACTER_TYPE::INTEGER_CONSTANT_CHAR);
setCtype('0', '9', SH_CHARACTER_TYPE::FLOATING_CONSTANT_CHAR);
setCtype('e', SH_CHARACTER_TYPE::FLOATING_CONSTANT_CHAR);
setCtype('E', SH_CHARACTER_TYPE::FLOATING_CONSTANT_CHAR);
setCtype('l', SH_CHARACTER_TYPE::FLOATING_CONSTANT_CHAR);
setCtype('L', SH_CHARACTER_TYPE::FLOATING_CONSTANT_CHAR);
setCtype('f', SH_CHARACTER_TYPE::FLOATING_CONSTANT_CHAR);
setCtype('F', SH_CHARACTER_TYPE::FLOATING_CONSTANT_CHAR);
//标点符号字符
setCtype(0x21,0x2F, SH_CHARACTER_TYPE::PUNCTUATOR_CHAR);
setCtype(0x3a,0x40, SH_CHARACTER_TYPE::PUNCTUATOR_CHAR);
setCtype(0x5B,0x60, SH_CHARACTER_TYPE::PUNCTUATOR_CHAR);
//空白字符
setCtype(0x20, SH_CHARACTER_TYPE::WHITESPACE_CHAR);
setCtype(0x09, 0x0D, SH_CHARACTER_TYPE::WHITESPACE_CHAR);
setCtype('+', SH_CHARACTER_TYPE::OPERATOR_CHAR);
}
//断定是否是某种类型
bool assert(char ch, SH_CHARACTER_TYPE::CHARACTER_TYPE type)
{
if ((getCtype(ch)&type) == type)
{
return true;
}
return false;
}
void setCtype(char from, char to, SH_CHARACTER_TYPE::CHARACTER_TYPE type)
{
for (char ch = from; ch <= to ; ++ ch)
{
ctypeTable[ch] = ctypeTable[ch]|type;
}
}
void setCtype(char ch, SH_CHARACTER_TYPE::CHARACTER_TYPE type)
{
ctypeTable[ch] = ctypeTable[ch]|type;
}
int getCtype(char ch)
{
return ctypeTable[ch];
}
};
int getLevel(char ch)
{
switch(ch)
{
case ',':
return 0;
case '=':
return 1;
case '|':
return 2;
case '^':
return 3;
case '>': case '<':
return 5;
case '+' : case '-':
return 6;
case '*': case '/':
return 7;
case '&':
return 8;
case '.':
return 9;
default:
return INT_MAX;
}
}
static int const INVALID = 0;
static int const EXPRESSION = 0x0001;
static int const LVALUE = 0x0002;
static int const OPERATOR_LEFT = 0x0004;
static int const OPERATOR_RIGHT = 0x0008;
int main(int argc, char* argv[])
{
sh_ctype ct;
Node* pHead = NULL;
Node* preTreeOperator = NULL;
char ch;
while (cin.get(ch) && ch != '\n')
{
//如果目前没有操作子Tree
if (preTreeOperator == NULL)
{
preTreeOperator = new Node(ch);
preTreeOperator->m_priority_level = getLevel(ch);
pHead = preTreeOperator;
continue;
}
else
{
Node* no = new Node(ch);
no->m_priority_level = getLevel(ch);
Node* theTree = preTreeOperator;
if(pHead->m_priority_level == no->m_priority_level
&& pHead->m_priority_level == INT_MAX)
{
std::cout<<"Unkown symbol "<<"\""<<ch<<"\"";
system("PAUSE");
return;
}
else if (no->m_priority_level <= pHead->m_priority_level)
{
pHead->m_pParent = no;
no->m_pLeftChild = pHead;
pHead = no;
preTreeOperator = no;
continue;
}
if(theTree->m_priority_level == no->m_priority_level
&& theTree->m_priority_level == INT_MAX)
{
std::cout<<"Unkown symbol "<<"\""<<ch<<"\"";
system("PAUSE");
return;
}
while (theTree!= NULL&& theTree->m_priority_level >= no->m_priority_level)
{
theTree = theTree->m_pParent;
}
if (theTree != NULL)
{
no->m_pLeftChild = theTree->m_pRightChild;
no->m_pParent = theTree;
theTree->m_pRightChild = no;
preTreeOperator = no;
}
}
}
Tree tree;
ExpressionVisiter vister;
tree.setValue(pHead);
tree.Inorder_Traversal(&vister);
system("PAUSE");
}
#endif

自己动手写C语言编译器（1）相关推荐

自己动手写C语言编译器（5）
依托Bison强大的工具, 写编译器,先用编译原理搞个计算器玩玩.如下: Bison工具: http://coolshell.cn/articles/1547.html http://www.gnu. ...
自己动手写C语言编译器（4）
Statement由Expression构成,Expression由Token构成,Token由char构成. 从上到下呈现树形结构. 程序是由statement组成的,其实我们要的就是一种判断Sta ...
自己动手写C语言编译器（暂停）
1.开源的东西其实并不好用. 2.需要更多的关注Android开发和linux的工具的使用. 3.llvm编译出错的问题一直没有解决. 4.也许需要亲自与llvm的官方进行沟通.这是一个庞大的任务.
自己动手写C语言编译器（3）
词法分析器部分完成. 支持: 1.支持单词分割 2.支持数字类型 3.支持字符串 4.支持换行 6.支持注释不支持: 1.不支持关键字 2.不支持变量. 3.不支持关键字. 4.不支操作符. 偶没有 ...
自己动手写C语言编译器（2）
直接上代码 : 支持:左右结合性 // MyCompiler.cpp : Defines the entry point for the console application. // #includ ...
java c语言 for_Java能写C语言编译器吗
用java是可以写语言编译器的,用任何语言工具写都可以.各语言开发工具间只有开发效率的差异,没有可行不可行的区别. 编译器其实就是一个翻译工具,它可以把代码直接翻译成二进制文件交给CPU执行(二进制指 ...
自己动手制作C 语言编译器（8）：表达式
这是整个编译器的最后一部分,解析表达式.什么是表达式?表达式是将各种语言要素的一个组合,用来求值.例如:函数调用.变量赋值.运算符运算等等. 表达式的解析难点有二:一是运算符的优先级问题,二是如何将表 ...
自己动手制作C 语言编译器（7）：语句
整个编译器还剩下最后两个部分:语句和表达式的解析.它们的内容比较多,主要涉及如何将语句和表达式编译成汇编代码.这章讲解语句的解析,相对于表达式来说它还是较为容易的. 语句 C 语言区分"语句 ...
keil c语言编译运行,Keil的c语言编译器
我曾经通过查看反汇编代码对KEILC编译器进行了测试,大概有这么一下内容,也得出一些结论. (1)全局变量:如果程序中定义了全局变量,而且初始值不是0.此时,在程序调到main()函数执行前,除了要进 ...

自己动手写C语言编译器（1）

自己动手写C语言编译器（1）相关推荐

最新文章

热门文章