Python实现SLR（1）语法分析器，编译原理yyds！

Python实现SLR（1）语法分析器

实验课前一天晚上肝了个SLR语法分析器，当时还发朋友圈语法分析器和我晚上总得走一个，从第二天状态来看，应该是我们俩一起走了（笑

• 编写的时间比较仓促，所以代码有些地方实现不是很好，存在一些问题，以后有时间的话再来修补一下，比如在对两个项目规范族进行比较时效率比较低，first集和follow集中对连续多个非终结符推到ε的情况下可能会有bug，但在我的文法定义中特意绕开了ε，勉强能用。
• 为了方便代码阅读，加入了许多注释后的打印语句，将这些打印语句取消注释后运行，能够告诉你当前这段代码所做的事。
• 在编写过程中，尽量对模块进行封装，把逻辑和文法定义分开，做到文法可以方便地修改，但时间过于仓促可能还是有点小问题。当文法修改时，修改

  getCol函数（该函数将终结符和非终结符映射到action和goto表中相应的列），initProduction函数（该函数定义了文法产生式（拓广文法），在本文中有28个产生式），source（输入单词序列），varset（非终结符集合），terminalset（终结符集合）
  

SLR（1）分析流程

• 输入文法
• 求first集
• 求follow集
• 构造LR（0）项目集DFA
• 构造Action和Goto
• 按照Action和Goto进行分析

完整代码附在文章末尾

1.主要数据结构定义和基础函数：

 基础函数

1. isVariable函数判断是不是非终结符

2. isTerminal函数判断是不是终结

3. transf(production_set, var)函数 production_set为一个LR（0）项目，尝试通过var（终结符或非终结符）进行转移

4. isSameStatus（status1, status2）函数：判断status1和status2是不是两个相同的LR（0）项目

5. isInPointset(production_set, pointset):#用来检验production_set是不是已经存在的point ，如果存在就把point返回（生成DFA时用到）

数据结构

1. 产生式采用类来存储，left和right分别为list，number‘为产生式编号
2. GraphPoint存储DFA转移，transfer为有向边集合，集合中的一个元素由var（终结符或非终结符），和另一个GraphPoint组成

class Production:
    def __init__(self, left, right, number):
        self.left = left
        self.right = right
        self.number = number

class GraphPoint:

    def __init__(self, begin_production, id):
        self.status = begin_production
        self.transfer = []
        self.id = id

    def add_transfer(self, var, graphPoint):
        self.transfer.append([var, graphPoint])

2.文法定义

1.分析目标代码：int lexicalanalysis(){  float a; int b; a=1.1; b=2; while(b<100){ b=b+1; a=a+3;}; if(a>5) {b=b-1;} else {b=b+1;}}

2.语法分析器输入为目标代码的词法分析器输出的单词序列

source = [[5, "int", " 关键字"], [1, "lexicalanalysis", " 标识符"], [13, "(", " 左括号"], [14, ")", " 右括号"], [20, "{", " 左大括号"],
          [4, "float", " 关键字"], [1, "a", " 标识符"], [15, ";", " 分号"], [5, "int", " 关键字"], [1, "b", " 标识符"],
          [15, ";", " 分号"], [1, "a", " 标识符"], [12, "=", " 赋值号"], [3, "1.1", " 浮点数"], [15, ";", " 分号"], [1, "b", " 标识符"],
          [12, "=", " 赋值号"], [2, "2", " 整数"], [15, ";", " 分号"], [8, "while", "  关键字"], [13, "(", " 左括号"],
          [1, "b", " 标识符"], [17, "<", " 小于号"], [2, "100", " 整数"], [14, ")", " 右括号"], [20, "{", " 左大括号"],
          [1, "b", " 标识符"], [12, "=", " 赋值号"], [1, "b", " 标识符"], [9, "+", " 加 号"], [2, "1", " 整数"], [15, ";", " 分号"],
          [1, "a", " 标识符"], [12, "=", " 赋值号"], [1, "a", " 标识符"], [9, "+", " 加号"], [2, "3", " 整数"], [15, ";", " 分号"],
          [21, "}", " 右大括号"], [15, ";", " 分号"], [6, "if", " 关键字"], [13, "(", " 左括号"], [1, "a", " 标识符"],
          [16, ">", " 大于号"], [2, "5", " 整数"], [14, ")", " 右括号"], [20, "{", " 左大括号"], [1, "b", " 标识符"],
          [12, "=", " 赋值号"], [1, "b", " 标识符"], [10, "-", " 减号"], [2, "1", " 整数"], [15, ";", " 分号"], [21, "}", " 右大括号"],
          [7, "else", " 关键字"], [20, "{", " 左大括号"], [1, "b", " 标识符"], [12, "=", " 赋值号"], [1, "b", " 标识符"],
          [9, "+", " 加号"], [2, "1", " 整数"], [15, ";", " 分号"], [21, "}", " 右大括号"], [21, "}", " 右大括号"]]

3.文法定义：拓广文法共有28个产生式，0号产生式为保证分析器只有一个接受状态，而拓广的产生式。

def initProduction():
    production_list = []
    production = Production(["A1"], ["A"], 0)
    production_list.append(production)
    production = Production(["A"], ["E", "I", "(", ")", "{", "D", "}"], 1)
    production_list.append(production)
    production = Production(["E"], ["int"], 2)
    production_list.append(production)
    production = Production(["E"], ["float"], 3)
    production_list.append(production)
    production = Production(["D"], ["D", ";", "B"], 4)
    production_list.append(production)
    production = Production(["B"], ["F"], 5)
    production_list.append(production)
    production = Production(["B"], ["G"], 6)
    production_list.append(production)
    production = Production(["B"], ["M"], 7)
    production_list.append(production)
    production = Production(["F"], ["E", "I"], 8)
    production_list.append(production)
    production = Production(["G"], ["I", "=", "P"], 9)
    production_list.append(production)
    production = Production(["P"], ["K"], 10)
    production_list.append(production)
    production = Production(["P"], ["K", "+", "P"], 11)
    production_list.append(production)
    production = Production(["P"], ["K", "-", "P"], 12)
    production_list.append(production)
    production = Production(["I"], ["id"], 13)
    production_list.append(production)
    production = Production(["K"], ["I"], 14)
    production_list.append(production)
    production = Production(["K"], ["number"], 15)
    production_list.append(production)
    production = Production(["K"], ["floating"], 16)
    production_list.append(production)
    production = Production(["M"], ["while", "(", "T", ")", "{", "D", ";", "}"], 18)
    production_list.append(production)
    production = Production(["N"], ["if", "(", "T", ")", "{", "D",";", "}", "else", "{", "D", ";","}"], 19)
    production_list.append(production)
    production = Production(["T"], ["K", "L", "K"], 20)
    production_list.append(production)
    production = Production(["L"], [">"], 21)
    production_list.append(production)
    production = Production(["L"], ["<"], 22)
    production_list.append(production)
    production = Production(["L"], [">="], 23)
    production_list.append(production)
    production = Production(["L"], ["<="], 24)
    production_list.append(production)
    production = Production(["L"], ["=="], 25)
    production_list.append(production)
    production = Production(["D"], ["B"], 26)
    production_list.append(production)
    production = Production(["B"], ["N"], 27)
    production_list.append(production)
    return production_list

3.求First集

根据此算法即可求解first集，第8，9步可以采用递归的方式进行求解。

def getFirst(production_list, varset, terminalset):
    first_dic = {}
    # 用来标记first集是否计算完毕，防止重复计算浪费时间
    done = {}
    for var in varset:
        first_dic[var] = set()
        done[var] = 0
    # 所有终结符的first集是他自身
    for var in terminalset:
        first_dic[var] = {var}
        done[var] = 1
    # print("初始化后的done",done)
    # print("初始化的first_dic",first_dic)
    for var in varset:
        if done[var] == 0:
            # print("计算",var)
            getFirstForVar(var, first_dic, varset, terminalset, done)
            # print("计算完毕",var)
            # print("此时的done", done)
            # print("此时的first_dic", first_dic)
        else:
            pass
    return first_dic


def getFirstForVar(var, first_dic, varset, terminalset, done):
    # 已经推导过直接结束

    if done[var] == 1:
        # print("我已经推导过了吼")
        return

    # 对非终结符求first集合,先看右边第一个元素为终结符
    for production in production_list:
        if var in production.left:
            if isTerminal(production.right[0], terminalset):
                first_dic[var].add(production.right[0])
                # 用null表示空字符
            if production.right[0] == "null":
                # print("出现右侧为空")
                first_dic[var].add("null")
    # 右边第一个元素为非终结符
    for production in production_list:
        if var in production.left:
            if isVariable(production.right[0], varset):
                if var == production.right[0]:
                    continue
                if done[production.right[0]] == 0:
                    getFirstForVar(production.right[0], first_dic, varset, terminalset, done)
                if "null" in first_dic[production.right[0]]:
                    first_dic[production.right[0]].remove("null")
                first_dic[var] = first_dic[var] | first_dic[production.right[0]]
                # print("将 ",production.right[0],"的集合 ",first_dic[production.right[0]],"并入",var,"的集合中",first_dic[var],"中","得到",)
            if isVariable(production.right[0], varset) and len(production.right) > 1:

                index = 1
                count = 1
                while isVariable(production.right[index], varset):
                    index = index + 1
                    count = c