构造与正则表达式对应的NFA

构造与正则表达式对应的NFA

接上一节
https://blog.csdn.net/greatyoulv/article/details/104186619

给你一个字符串 s 和一个字符规律 p,请你来实现一个支持 ‘.’ 和 ‘’ 的正则表达式匹配。
‘.’ 匹配任意单个字符
’ 匹配零个或多个前面的那一个元素
所谓匹配,是要涵盖 整个 字符串 s的,而不是部分字符串。

public class regexNFA {

	public static void main(String[] args) {
		// TODO Auto-generated method stub
		String txt = "abac";
		String regex = "aba*c";
		boolean result = isMatch(txt, regex);
		System.out.println(result);
	}
	
	public static boolean isMatch(String s, String p) {
        char[] sc = s.toCharArray();
        
        char[] pc = p.toCharArray();

        
        //图的状态数比正则字符数多1(起始状态)
        int m = pc.length + 1;
        
        boolean[][] graph = new boolean[m][m];
        for (int i = 0; i < m - 1; i++) {
        	//存放当前位
        	int lp = i;
        	// 当下一位为闭包操作(m-1是正则串的长度,pc[m-2]指向正则串的最后一位,< m-2指向倒数第二位)
        	if (i< m - 2 && pc[i + 1] == '*') {
        		// 进行双向链接
        		graph[lp][i + 1] = true;
                graph[i + 1][lp] = true;
        	}
        	// 特定符号直接链接下一位
            if (pc[i] == '*') {
                graph[i][i + 1] = true;
            }
        }
        
        // 打印图
        for (int i = 0; i < m; i++) {
        	for (int j = 0; j < m; j++) {
        		System.out.print(graph[i][j] + " ");
        	}
        	System.out.println();
        }
        
        // 深度搜索第一个顶点的可达顶点
        boolean[] mark = dfs(graph, 0);
        
        // 打印访问标记
        for (int i = 0; i < mark.length; i++) {
        	System.out.print(mark[i] + "-");
        }
        System.out.println();
        
        // 遍历字符串
        for (int i = 0; i < sc.length; i++) {
            boolean[] match = new boolean[m];
            for (int j = 0; j < m; j++) {
            	// 标记当前访问点
                if (mark[j]) {
                	// 标记最后一位
                    if (j == m - 1) {
                        continue;
                    }
                    // 当前标志访问点的值点为(".")或与给定字符串相匹配的第一个字符
                    if (pc[j] == sc[i] || pc[j] == '.') {
                    	// 无需读入字符直接状态转移到下一位
                        match[j + 1] = true;
                    }
                }
            }

            // 对状态转移到的下一位进行深度搜索
            mark = dfs(graph, match);
            
            // 打印访问标记
            System.out.println(i);
            for (int k = 0; k < mark.length; k++) {
            	System.out.print(mark[k] + "--");
            }
            System.out.println();
            
        }

        return mark[m - 1];
    }

    private static boolean[] dfs(boolean[][] graph, int s) {
        boolean[] re = new boolean[graph.length];
        doDfs(graph, re, s);

        return re;
    }

    private static boolean[] dfs(boolean[][] graph, boolean[] mark) {
        boolean[] re = new boolean[graph.length];
        for (int i = 0; i < mark.length; i++) {
            if (mark[i]) {
                doDfs(graph, re, i);
            }
        }

        return re;
    }

    private static void doDfs(boolean[][] graph, boolean[] mark, int v) {
        mark[v] = true;
        for (int i = 0; i < graph.length; i++) {
            if (!mark[i] && graph[v][i]) {
                doDfs(graph, mark, i);
            }
        }
    }

}

参考链接:
https://blog.csdn.net/weixin_41427400/article/details/80958777

https://leetcode-cn.com/problems/regular-expression-matching/solution/yi-bu-dao-wei-zhi-jie-an-zheng-ze-biao-da-shi-de-s/

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
### 回答1: 正则表达式转换为NFA(非确定有限自动机)可以通过构造Thompson算法实现。代码示例如下: ```python # 定义NFA状态和边的类 class NFAState: def __init__(self, label=None): self.label = label self.transitions = [] # 定义NFA类 class NFA: def __init__(self, start_state, accept_states): self.start_state = start_state self.accept_states = accept_states def add_transition(self, state1, input, state2): state1.transitions.append((input, state2)) # 正则表达式NFA的函数 def regex_to_nfa(regex): stack = [] for char in regex: if char == '*': # 闭包操作 nfa = stack.pop() accept_state = NFAState() nfa.add_transition(accept_state, None, nfa.start_state) nfa.add_transition(accept_state, None, accept_state) stack.append(NFA(accept_state, [accept_state])) elif char == '|': # 或操作 nfa2 = stack.pop() nfa1 = stack.pop() start_state = NFAState() accept_state = NFAState() start_state.transitions.append((None, nfa1.start_state)) start_state.transitions.append((None, nfa2.start_state)) nfa1.accept_states[0].transitions.append((None, accept_state)) nfa2.accept_states[0].transitions.append((None, accept_state)) stack.append(NFA(start_state, [accept_state])) elif char == '.': # 连接操作 nfa2 = stack.pop() nfa1 = stack.pop() nfa1.accept_states[0].transitions.append((None, nfa2.start_state)) stack.append(NFA(nfa1.start_state, nfa2.accept_states)) else: # 创建单个字符的NFA accept_state = NFAState() start_state = NFAState() start_state.transitions.append((char, accept_state)) stack.append(NFA(start_state, [accept_state])) return stack.pop() ``` NFA转换为DFA可以使用子集构造算法实现。代码示例如下: ```python # 定义DFA状态和边的类 class DFAState: def __init__(self, label=None): self.label = label self.transitions = {} # 定义DFA类 class DFA: def __init__(self, start_state, accept_states): self.start_state = start_state self.accept_states = accept_states def add_transition(self, state1, input, state2): state1.transitions[input] = state2 # NFA转DFA的函数 def nfa_to_dfa(nfa): start_state = DFAState(nfa.start_state.label) dfa_states = [start_state] unmarked_states = [start_state] while unmarked_states: dfa_state = unmarked_states.pop(0) transitions = {} for nfa_state in get_nfa_states(dfa_state, nfa): for transition in nfa_state.transitions: input_symbol = transition[0] next_nfa_state = transition[1] if input_symbol not in transitions: transitions[input_symbol] = set() transitions[input_symbol].add(next_nfa_state) for input_symbol, next_state_states in transitions.items(): next_state_label = ",".join(sorted([n.label for n in next_state_states])) next_state = get_or_create_dfa_state(next_state_label, dfa_states) dfa_state.transitions[input_symbol] = next_state if next_state not in dfa_states: dfa_states.append(next_state) unmarked_states.append(next_state) accept_states = [s for s in dfa_states if nfa.accept_states[0].label in s.label.split(",")] return DFA(start_state, accept_states) # 获取NFA状态的ε闭包 def get_nfa_states(dfa_state, nfa): nfa_states = [] def get_nfa_states_recursive(nfa_state): nfa_states.append(nfa_state) for transition in nfa_state.transitions: input_symbol = transition[0] next_nfa_state = transition[1] if input_symbol is None and next_nfa_state not in nfa_states: get_nfa_states_recursive(next_nfa_state) for nfa_state_label in dfa_state.label.split(","): nfa_state = get_nfa_state_by_label(nfa_state_label, nfa) get_nfa_states_recursive(nfa_state) return nfa_states # 根据NFA状态标签获取NFA状态 def get_nfa_state_by_label(label, nfa): for state in nfa.accept_states: if state.label == label: return state if nfa.start_state.label == label: return nfa.start_state # 根据DFA状态标签获取DFA状态,如果不存在则创建 def get_or_create_dfa_state(label, dfa_states): for state in dfa_states: if state.label == label: return state return DFAState(label) ``` 以上就是将正则表达式转换为NFA,以及将NFA转换为DFA的代码示例。 ### 回答2: 正则表达式NFA主要包括两个步骤:正则表达式转后缀表达式和后缀表达式转NFA。 首先,将给定的正则表达式转换为后缀表达式。可以通过使用栈和运算符优先级来实现。遍历正则表达式的每个字符,如果是操作数,则直接输出到后缀表达式。如果是运算符,则根据优先级进行相应的操作,将栈中优先级大于或等于当前运算符的运算符输出到后缀表达式,再将当前运算符压入栈。当所有字符都被处理完后,将栈中剩余的运算符依次输出到后缀表达式中。 然后,根据后缀表达式构建对应NFA。可以使用Thompson算法来实现此过程。首先,创建一个空的NFA栈。然后,遍历后缀表达式的每个字符。如果是操作符,如'a'、'b',则创建一个新的NFA,其中有两个状态,一个初始状态和一个接受状态,通过一条连接状态的边进行连接,并将该NFA压入NFA栈。如果是运算符,如'|'、'.'、'*',则从NFA栈中弹出对应NFA,并根据运算符创建新的NFA,并将该NFA压入NFA栈。 最后,将得到的NFA转换为DFA。可以使用子集构造算法来实现此过程。首先,将NFA的初始状态作为DFA的初始状态,并计算该状态的ε-闭包。然后,将ε-闭包作为DFA的一个状态,如果该状态中包含NFA的接受状态,则将该状态标记为接受状态。接着,对于每个输入符号,计算该输入符号在当前状态下,通过ε-闭包能够到达的NFA状态,并将其作为DFA的一个新状态。重复以上步骤,直到所有的DFA状态都被生成。最终得到的DFA即为所求。 以上是正则表达式转换为NFA再转换为DFA的基本过程。可以根据具体的编程语言和数据结构进行具体的实现。 ### 回答3: 正则表达式NFA(Nondeterministic Finite Automaton)的过程可以通过使用Thompson算法来实现,以下是一个简单的Python代码示例: ```python class State: def __init__(self, label=None): self.label = label self.edges = [] class NFA: def __init__(self, start=None, end=None): self.start = start self.end = end def regex_to_nfa(regex): stack = [] for char in regex: if char == '.': nfa2 = stack.pop() nfa1 = stack.pop() nfa1.end.edges.append(nfa2.start) stack.append(NFA(nfa1.start, nfa2.end)) elif char == '|': nfa2 = stack.pop() nfa1 = stack.pop() start = State() start.edges.extend([nfa1.start, nfa2.start]) end = State() nfa1.end.edges.append(end) nfa2.end.edges.append(end) stack.append(NFA(start, end)) elif char == '*': nfa = stack.pop() start = State() end = State() start.edges.extend([nfa.start, end]) nfa.end.edges.extend([nfa.start, end]) stack.append(NFA(start, end)) else: start = State() end = State() start.edges.append(end) stack.append(NFA(start, end)) return stack.pop() def nfa_to_dfa(nfa): dfa_start = State() dfa = NFA(dfa_start) dfa_states = [dfa_start] state_map = {} state_queue = [dfa_start] while len(state_queue) > 0: current_state = state_queue.pop(0) state_map[current_state] = {} for char in nfa.alphabet: new_state = State() state_map[current_state][char] = new_state for nfa_state in current_state: if nfa_state.label == char: new_state.append(nfa_state.edges) for edge in nfa_state.edges: if edge not in dfa_states: state_queue.append(edge) dfa_states.append(edge) return dfa regex = "(ab)*c" nfa = regex_to_nfa(regex) dfa = nfa_to_dfa(nfa) ``` 以上代码实现了将正则表达式转化为NFA,以及将NFA转化为DFA的过程。在这个示例中,我们使用Thompson算法将正则表达式转换为NFA,并使用子集构造法将NFA转换为DFA。最终得到的DFA可以用于模式匹配和字符串匹配等应用。该示例代码仅为简化版本,实际实现中可能会有更多的细节和优化。

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值