[copy]正在修复语法问题

CMakeLists.txt

@@ -2,7 +2,7 @@ cmake_minimum_required(VERSION 3.10)
 project(compiler-bin)
 
 # 收集所有的cpp源文件
-file(GLOB SOURCES_LL "LL1/*.cpp")
+file(GLOB SOURCES_LL1 "LL1/src/*.cpp")
 file(GLOB SOURCES_NFA "nfa/src/*.cpp")
 file(GLOB SOURCES_MAIN "main/*.cpp")
 
@@ -10,15 +10,15 @@ file(GLOB SOURCES_MAIN "main/*.cpp")
 # set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/bin)
 
 # 创建静态链接库
-add_library(LL STATIC ${SOURCES_LL})
+add_library(LL1 STATIC ${SOURCES_LL1})
 add_library(nfa STATIC ${SOURCES_NFA})
 
 # 添加头文件目录
-target_include_directories(LL PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/LL1)
+target_include_directories(LL1 PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/LL1/include)
 target_include_directories(nfa PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/nfa/include)
 
 # 添加可执行文件
 add_executable(main ${SOURCES_MAIN})
 
 # 链接静态库
-target_link_libraries(main PRIVATE LL nfa)
+target_link_libraries(main PRIVATE LL1 nfa)

LL1/CMakeLists.txt

@@ -1,14 +1,15 @@
 cmake_minimum_required(VERSION 3.10)
-project(LL)
+project(LL1)
 
 # 收集所有的cpp源文件
-file(GLOB SOURCES "*.cpp")
-
-# 设置输出目录为 bin
-set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR})
+file(GLOB SOURCES "src/*.cpp")
 
 # 创建静态链接库
-add_library(LL STATIC ${SOURCES})
+add_library(LL1 STATIC ${SOURCES})
+
+# Test:添加可执行文件并链接目标库
+add_executable(test_LL1 test/test_main.cpp)
+target_link_libraries(test_LL1 LL1)
 
 # 添加头文件目录
-target_include_directories(LL PUBLIC ${CMAKE_CURRENT_SOURCE_DIR})
+target_include_directories(LL PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/include)

LL1/archive/CMakeLists.txt

@@ -0,0 +1,14 @@
+cmake_minimum_required(VERSION 3.10)
+project(LL)
+
+# 收集所有的cpp源文件
+file(GLOB SOURCES "*.cpp")
+
+# 设置输出目录为 bin
+set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR})
+
+# 创建静态链接库
+add_library(LL STATIC ${SOURCES})
+
+# 添加头文件目录
+target_include_directories(LL PUBLIC ${CMAKE_CURRENT_SOURCE_DIR})

LL1/include/LL1.h

@@ -0,0 +1,32 @@
+// LL1 语法分析器
+#ifndef LL1_H
+#define LL1_H
+
+#include "grammar.h"
+
+using namespace std;
+
+class LL1:public Grammar{
+public:
+	LL1();
+	~LL1();
+
+	bool IsLL1();														// 判断该文法是否为 LL1 文法
+	void build_LL1_predict();											// 构建 LL1 的预测分析表
+	void print_LL1_predict();											// 打印 LL1 的预测分析表
+	void build_LL1_grammar();											// 构建规约序列
+	void print_LL1_grammar_log();
+	void fileout_LL1_grammar_log(string file_name);
+
+
+private:
+	unordered_map<string, vector<string>> select;						// 计算符号的 SELECT 集合
+	unordered_map<string, unordered_map<string, int>> LL1_predict;		// LL1 的预测分析表
+	vector<string> LL1_grammar_log;									    // 规约序列
+
+	int insert_rule(pair<string, vector<string>>& new_rule);			// 增加新的规则
+};
+
+
+
+#endif // !LL1_H

LL1/include/grammar.h

@@ -0,0 +1,55 @@
+// 语法生成器
+#ifndef GRAMMAR_H
+#define GRAMMAR_H
+
+
+#include <string>
+#include <vector>
+#include <map>
+#include <unordered_set>
+#include <unordered_map>
+
+using namespace std;
+
+class Grammar
+{
+public:
+	const string grammar_file = "./tests/grammar.txt";
+	
+	Grammar();
+	~Grammar();
+	void read_grammar();									// 读取语法规则
+	void print_grammar();									// 打印语法规则
+	void expand_grammar();									// 拓展语法规则
+	void init_grammar_set();								// 初始化语法相关集合
+	void print_grammar_set();								// 打印语法相关集合
+	void get_token_strings(vector<string> &);				// 获取 token_stirngs
+	void print_token_strings();
+
+protected:
+	vector<pair<string, vector<string>>> grammar_rules;		// 产生式规则
+	string start;											// 起始字符
+	vector<string> symbols;									// 符号
+	vector<string> VTs;										// 终结符								
+	vector<string> VNs;										// 非终结符
+	unordered_map<string, vector<string>> first;			// FIRST 集
+	unordered_map<string, vector<string>> follow;			// FOLLOW 集
+	unordered_map<string, bool> infer_empty;				// 是否可以推导出 $ 空字符
+	vector<string> token_strings;
+
+
+private:
+	unordered_map<string, vector<int>> left_appears;		// 该符号出现在哪些产生式左侧
+	unordered_map<string, vector<int>> right_appears;		// 该符号出现在哪些产生式右侧
+	unordered_map<string, vector<string>> depend;			// FOLLOW 集的依赖关系
+	
+
+	void init_appears_depend();								// 获取 appear depend 集合
+	bool symbol_infer_empty(const string& symbol);			// 判断符号是否可以推导出 $ 空字符
+	vector<string> symbol_infer_first(const string& symbol);// 推导符号的 FIRST 集
+	vector<string> symbol_infer_follow(const string& symbol);// 推导符号的 FOLLOW 集
+
+};
+
+
+#endif // !GRAMMAR_H

LL1/src/LL1.cpp

@@ -0,0 +1,351 @@
+#include <set>
+#include <algorithm>
+#include <stack>
+#include <cstring>
+#include <fstream>
+#include <iostream>
+#include <sstream>
+
+#include "LL1.h"
+
+
+LL1::LL1()
+{
+	read_grammar();
+	init_grammar_set();
+}
+
+LL1::~LL1()
+{
+}
+
+bool LL1::IsLL1()
+{
+	string symbol;
+	vector<string> right_first = vector<string>();
+	vector<string> left_follow;
+	for (int i = 0; i < grammar_rules.size(); i++) {
+		symbol.clear();
+		right_first.clear();
+		left_follow.clear();
+
+		symbol = grammar_rules[i].first;
+		
+		
+		
+		// 计算 产生式左侧 FOLLOW 集
+		left_follow = follow[symbol];
+
+		// 计算 产生式右侧 FIRST 集
+
+		// 对 X1 的 非 $ 符号 加入
+		for (int j = 0; j < first[grammar_rules[i].second[0]].size(); j++) {
+			if (first[grammar_rules[i].second[0]][j] == "$") {
+				continue;
+			}
+			right_first.push_back(first[grammar_rules[i].second[0]][j]);
+		}
+
+		int cnt;
+		for (cnt = 1; cnt < grammar_rules[i].second.size(); cnt++) {
+			
+			// 当且仅当 有 $ 符号时 继续加入
+			if (!infer_empty[grammar_rules[i].second[cnt - 1]]) {
+				break;
+			}
+			for (int j = 0; j < first[grammar_rules[i].second[cnt]].size(); j++) {
+				if (first[grammar_rules[i].second[cnt]][j] == "$") {
+					continue;
+				}
+				right_first.push_back(first[grammar_rules[i].second[cnt]][j]);
+			}
+		}
+
+		// 若都能推导至 $ 符号时 加入
+		if (cnt == grammar_rules[i].second.size() && infer_empty[grammar_rules[i].second[0]]) {
+			right_first.push_back("$");
+		}
+		
+		// 对产生式右侧 FIRST 集 进行 去重
+		set<string> sright_first(right_first.begin(), right_first.end());
+		right_first.clear();
+		right_first.resize(sright_first.size());
+		right_first.assign(sright_first.begin(), sright_first.end());
+
+		
+
+		vector<string> symbol_select;
+
+		// 若产生式右侧 FIRST 集为 {$} 时
+		if (right_first.size() == 1 && right_first[0] == "$") {
+			
+			// SELECT 集为 产生式右侧 FOLLOW 集 与 {$} 的交集
+			symbol_select = left_follow;
+			if (find(left_follow.begin(), left_follow.end(), "$") == left_follow.end()) {
+				symbol_select.push_back("$");
+			}
+		}
+		else
+		{
+			// SELECT 集为 产生式左侧 FIRST 集
+			symbol_select = right_first;
+		}
+
+		// 对 SELECT 集 进行排序 方便接下来进行集合运算
+		sort(symbol_select.begin(), symbol_select.end());
+
+		vector<string> new_select = vector<string>();
+		
+		// 判断 SELECT 表中有无现有数据
+		if (select.find(symbol) == select.end()) {
+			
+			select[symbol] = symbol_select;
+		}
+		else {
+
+			// 判断两个相同产生式左侧 SELECT 集 是否相交
+			set_intersection(symbol_select.begin(), symbol_select.end(), select[symbol].begin(), select[symbol].end(), back_inserter(new_select));
+
+			if (new_select.size() == 0) {
+				// 不相交，继续运算，存入两者并集
+				set_union(symbol_select.begin(), symbol_select.end(), select[symbol].begin(), select[symbol].end(), back_inserter(new_select));
+			}
+			else
+			{
+				// 非 LL(1) 文法，退出
+				cout << "This grammar is not LL (1) grammar" << endl;
+				return false;
+			}
+
+		}
+
+	}
+
+	// cout << "该文法为 LL(1) 文法！" << endl;
+	return true;
+}
+
+void LL1::build_LL1_predict()
+{
+	// 对每一个 非终结符 进行初始化行
+	for (int i = 0; i < VNs.size(); i++) {
+		if (LL1_predict.find(VNs[i]) == LL1_predict.end()) {
+			LL1_predict[VNs[i]] = unordered_map<string, int>();
+		}
+	}
+
+	string symbol;
+	vector<string> right_first = vector<string>();
+	vector<string> left_follow;
+
+	// 遍历 产生式 构建 预测分析表
+	for (int i = 0; i < grammar_rules.size(); i++) {
+		symbol.clear();
+		right_first.clear();
+		left_follow.clear();
+
+		symbol = grammar_rules[i].first;
+
+
+		// 计算 产生式左侧 FOLLOW 集
+		left_follow = follow[symbol];
+
+		unordered_map<string, int> &symbol_predict = LL1_predict[symbol];
+		
+		
+		// 计算 产生式右侧 FIRST 集
+
+		// 对 X1 的 非 $ 符号 加入
+		for (int j = 0; j < first[grammar_rules[i].second[0]].size(); j++) {
+			if (first[grammar_rules[i].second[0]][j] == "$") {
+				continue;
+			}
+			right_first.push_back(first[grammar_rules[i].second[0]][j]);
+		}
+
+		int cnt;
+		for (cnt = 1; cnt < grammar_rules[i].second.size(); cnt++) {
+
+			// 当且仅当 有 $ 符号时 继续加入
+			if (!infer_empty[grammar_rules[i].second[cnt - 1]]) {
+				break;
+			}
+			for (int j = 0; j < first[grammar_rules[i].second[cnt]].size(); j++) {
+				if (first[grammar_rules[i].second[cnt]][j] == "$") {
+					continue;
+				}
+				right_first.push_back(first[grammar_rules[i].second[cnt]][j]);
+			}
+		}
+
+		// 若都能推导至 $ 符号时 加入
+		if (cnt == grammar_rules[i].second.size() && infer_empty[grammar_rules[i].second[0]]) {
+			right_first.push_back("$");
+		}
+
+		// 对产生式右侧 FIRST 集 进行 去重
+		set<string> sright_first(right_first.begin(), right_first.end());
+		right_first.clear();
+		right_first.resize(sright_first.size());
+		right_first.assign(sright_first.begin(), sright_first.end());
+
+		// 循环遍历 FIRST 集进行初始化
+		for (int j = 0; j < right_first.size(); j++) {
+			if (right_first[j] == "$") {
+				pair<string, vector<string>> new_rule (grammar_rules[i].first, vector<string>());
+				new_rule.second.push_back("$");
+				int rule_id = insert_rule(new_rule);
+
+				for (int k = 0; k < left_follow.size(); k++) {
+					symbol_predict[left_follow[k]] = rule_id;
+				}
+			}
+			symbol_predict[right_first[j]] = i;
+
+		}
+
+	}
+
+
+}
+
+void LL1::print_LL1_predict()
+{
+	cout << "[LL1_predict]:" << endl;
+	for (auto iter = LL1_predict.begin(); iter != LL1_predict.end(); ++iter) {
+		cout << (*iter).first << " ";
+		for (auto j = (*iter).second.begin(); j != (*iter).second.end(); ++j) {
+			cout << (*j).first << "," << (*j).second << " ";
+		}
+		cout << endl;
+
+	}
+	cout << endl << endl;
+
+}
+
+void LL1::build_LL1_grammar()
+{
+	// 符号栈
+	stack<string> stack;
+	int token_cnt = 0;
+	
+	// 起始符 入栈
+	stack.push(start);
+
+	while (!stack.empty())
+	{
+		LL1_grammar_log.push_back(string());
+		
+		// 栈顶符号
+		// 判断栈顶是否为 空符号 
+		if (stack.top() == "$") {
+			// 栈空 以 EOF 表示
+			LL1_grammar_log.back() += "EOF";
+		}
+		else
+		{
+			LL1_grammar_log.back() += stack.top();
+		}
+		
+		// 添加 # 分割
+		LL1_grammar_log.back() += "#";
+
+		// 面临输入的符号
+		string this_token;
+		if (token_cnt == token_strings.size()) {
+			// 栈空 以 EOF 表示
+			this_token = "$";
+			LL1_grammar_log.back() += "EOF";
+		}
+		else
+		{
+			this_token = token_strings[token_cnt];
+			LL1_grammar_log.back() += token_strings[token_cnt];
+		}
+		
+		// 对栈顶元素与即将输入的符号进行比较
+		if (stack.top() == this_token) {
+			// 栈顶出栈 token 指向下一位
+			token_cnt++;
+			stack.pop();
+
+			if (this_token == "$") {
+				// 分析成功 结束分析
+				LL1_grammar_log.back() += "\taccept";
+			}
+			else
+			{
+				// 跳过
+				LL1_grammar_log.back() += "\tmove";
+			}
+		}
+		// 若为终结符
+		else if (find(VTs.begin(), VTs.end(), stack.top()) != VTs.end()) {
+			if (stack.top() == "$") {
+				stack.pop();
+				LL1_grammar_log.pop_back();
+			}
+			else {
+				LL1_grammar_log.back() += "\terror";
+				return;
+			}
+		}
+		else
+		{
+			auto tab = LL1_predict[stack.top()];
+			
+			if (tab.find(this_token) == tab.end()) {
+				LL1_grammar_log.back() += "\terror";
+				return;
+			}
+			else
+			{
+				auto this_rule = grammar_rules[tab[this_token]];
+				stack.pop();
+				for (int i = this_rule.second.size() - 1; i >= 0; i--) {
+					stack.push(this_rule.second[i]);
+				}
+				LL1_grammar_log.back() += "\treduction";
+			}
+		}
+	}
+
+}
+
+void LL1::print_LL1_grammar_log()
+{
+	for (int i = 0; i < LL1_grammar_log.size(); ++i) {
+		cout << LL1_grammar_log[i] << endl;
+	}
+}
+
+void LL1::fileout_LL1_grammar_log(string file_name)
+{
+	//打开结果输出文件
+	fstream outfile(file_name);
+
+	if (!outfile.is_open()) {
+		cout << "[FILEOUT] fail to open file" << endl;
+	}
+
+	for (int i = 0; i < LL1_grammar_log.size(); ++i) {
+		outfile << LL1_grammar_log[i] << endl;
+	}
+	outfile.close();
+}
+
+int LL1::insert_rule(pair<string, vector<string>>& new_rule)
+{
+	int cnt;
+	for (cnt = 0; cnt < grammar_rules.size(); cnt++) {
+		// 当 产生式规则 中存在这条产生式时 返回序号
+		if (grammar_rules[cnt].first == new_rule.first && grammar_rules[cnt].second == new_rule.second) {
+			return cnt;
+		}
+	}
+	// 若不存在 返回序号的同时加入
+	grammar_rules.push_back(new_rule);
+	return cnt;
+}
+

LL1/src/grammar.cpp

@@ -0,0 +1,520 @@
+#include <deque>
+#include <cstring>
+#include <fstream>
+#include <iostream>
+#include <sstream>
+#include <algorithm>
+#include <set>
+
+#include "grammar.h"
+
+
+Grammar::Grammar()
+{
+
+}
+
+Grammar::~Grammar()
+{
+
+}
+
+
+
+void Grammar::read_grammar() {
+	fstream infile;
+	infile.open(grammar_file,ios::in);
+	if (!infile.is_open())
+	{
+		cout << "[READ_GRAMMAR] fail to open file: "<< grammar_file << endl;
+		return;
+	}
+
+	string  buf;
+	string arrow = "->";
+	string farrow;
+	bool start_flag = true;
+	string left;
+	string forms;
+
+	while (!infile.eof()) {	
+		// 清理 string
+		buf.clear();
+		left.clear();
+		forms.clear();
+		farrow.clear();
+
+
+		grammar_rules.push_back(pair<string, vector<string>>());
+		getline(infile, buf);
+
+		stringstream ss(buf);
+
+		// 读取产生式左侧
+		ss >> left;
+		grammar_rules.back().first = left;
+		symbols.push_back(left);
+		VNs.push_back(left);
+		
+		// 存储 start
+		if (start_flag) {
+			start = left;
+			start_flag = false;
+		}
+
+		// 读取 -> 符号 并保证合法
+		ss >> farrow;
+		if (farrow != arrow) {
+			cout << "Grammar reading error" << endl;
+		}
+
+		// 读取产生式右侧
+		while (ss >> forms)
+		{
+			grammar_rules.back().second.push_back(forms);
+			symbols.push_back(forms);
+			forms.clear();
+		}
+	}
+
+	// 符号集 和 非终结符 去重
+	set<string> ssymbols(symbols.begin(), symbols.end());
+	symbols.clear();
+	symbols.resize(ssymbols.size());
+	symbols.assign(ssymbols.begin(), ssymbols.end());
+
+	set<string> sVNs(VNs.begin(), VNs.end());
+	VNs.clear();
+	VNs.resize(sVNs.size());
+	VNs.assign(sVNs.begin(), sVNs.end());
+
+	// 符号集 和 非终结符 排序 以保证差集的成功
+	sort(symbols.begin(), symbols.end());
+	sort(VNs.begin(), VNs.end());
+
+	// 取差集 得到终极符
+	set_difference(symbols.begin(), symbols.end(), VNs.begin(), VNs.end(), back_inserter(VTs));
+
+	infile.close();
+
+}
+
+void Grammar::print_grammar()
+{
+	cout << "[start]: " << endl << start << endl << endl;
+
+	cout << "[VTs]:" << endl;
+	for (int i = 0; i < VTs.size(); i++) {
+		cout << VTs[i] << " ";
+		if (((i + 1) % 5) == 0)
+			cout << endl;
+	}
+	cout << endl << endl;
+
+	cout << "[VNs]:" << endl;
+	for (int i = 0; i < VNs.size(); i++) {
+		cout << VNs[i] << " ";
+		if (((i + 1) % 5) == 0)
+			cout << endl;
+	}
+	cout << endl << endl;
+
+	cout << "[symbols]:" << endl;
+	for (int i = 0; i < symbols.size(); i++) {
+		cout << symbols[i] << " ";
+		if (((i + 1) % 5) == 0)
+			cout << endl;
+	}
+	cout << endl << endl;
+
+	cout << "[grammar_rules]: " << grammar_rules.size() << endl;
+	for (int i = 0; i < grammar_rules.size(); ++i) {
+		cout << grammar_rules[i].first << " -> ";
+		for (int j = 0; j < grammar_rules[i].second.size(); ++j) {
+			cout << "\"" << grammar_rules[i].second[j] << "\" ";
+		}
+		cout << endl;
+	}
+	cout << endl << endl;
+}
+
+void Grammar::expand_grammar()
+{
+	string new_start = start + "\'";
+	pair<string, vector<string>> new_rule = pair<string, vector<string>>(new_start, vector<string>());
+	new_rule.second.push_back(start);
+
+	VNs.push_back(new_start);
+	symbols.push_back(new_start);
+	grammar_rules.insert(grammar_rules.begin(), new_rule);
+	start = new_start;
+
+	// 符号集排序
+	sort(symbols.begin(), symbols.end());
+
+}
+
+void Grammar::init_grammar_set()
+{
+	string symbol;
+
+	
+
+	// 对符号集中各符号进行推导 是否可以到达 $ 空符号
+	for (int i = 0; i < symbols.size(); i++) {
+		symbol = symbols[i];
+		this->symbol_infer_empty(symbol);
+		symbol.clear();
+	}
+
+	// 初始化符号在产生式的 出现 依赖 情况
+	init_appears_depend();
+	
+	// 对符号集中各符号进行推导 FIRST 集
+	for (int i = 0; i < symbols.size(); i++) {
+		symbol = symbols[i];
+		this->symbol_infer_first(symbol);
+		symbol.clear();
+	}
+
+	// 对符号集中各符号进行推导 FOLLOW 集
+	
+	// 符号队列
+	deque<string> queue;
+
+	// 初次遍历所有符号 生成初始的 FOLLOW 集 
+
+	// 构建 start 的 FOLLOW 集
+	follow[start] = this->symbol_infer_follow(start);
+	follow[start].push_back("$");
+	queue.push_back(start);
+
+	// 构建除 start 的 FOLLOW 集
+	for (int i = 0; i < symbols.size(); i++) {
+		symbol = symbols[i];
+		if (symbol == start) {
+			symbol.clear();
+			continue;
+		}
+		follow[symbol] = this->symbol_infer_follow(symbol);
+		queue.push_back(symbol);
+		symbol.clear();
+	}
+
+	// 对 符号队列 进行进一步生成
+	while (!queue.empty()) {
+		// 读取 符号队列 开头
+		symbol = queue.front();
+		queue.pop_front();
+
+		// 若 FOLLOW 集发生改变
+		vector<string> new_symbol_follow = this->symbol_infer_follow(symbol);
+		if (follow[symbol].size() < new_symbol_follow.size()) {
+			// 对依赖 该符号 的所有符号添加至 符号队列
+			vector<string> dep = depend[symbol];
+			for (int i = 0; i < dep.size(); i++) {
+				queue.push_back(dep[i]);
+			}
+			follow[symbol] = new_symbol_follow;
+		}
+		symbol.clear();
+	}
+
+
+}
+
+void Grammar::print_grammar_set()
+{
+	// 打印符号在产生式的出现情况
+	cout << "[left_appears]:" << endl;
+	for (int i = 0; i < symbols.size(); i++) {
+		cout << "LEFT( " << symbols[i] << " ) = {";
+		for (int j = 0; j < left_appears[symbols[i]].size(); j++) {
+			cout << " " << left_appears[symbols[i]][j] << " ";
+		}
+		cout << "}" << endl;
+	}
+	cout << endl << endl;
+
+	cout << "[right_appears]:" << endl;
+	for (int i = 0; i < symbols.size(); i++) {
+		cout << "RIGHT( " << symbols[i] << " ) = {";
+		for (int j = 0; j < right_appears[symbols[i]].size(); j++) {
+			cout << " " << right_appears[symbols[i]][j] << " ";
+		}
+		cout << "}" << endl;
+	}
+	cout << endl << endl;
+
+	// 打印 FOLLOW 集的依赖关系
+	cout << "[depend]:" << endl;
+	for (int i = 0; i < symbols.size(); i++) {
+		cout << "DEPEND( " << symbols[i] << " ) = {";
+		for (int j = 0; j < depend[symbols[i]].size(); j++) {
+			cout << " " << depend[symbols[i]][j] << " ";
+		}
+		cout << "}" << endl;
+	}
+	cout << endl << endl;
+
+
+	// 打印是否可以推导出 $ 空符号
+	cout << "[infer_empty]:" << endl;
+	for (int i = 0; i < symbols.size(); i++) {
+		cout << symbols[i]<<" -> " << infer_empty[symbols[i]] << endl;
+	}
+	cout << endl << endl;
+
+	// 打印 FIRST 集
+	cout << "[FIRST]:" << endl;
+	for (int i = 0; i < symbols.size(); i++) {
+		cout << "FIRST( " << symbols[i] << " ) = {";
+		for (int j = 0; j < first[symbols[i]].size(); j++) {
+			cout << " " << first[symbols[i]][j] << " ";
+		}
+		cout << "}" << endl;
+	}
+	cout << endl << endl;
+
+	// 打印 FOLLOW 集
+	cout << "[FOLLOW]:" << endl;
+	for (int i = 0; i < symbols.size(); i++) {
+		cout << "FOLLOW( " << symbols[i] << " ) = {";
+		for (int j = 0; j < follow[symbols[i]].size(); j++) {
+			cout << " " << follow[symbols[i]][j] << " ";
+		}
+		cout << "}" << endl;
+	}
+	cout << endl << endl;
+
+}
+
+void Grammar::get_token_strings(vector<string>& my_token_strings)
+{
+	token_strings.resize(my_token_strings.size());
+	token_strings.assign(my_token_strings.begin(), my_token_strings.end());
+
+}
+
+void Grammar::print_token_strings()
+{
+	for (int i = 0; i < token_strings.size(); i++) {
+		cout << token_strings[i] << endl;
+	}
+}
+
+void Grammar::init_appears_depend()
+{
+	for (int k = 0; k < symbols.size(); k++) {
+		left_appears[symbols[k]] = vector<int>();
+		right_appears[symbols[k]] = vector<int>();
+		depend[symbols[k]] = vector<string>();
+		for (int i = 0; i < grammar_rules.size(); i++) {
+			if (grammar_rules[i].first == symbols[k]) {
+				// 产生式左侧相等 存入 left
+				left_appears[symbols[k]].push_back(i);
+
+				// 对该产生式构建依赖关系
+				for (int m = 0; m < grammar_rules[i].second.size(); m++) {
+					int n;
+					
+					// 判断该产生式右侧符号是否可以推导至 $ 空符号
+					for (n = m + 1; n < grammar_rules[i].second.size(); n++) {
+						if (!infer_empty[grammar_rules[i].second[n]]) {
+							break;
+						}
+					}
+					// 若可以推导 按照入栈的方式依次加入
+					if (n == grammar_rules[i].second.size()) {
+						if (symbols[k] != grammar_rules[i].second[m]) {
+							depend[symbols[k]].push_back(grammar_rules[i].second[m]);
+						}
+					}
+
+				}
+			}
+			for (int j = 0; j < grammar_rules[i].second.size(); j++) {
+				// 产生式右侧相等 存入 left
+				if (grammar_rules[i].second[j] == symbols[k]) {
+					right_appears[symbols[k]].push_back(i);
+					break;
+				}
+			}
+		}
+	}
+
+}
+
+bool Grammar::symbol_infer_empty(const string& symbol) {
+	
+	// 已经进行推导过
+	if (infer_empty.find(symbol) != infer_empty.end()) {
+		return infer_empty[symbol];
+	}
+	
+	// 当符号为终结符时，当且仅当为 $ 可以推导出 $
+	if (find(VTs.begin(), VTs.end(), symbol) != VTs.end()) {
+		infer_empty[symbol] = (symbol == "$") ;
+		return infer_empty[symbol];
+	}
+
+	// 当符号为非终结符时，通过产生式进行推导
+	for (int i = 0; i < grammar_rules.size(); i++) {
+		// 当该符号为产生式左侧时
+		if (grammar_rules[i].first == symbol) {
+			int j;
+			vector<string> rule_right = grammar_rules[i].second;
+			for (j = 0; j < rule_right.size(); j++) {
+				// 递归推导 产生式右侧无法推导至 $ 时
+				if (!(this->symbol_infer_empty(rule_right[j]))) {
+					break;
+				}
+			}
+
+			// 当且仅当产生式右侧可以推导至 $ 时
+			if (j == rule_right.size()) {
+				infer_empty[symbol] = true;
+				return infer_empty[symbol];
+			}
+		}
+	}
+
+	// 当各产生式都无法推导至 $ 时，则无法推导
+	infer_empty[symbol] = false;
+	return infer_empty[symbol];
+
+}
+
+vector<string> Grammar::symbol_infer_first(const string& symbol)
+{
+	// 已经推导过 FIRST 集
+	if (first.find(symbol) != first.end()) {
+		return first[symbol];
+	}
+
+	vector<string> symbol_first;
+	
+	// 当符号为终结符时 FIRST 集为它本身
+	if (find(VTs.begin(), VTs.end(), symbol) != VTs.end()) {	
+		symbol_first.push_back(symbol);
+		first[symbol] = symbol_first;
+		return first[symbol];
+	}
+
+	// 当符号为非终结符时，通过产生式进行推导
+	for (int i = 0; i < grammar_rules.size(); i++) {
+		// 当该符号为产生式左侧时
+		if (grammar_rules[i].first == symbol) {
+			int j;
+			for (j = 0; j < grammar_rules[i].second.size(); j++) {
+
+				// 依次添加所有产生式右侧的
+				vector<string> firsts = symbol_infer_first(grammar_rules[i].second[j]);
+				for (int k = 0; k < firsts.size(); k++) {
+					symbol_first.push_back(firsts[k]);
+				}
+				
+				// 若产生式右侧无法推导至 $ 空字符时 中断
+				if (!infer_empty[grammar_rules[i].second[j]]) {
+					break;
+				}
+
+			}
+
+			// 当且仅当产生式右侧可以推导至 $ 时 将 $ 加入到 FIRST 集中
+			if (j == grammar_rules[i].second.size()) {
+				symbol_first.push_back("$");
+			}
+		}
+	}
+
+	// 对当前 FIRST 集进行 去重 与 排序
+	set<string> ssymbol_first(symbol_first.begin(), symbol_first.end());
+	symbol_first.clear();
+	symbol_first.resize(ssymbol_first.size());
+	symbol_first.assign(ssymbol_first.begin(), ssymbol_first.end());
+
+	sort(symbol_first.begin(), symbol_first.end());
+
+	// 返回非终结符的 FIRST 集
+	first[symbol] = symbol_first;
+	return first[symbol];
+}
+
+vector<string> Grammar::symbol_infer_follow(const string& symbol)
+{
+	vector<string> symbol_follow;
+	
+	// 获取该符号出现在哪些产生式右侧
+	vector<int> right_appear = right_appears[symbol];
+	for (int i = 0; i < right_appear.size(); i++) {
+		int cnt;
+		
+		// 获取该产生式右侧的符号
+		vector<string> rule_right = grammar_rules[right_appear[i]].second;
+		
+		// 依次遍历 该产生式右侧 至 该符号 后一位
+		for (cnt = 0; cnt < rule_right.size(); cnt++) {
+			if (rule_right[cnt] == symbol) {
+				break;
+			}
+		}
+		cnt++;
+
+		// 遍历 剩余产生式右侧
+		for (; cnt < rule_right.size(); cnt++) {
+			
+			// 依次获取 后置元素 的 FIRST 集
+			vector<string> symbol_first = first[rule_right[cnt]];
+
+			// 将 该 FIRST 集 循环添加至 symbol_follow 中
+			for (int j = 0; j < symbol_first.size(); j++) {
+				symbol_follow.push_back(symbol_first[j]);
+			}
+
+			// 若不可达 $ 中断遍历
+			if (!infer_empty[rule_right[cnt]]) {
+					break;
+			}
+		}
+
+		// 当剩余产生式右侧均可到达 $ 时 
+		if (cnt == rule_right.size()) {
+			if (follow.find(grammar_rules[right_appear[i]].first) != follow.end()) {
+				
+				// 将产生式左侧的 FOLLOW 集 加入到 当前符号的 FOLLOW 集中
+				vector<string> first_follow = follow[grammar_rules[right_appear[i]].first];
+				for (int j = 0; j < first_follow.size(); j++) {
+					symbol_follow.push_back(first_follow[j]);
+				}
+
+			}
+
+		}
+		
+	}
+
+	// 删除不需要的 $ 空字符
+	auto it = remove(symbol_follow.begin(), symbol_follow.end(), "$");
+	auto it1 = symbol_follow.erase(it, symbol_follow.end());
+
+
+	// 对当前 FOLLOW 集 进行去重排序
+	set<string> ssymbol_follow(symbol_follow.begin(), symbol_follow.end());
+	symbol_follow.clear();
+	symbol_follow.resize(ssymbol_follow.size());
+	symbol_follow.assign(ssymbol_follow.begin(), ssymbol_follow.end());
+
+	sort(symbol_follow.begin(), symbol_follow.end());
+
+
+
+	return symbol_follow;
+}
+
+
+
+
+
+
+

LL1/test/test_main.cpp

@@ -0,0 +1,73 @@
+#include <iostream>
+#include <fstream>
+#include <cassert>
+
+#include "grammar.h"
+#include "LL1.h"
+using namespace std;
+
+string inputs[] = {
+    "tests/00.txt",
+    "tests/20.txt"
+};
+
+string outputs_lexical[] = {
+"tests/00_my_lexical.txt",
+"tests/20_my_lexical.txt"
+};
+
+string outputs_grammar[] = {
+"tests/00_my_grammar.txt",
+"tests/20_my_grammar.txt",
+};
+
+// 读取文件
+string readfile(const string& filename)
+{
+    ifstream file(filename);
+    string content((istreambuf_iterator<char>(file)),istreambuf_iterator<char>());
+    return content;
+}
+
+int main(int argc, char** argv) {
+
+    try{
+        int i=0;
+        for (auto input : inputs) {
+                LL1 ll;
+                //ll.print_grammar_set();
+
+                string content = readfile(input);
+                vector<string> token_strings = recognize(minimizedDFA, content,outputs_lexical[i]);
+
+                bool flag = ll.IsLL1();
+                ll.build_LL1_predict();
+
+
+                //ll.print_LL1_predict();
+                ll.get_token_strings(token_strings);
+
+
+                ll.print_token_strings();
+                ll.build_LL1_grammar();
+
+
+                ll.fileout_LL1_grammar_log(outputs_grammar[i]);
+
+
+                //ll.print_LL1_grammar_log();
+                cout << endl;
+                cout<<outputs_grammar[i]<<endl;
+                cout<<"end"<<endl;
+                i++;
+            }
+            cout<<"LL1 FINISH"<<endl;
+    }
+    catch(...){
+        cout<<"ERROR"<<endl;
+    }
+
+    cout<<"SYSTEM END"<<endl;
+
+	return 0;
+}

main/main.cpp

@@ -7,73 +7,91 @@
 #include "LL1.h"
 using namespace std;
 
+
+string inputs[] = {
+    "tests/00/00.txt",
+    "tests/01/01.txt",
+    "tests/02/02.txt",
+    "tests/07/07.txt",
+    "tests/08_err/08.txt",
+    "tests/10_err/10.txt",
+    "tests/20/20.txt"
+};
+
+string outputs_lexical[] = {
+"tests/00/00_my_lexical.txt",
+"tests/01/01_my_lexical.txt",
+"tests/02/02_my_lexical.txt",
+"tests/07/07_my_lexical.txt",
+"tests/08_err/08_my_lexical.txt",
+"tests/10_err/10_my_lexical.txt",
+"tests/20/20_my_lexical.txt"
+};
+
+string outputs_grammar[] = {
+"tests/00/00_my_grammar.txt",
+"tests/01/01_my_grammar.txt",
+"tests/02/02_my_grammar.txt",
+"tests/07/07_my_grammar.txt",
+"tests/08_err/08_my_grammar.txt",
+"tests/10_err/10_my_grammar.txt",
+"tests/20/20_my_grammar.txt",
+};
+
 int main(int argc, char** argv) {
     
-    NFA nfa = RexToNFA();
-    //printNFA(nfa);
 
-    DFA dfa = nfaToDFA(nfa);
-    //printDFA(dfa);
-    DFA minimizedDFA = minimizeDFA(minimizeDFA(dfa));
-    removeUnreachableStates(minimizedDFA);
-    //printDFA(minimizedDFA);
+        NFA nfa = RexToNFA();
+        //printNFA(nfa);
 
-    string inputs[6] = {
-        "tests/00/00.txt",
-        "tests/01/01.txt",
-        "tests/02/02.txt",
-        "tests/07/07.txt",
-        "tests/08_err/08.txt",
-        "tests/10_err/10.txt"
-    };
+        DFA dfa = nfaToDFA(nfa);
+        //printDFA(dfa);
+        DFA minimizedDFA = minimizeDFA(minimizeDFA(dfa));
+        removeUnreachableStates(minimizedDFA);
+        //printDFA(minimizedDFA);
 
-    string outputs_lexical[6] = {
-    "tests/00/00_my_lexical.txt",
-    "tests/01/01_my_lexical.txt",
-    "tests/02/02_my_lexical.txt",
-    "tests/07/07_my_lexical.txt",
-    "tests/08_err/08_my_lexical.txt",
-    "tests/10_err/10_my_lexical.txt"
-    };
+        cout<<"DFA FINISH"<<endl;
+        int i = 0;
+    try{
+        for (auto input : inputs) {
+            LL1 ll;
+            //ll.print_grammar_set();
 
-    string outputs_grammar[6] = {
-    "tests/00/00_my_grammar.txt",
-    "tests/01/01_my_grammar.txt",
-    "tests/02/02_my_grammar.txt",
-    "tests/07/07_my_grammar.txt",
-    "tests/08_err/08_my_grammar.txt",
-    "tests/10_err/10_my_grammar.txt"
-    };
+            string content = readfile(input);
+            vector<string> token_strings = recognize(minimizedDFA, content,outputs_lexical[i]);
+
+            bool flag = ll.IsLL1();
+            ll.build_LL1_predict();
 
 
-    int i = 0;
-    for (auto input : inputs) {
-        LL1 ll;
-        //ll.print_grammar_set();
-
-        string content = readfile(input);
-        vector<string> token_strings = recognize(minimizedDFA, content,outputs_lexical[i]);
-
-	    bool flag = ll.IsLL1();
-	    ll.build_LL1_predict();
+            //ll.print_LL1_predict();
+            ll.get_token_strings(token_strings);
 
 
-	    //ll.print_LL1_predict();
-        ll.get_token_strings(token_strings);
+            ll.print_token_strings();
+            ll.build_LL1_grammar();
 
 
-        ll.print_token_strings();
-        ll.build_LL1_grammar();
+            ll.fileout_LL1_grammar_log(outputs_grammar[i]);
 
 
-        ll.fileout_LL1_grammar_log(outputs_grammar[i]);
-
-
-        //ll.print_LL1_grammar_log();
-        cout << endl;
-        cout<<"end"<<endl;
-        i++;
+            //ll.print_LL1_grammar_log();
+            cout << endl;
+            cout<<outputs_grammar[i]<<endl;
+            cout<<"end"<<endl;
+            i++;
+        }
+        cout<<"LL1 FINISH"<<endl;
     }
+    catch(...){
+        cout<<"FINAL ERROR";
+    }
+
+
+
+
+
+