正则表达式LL1分析算符优先LR分析.docx

1、正则表达式LL1分析算符优先LR分析目 录1 正则表达式 11.1 正则表达式 11.2 确定化(化简)后的状态转换图 11.3 分析程序代码 11.4 程序运行截图 11.5 小结 12 LL(1)分析 22.1 LL(1)文法 22.2 LL(1)预测分析表 22.3 分析程序代码 22.4 程序运行截图 22.5 小结 23 算符优先分析 33.1 算符优先文法 33.2 算符优先关系表 33.3 分析程序代码 33.4 程序运行截图 33.5 小结 34 LR分析 44.1 LR文法 44.2 LR分析表 44.3 分析程序代码 44.4 程序运行截图 44.5 小结 4参考文献： 4

2、1 正则表达式1.1 正则表达式 (a|b)*(aa|bb)(a|b)* (注：该正规式为示例，可更改）1.2 确定化(化简)后的状态转换图 1.3 分析程序代码#include #include #include using namespace std;/定义一个存储状态转换的类class Statupublic: map MAP;int main() /定义4个状态并初始化各个状态间的关系，以及定义一个存储当前状态的变量status Statu A, B, C, D, *status; A.MAPa = &B; A.MAPb = &C; B.MAPa = &D; B.MAPb = &C;

3、C.MAPa = &B; C.MAPb = &D; D.MAPa = &D; D.MAPb = &D; while(1) string:size_type size=0; string str; bool flag = true; cout请输入符合(a|b)*(aa|bb)(a|b)* 的正则表达式 str; status = &A;/刚开始指向初始状态 const char *ch = str.c_str(); while(chsize) if(chsize != a & chsize != b) flag = false; break; status = status-MAPchsize

4、+;/根据输入字符串调整状态 if(status = &D & flag)/若为终结状态则符合此正则表达式 cout正确endl; else cout错误endl; return 0;1.4 程序运行截图 1.5 小结 通过定义各个状态，并生成各个状态间的关系可以方便地表示状态转换图，使用关联容器则可以方便的管理状态。2 LL(1)分析2.1 LL(1)文法 ETE (注：该文法为示例，可更改） E+TE| TFT T*FT| F(E)|i2.2 LL(1)预测分析表i+*()#EETEETEEE+TEEETTFTTFTTTT*FTTTFFiF(E)2.3 分析程序代码#include #in

5、clude #include #include using namespace std;enum Symbols / 终结符号 Terminal symbols：TS TS_I, / i TS_PLUS, / + TS_MULTIPLY, / * TS_L_PARENS, / ( TS_R_PARENS, / ) TS_EOS, / #, 0 TS_INVALID, / 非法字符 /非终结符号 None terminal symbols：NTS NTS_E, /E NTS_EE, /E NTS_T, /T NTS_TT, /T NTS_F /F;enum Symbols lexer(char

6、c) switch(c) case i: return TS_I; case +: return TS_PLUS; case *: return TS_MULTIPLY; case (: return TS_L_PARENS; case ): return TS_R_PARENS; case #: return TS_EOS; / 栈底：#终结符号 default: return TS_INVALID; int main() while(1) cout 输入一个以#结束的字符串 str; p = str.c_str(); map enum Symbols, map table; stack s

7、s; /设置词法分析表 tableNTS_ETS_I = 1; tableNTS_ETS_L_PARENS = 1; tableNTS_EETS_PLUS = 2; tableNTS_EETS_R_PARENS = 3; tableNTS_EETS_EOS = 3; tableNTS_TTS_I = 4; tableNTS_TTS_L_PARENS = 4; tableNTS_TTTS_PLUS = 6; tableNTS_TTTS_MULTIPLY = 5; tableNTS_TTTS_R_PARENS = 6; tableNTS_TTTS_EOS = 6; tableNTS_FTS_I =

8、 8; tableNTS_FTS_L_PARENS = 7; /初始化符号栈 ss.push(TS_EOS); ss.push(NTS_E); while(ss.size() 0) if(lexer(*p) = ss.top() p+; ss.pop(); else switch(tabless.top()lexer(*p) case 1:/1.ETE ss.pop(); ss.push(NTS_EE); ss.push(NTS_T); break; case 2:/2.E+TE ss.pop(); ss.push(NTS_EE); ss.push(NTS_T); ss.push(TS_PLU

9、S); break; case 3:/2.E ss.pop(); break; case 4:/4.TFT ss.pop(); ss.push(NTS_TT); ss.push(NTS_F); break; case 5:/5.T*FT ss.pop(); ss.push(NTS_TT); ss.push(NTS_F); ss.push(TS_MULTIPLY); break; case 6:/6.T ss.pop(); break; case 7:/7.F(E) ss.pop(); ss.push(TS_R_PARENS); ss.push(NTS_E); ss.push(TS_L_PARE

10、NS); break; case 8:/7.Fi ss.pop(); ss.push(TS_I); break; default: cout 错误的句子 endl; goto END; break; cout 正确的句子 endl; END: ; return 0;2.4 程序运行截图2.5 小结在程序中要跳出多层语句只能用goto语句实现。用枚举类型表示各个终结符与非终结符，再用关联容器生成预测分析表即可方便地实现此算法。3 算符优先分析3.1 算符优先文法 ET | E+T | E-T (注：该文法为示例，可更改） TF | T*F | T/F F(E) | i3.2 算符优先关系表+-*

11、/()i#+-*/()i#3.3 分析程序代码#include #include #include #include using namespace std;enum Symbols / 终结符号 TS_PLUS, / + TS_MINUS, / - TS_MULTIPLY, / * TS_DIVISION, / / TS_L_PARENS, / ( TS_R_PARENS, / ) TS_I, / i TS_EOS, / #, 0 TS_INVALID, / 非法字符;map enum Symbols, map table; enum Symbols lexer(char c) switch

12、(c) case +: return TS_PLUS; case -: return TS_MINUS; case *: return TS_MULTIPLY; case /: return TS_DIVISION; case (: return TS_L_PARENS; case ): return TS_R_PARENS; case i: return TS_I; case #: return TS_EOS; / 栈底：#终结符号 default: return TS_INVALID; /初始化算符优先关系表，0、1、2分别表示=、，-1表示不存在void setup() tableTS_

13、PLUSTS_PLUS = 2; tableTS_PLUSTS_MINUS = 2; tableTS_PLUSTS_MULTIPLY = 1; tableTS_PLUSTS_DIVISION = 1; tableTS_PLUSTS_L_PARENS = 1; tableTS_PLUSTS_R_PARENS = 2; tableTS_PLUSTS_I = 1; tableTS_PLUSTS_EOS = 2; tableTS_MINUSTS_PLUS = 2; tableTS_MINUSTS_MINUS = 2; tableTS_MINUSTS_MULTIPLY = 1; tableTS_MINU

14、STS_DIVISION = 1; tableTS_MINUSTS_L_PARENS = 1; tableTS_MINUSTS_R_PARENS = 2; tableTS_MINUSTS_I = 1; tableTS_MINUSTS_EOS = 2; tableTS_MULTIPLYTS_PLUS = 2; tableTS_MULTIPLYTS_MINUS = 2; tableTS_MULTIPLYTS_MULTIPLY = 2; tableTS_MULTIPLYTS_DIVISION = 2; tableTS_MULTIPLYTS_L_PARENS = 1; tableTS_MULTIPLY

15、TS_R_PARENS = 2; tableTS_MULTIPLYTS_I = 1; tableTS_MULTIPLYTS_EOS = 2; tableTS_DIVISIONTS_PLUS = 2; tableTS_DIVISIONTS_MINUS = 2; tableTS_DIVISIONTS_MULTIPLY = 2; tableTS_DIVISIONTS_DIVISION = 2; tableTS_DIVISIONTS_L_PARENS = 1; tableTS_DIVISIONTS_R_PARENS = 2; tableTS_DIVISIONTS_I = 1; tableTS_DIVI

16、SIONTS_EOS = 2; tableTS_L_PARENSTS_PLUS = 1; tableTS_L_PARENSTS_MINUS = 1; tableTS_L_PARENSTS_MULTIPLY = 1; tableTS_L_PARENSTS_DIVISION = 1; tableTS_L_PARENSTS_L_PARENS = 1; tableTS_L_PARENSTS_R_PARENS = 0; tableTS_L_PARENSTS_I = 1; tableTS_L_PARENSTS_EOS = -1; tableTS_R_PARENSTS_PLUS = 2; tableTS_R

17、_PARENSTS_MINUS = 2; tableTS_R_PARENSTS_MULTIPLY = 2; tableTS_R_PARENSTS_DIVISION = 2; tableTS_R_PARENSTS_L_PARENS = -1; tableTS_R_PARENSTS_R_PARENS = 2; tableTS_R_PARENSTS_I = -1; tableTS_R_PARENSTS_EOS = 2; tableTS_ITS_PLUS = 2; tableTS_ITS_MINUS = 2; tableTS_ITS_MULTIPLY = 2; tableTS_ITS_DIVISION

18、 = 2; tableTS_ITS_L_PARENS = -1; tableTS_ITS_R_PARENS = 2; tableTS_ITS_I = -1; tableTS_ITS_EOS = 2; tableTS_EOSTS_PLUS = 1; tableTS_EOSTS_MINUS = 1; tableTS_EOSTS_MULTIPLY = 1; tableTS_EOSTS_DIVISION = 1; tableTS_EOSTS_L_PARENS = 1; tableTS_EOSTS_R_PARENS = -1; tableTS_EOSTS_I = 1; tableTS_EOSTS_EOS

19、 = 0;int main() setup();/初始化优先关系表 while(1) cout 输入一个以#结束的符号串 str; p = str.c_str(); stack ss; ss.push(TS_EOS); while(1) if(lexer(*p) = TS_INVALID)/读取到不在集合中的非法字符 cout 不是给定算符优先文法的句子 endl; break; if(ss.top() = TS_EOS & lexer(*p) = TS_EOS)/成功规约 cout 是给定算符优先文法的句子 endl; break; else if(tabless.top()lexer(*p

20、) = 1 | tabless.top()lexer(*p) = 0) /继续读入字符，暂不规约 ss.push(lexer(*p); +p; else if(tabless.top()lexer(*p) = 2) /符合规约要求，开始规约 enum Symbols temp; while(1) temp = ss.top(); ss.pop(); if(tabless.top()temp = 1) break; else/语法错误 cout 不是给定算符优先文法的句子 endl; break; return 0;3.4 程序运行截图3.5 小结 此方法初始化优先关系法显得有点繁琐，有待改进。

21、4 LR分析4.1 LR文法 (0) SS (注：该文法为示例，可更改） (1) SBB (2) BaB (3) Bb4.2 LR分析表ACTIONGOTOab#SB0S3S4121acc2S3S453S3S464r3r3r35r1r1r16r2r2r24.3 分析程序代码#include #include #include #include using namespace std;enum Symbols / 终结符号 TS_A, / A TS_B, / B TS_EOS, / #, 0 TS_INVALID, / 非法字符 /非终结符号 NTS_SS, /S NTS_S, /S NTS_B

22、, /B /状态status S_0, S_1, S_2, S_3, S_4, S_5, S_6, S_ACC, /GOGO G_S, G_B, /规约式 R_1, R_2, R_3,;enum Symbols lexer(char c) switch(c) case a: return TS_A; case b: return TS_B; case #: return TS_EOS; / 栈底：#终结符号 default: return TS_INVALID; map enum Symbols, map table; int main() while(1) cout 输入一个以#结束的字符串

23、 str; p = str.c_str(); stack status; /设置LR分析表 tableS_0TS_A = S_3; tableS_0TS_B = S_4; tableS_0NTS_S = S_1; tableS_0NTS_B = S_2; tableS_1TS_EOS = S_ACC; tableS_2TS_A = S_3; tableS_2TS_B = S_4; tableS_2NTS_B = S_5; tableS_3TS_A = S_3; tableS_3TS_B = S_4; tableS_3NTS_B = S_6; tableS_4TS_A = R_3; tableS_4TS_B = R_3;