diff options
author | Mike Buland <eichlan@xagasoft.com> | 2012-03-25 20:00:08 +0000 |
---|---|---|
committer | Mike Buland <eichlan@xagasoft.com> | 2012-03-25 20:00:08 +0000 |
commit | 469bbcf0701e1eb8a6670c23145b0da87357e178 (patch) | |
tree | b5b062a16e46a6c5d3410b4e574cd0cc09057211 /src/experimental/parser.h | |
parent | ee1b79396076edc4e30aefb285fada03bb45e80d (diff) | |
download | libbu++-469bbcf0701e1eb8a6670c23145b0da87357e178.tar.gz libbu++-469bbcf0701e1eb8a6670c23145b0da87357e178.tar.bz2 libbu++-469bbcf0701e1eb8a6670c23145b0da87357e178.tar.xz libbu++-469bbcf0701e1eb8a6670c23145b0da87357e178.zip |
Code is all reorganized. We're about ready to release. I should write up a
little explenation of the arrangement.
Diffstat (limited to 'src/experimental/parser.h')
-rw-r--r-- | src/experimental/parser.h | 135 |
1 files changed, 135 insertions, 0 deletions
diff --git a/src/experimental/parser.h b/src/experimental/parser.h new file mode 100644 index 0000000..a168c7b --- /dev/null +++ b/src/experimental/parser.h | |||
@@ -0,0 +1,135 @@ | |||
1 | /* | ||
2 | * Copyright (C) 2007-2011 Xagasoft, All rights reserved. | ||
3 | * | ||
4 | * This file is part of the libbu++ library and is released under the | ||
5 | * terms of the license contained in the file LICENSE. | ||
6 | */ | ||
7 | |||
8 | #ifndef BU_PARSER_H | ||
9 | #define BU_PARSER_H | ||
10 | |||
11 | #include "bu/string.h" | ||
12 | #include "bu/list.h" | ||
13 | #include "bu/array.h" | ||
14 | #include "bu/hash.h" | ||
15 | #include "bu/signals.h" | ||
16 | |||
17 | #include "bu/lexer.h" | ||
18 | |||
19 | namespace Bu | ||
20 | { | ||
21 | /** | ||
22 | * The base framework for a LR(1) grammar parser. Provided a proper set of | ||
23 | * ParserStates this will prase any input the lexer can provide. | ||
24 | */ | ||
25 | class Parser | ||
26 | { | ||
27 | public: | ||
28 | Parser(); | ||
29 | virtual ~Parser(); | ||
30 | |||
31 | /** | ||
32 | * When a Lexer is pushed onto the stack it becomes the source for | ||
33 | * future tokens read by the parser until it is popped off the stack. | ||
34 | * The Parser takes ownership of every Lexer pushed onto the stack, | ||
35 | * and will delete it when it is popped off the stack. | ||
36 | */ | ||
37 | void pushLexer( Lexer *pLex ); | ||
38 | |||
39 | /** | ||
40 | * Pop a lexer off the stack, and delete it. | ||
41 | */ | ||
42 | void popLexer(); | ||
43 | |||
44 | Lexer::Token *popToken(); | ||
45 | void pushToken( Lexer::Token *pTok ); | ||
46 | |||
47 | /** | ||
48 | * Execute a parse. | ||
49 | */ | ||
50 | void parse(); | ||
51 | |||
52 | void setRootNonTerminal( int iRoot ); | ||
53 | void setRootNonTerminal( const Bu::String &sRoot ); | ||
54 | |||
55 | typedef Bu::Signal1<void, Parser &> Reduction; | ||
56 | |||
57 | /** | ||
58 | * Represents a possible state, either a terminal or non-terminal symbol | ||
59 | * in a Production. | ||
60 | */ | ||
61 | class State | ||
62 | { | ||
63 | public: | ||
64 | enum Type | ||
65 | { | ||
66 | typeTerminal, | ||
67 | typeTerminalPush, | ||
68 | typeNonTerminal, | ||
69 | typeReduction | ||
70 | }; | ||
71 | |||
72 | State( Type eType, int iIndex ); | ||
73 | virtual ~State(); | ||
74 | |||
75 | //private: | ||
76 | Type eType; | ||
77 | int iIndex; | ||
78 | }; | ||
79 | |||
80 | typedef Bu::List<State> Production; | ||
81 | |||
82 | class NonTerminal | ||
83 | { | ||
84 | public: | ||
85 | NonTerminal(); | ||
86 | virtual ~NonTerminal(); | ||
87 | |||
88 | void addProduction( Production p ); | ||
89 | void setCanSkip(); | ||
90 | |||
91 | // private: | ||
92 | typedef Bu::List<Production> ProductionList; | ||
93 | ProductionList lProduction; | ||
94 | bool bCanSkip; | ||
95 | }; | ||
96 | |||
97 | int addNonTerminal( const Bu::String &sName, NonTerminal &nt ); | ||
98 | int addNonTerminal( const Bu::String &sName ); | ||
99 | void setNonTerminal( const Bu::String &sName, NonTerminal &nt ); | ||
100 | int getNonTerminalId( const Bu::String &sName ); | ||
101 | bool hasNonTerminal( const Bu::String &sName ); | ||
102 | |||
103 | int addReduction( const Bu::String &sName, const Reduction &r ); | ||
104 | int addReduction( const Bu::String &sName ); | ||
105 | void setReduction( const Bu::String &sName, const Reduction &r ); | ||
106 | int getReductionId( const Bu::String &sName ); | ||
107 | bool hasReduction( const Bu::String &sName ); | ||
108 | |||
109 | private: | ||
110 | bool selectProduction( int iNt, Lexer::Token *ptCur ); | ||
111 | void advanceState(); | ||
112 | |||
113 | private: | ||
114 | typedef Bu::List<Lexer *> LexerStack; | ||
115 | typedef Bu::List<Lexer::Token *> TokenStack; | ||
116 | typedef Bu::List<Production::const_iterator> StateStack; | ||
117 | typedef Bu::Array<Reduction> ReductionArray; | ||
118 | typedef Bu::Hash<Bu::String,int> NameIndexHash; | ||
119 | typedef Bu::Array<NonTerminal> NonTerminalArray; | ||
120 | |||
121 | LexerStack sLexer; | ||
122 | TokenStack sToken; | ||
123 | StateStack sState; | ||
124 | ReductionArray aReduction; | ||
125 | NameIndexHash hReductionName; | ||
126 | NonTerminalArray aNonTerminal; | ||
127 | NameIndexHash hNonTerminalName; | ||
128 | int iRootNonTerminal; | ||
129 | }; | ||
130 | Bu::Formatter &operator<<( Bu::Formatter &f, Bu::Parser::State::Type t ); | ||
131 | Bu::Formatter &operator<<( Bu::Formatter &f, const Bu::Parser::State &s ); | ||
132 | }; | ||
133 | |||
134 | |||
135 | #endif | ||