1 /****************************************************************************
2 **
3 ** Copyright (C) 2016 The Qt Company Ltd.
4 ** Contact: https://www.qt.io/licensing/
5 **
6 ** This file is part of the utils of the Qt Toolkit.
7 **
8 ** $QT_BEGIN_LICENSE:GPL-EXCEPT$
9 ** Commercial License Usage
10 ** Licensees holding valid commercial Qt licenses may use this file in
11 ** accordance with the commercial license agreement provided with the
12 ** Software or, alternatively, in accordance with the terms contained in
13 ** a written agreement between you and The Qt Company. For licensing terms
14 ** and conditions see https://www.qt.io/terms-conditions. For further
15 ** information use the contact form at https://www.qt.io/contact-us.
16 **
17 ** GNU General Public License Usage
18 ** Alternatively, this file may be used under the terms of the GNU
19 ** General Public License version 3 as published by the Free Software
20 ** Foundation with exceptions as appearing in the file LICENSE.GPL3-EXCEPT
21 ** included in the packaging of this file. Please review the following
22 ** information to ensure the GNU General Public License requirements will
23 ** be met: https://www.gnu.org/licenses/gpl-3.0.html.
24 **
25 ** $QT_END_LICENSE$
26 **
27 ****************************************************************************/
28 
29 #ifndef RE2NFA_H
30 #define RE2NFA_H
31 
32 #include "nfa.h"
33 #include <QSet>
34 
35 class RE2NFA
36 {
37 public:
38     RE2NFA(const QMap<QString, NFA> &macros, const QSet<InputType> &maxInputSet, Qt::CaseSensitivity cs);
39 
40     NFA parse(const QString &expression, int *errorColumn = 0);
41 
42 private:
43     NFA parseExpr();
44     NFA parseBranch();
45     NFA parsePiece();
46     NFA parseAtom();
47     NFA parseMaybeQuantifier(const NFA &nfa);
48     NFA parseSet();
49     NFA parseSet2();
50 
51     NFA createCharNFA();
52 
53 private:
54     friend class RegExpTokenizer;
55 
56     enum Token {
57         TOK_INVALID,
58         TOK_STRING,
59         TOK_LBRACE,   // {
60         TOK_RBRACE,   // }
61         TOK_LBRACKET, // [
62         TOK_RBRACKET, // ]
63         TOK_LPAREN,   // (
64         TOK_RPAREN,   // )
65         TOK_COMMA,
66         TOK_STAR,
67         TOK_OR,
68         TOK_QUESTION,
69         TOK_DOT,
70         TOK_PLUS,
71         TOK_SEQUENCE,
72         TOK_QUOTED_STRING
73     };
74 
75     struct Symbol
76     {
SymbolSymbol77         inline Symbol() : token(TOK_INVALID), column(-1) {}
tokenSymbol78         inline Symbol(Token t, const QString &l = QString()) : token(t), lexem(l), column(-1) {}
79         Token token;
80         QString lexem;
81         int column;
82     };
83 
hasNext()84     inline bool hasNext() const { return index < symbols.count(); }
next()85     inline Token next() { return symbols.at(index++).token; }
86     bool next(Token t);
87     bool test(Token t);
prev()88     inline void prev() { index--; }
symbol()89     inline const Symbol &symbol() const { return symbols.at(index - 1); }
90     QString lexemUntil(Token t);
91 
92     void tokenize(const QString &input);
93 
94     QMap<QString, NFA> macros;
95     QVector<Symbol> symbols;
96     int index;
97     int errorColumn;
98     const QSet<InputType> maxInputSet;
99     Qt::CaseSensitivity caseSensitivity;
100 };
101 
102 #endif // RE2NFA_H
103 
104