1//===-- Regex.h - Regular Expression matcher implementation -*- C++ -*-----===//
2//
3//                     The LLVM Compiler Infrastructure
4//
5// This file is distributed under the University of Illinois Open Source
6// License. See LICENSE.TXT for details.
7//
8//===----------------------------------------------------------------------===//
9//
10// This file implements a POSIX regular expression matcher.  Both Basic and
11// Extended POSIX regular expressions (ERE) are supported.  EREs were extended
12// to support backreferences in matches.
13// This implementation also supports matching strings with embedded NUL chars.
14//
15//===----------------------------------------------------------------------===//
16
17#ifndef LLVM_SUPPORT_REGEX_H
18#define LLVM_SUPPORT_REGEX_H
19
20#include "llvm/Support/Compiler.h"
21#include <string>
22
23struct llvm_regex;
24
25namespace llvm {
26  class StringRef;
27  template<typename T> class SmallVectorImpl;
28
29  class Regex {
30  public:
31    enum {
32      NoFlags=0,
33      /// Compile for matching that ignores upper/lower case distinctions.
34      IgnoreCase=1,
35      /// Compile for newline-sensitive matching. With this flag '[^' bracket
36      /// expressions and '.' never match newline. A ^ anchor matches the
37      /// null string after any newline in the string in addition to its normal
38      /// function, and the $ anchor matches the null string before any
39      /// newline in the string in addition to its normal function.
40      Newline=2,
41      /// By default, the POSIX extended regular expression (ERE) syntax is
42      /// assumed. Pass this flag to turn on basic regular expressions (BRE)
43      /// instead.
44      BasicRegex=4
45    };
46
47    /// Compiles the given regular expression \p Regex.
48    Regex(StringRef Regex, unsigned Flags = NoFlags);
49    Regex(const Regex &) LLVM_DELETED_FUNCTION;
50    Regex &operator=(Regex regex) {
51      std::swap(preg, regex.preg);
52      std::swap(error, regex.error);
53      return *this;
54    }
55    Regex(Regex &&regex) {
56      preg = regex.preg;
57      error = regex.error;
58      regex.preg = nullptr;
59    }
60    ~Regex();
61
62    /// isValid - returns the error encountered during regex compilation, or
63    /// matching, if any.
64    bool isValid(std::string &Error);
65
66    /// getNumMatches - In a valid regex, return the number of parenthesized
67    /// matches it contains.  The number filled in by match will include this
68    /// many entries plus one for the whole regex (as element 0).
69    unsigned getNumMatches() const;
70
71    /// matches - Match the regex against a given \p String.
72    ///
73    /// \param Matches - If given, on a successful match this will be filled in
74    /// with references to the matched group expressions (inside \p String),
75    /// the first group is always the entire pattern.
76    ///
77    /// This returns true on a successful match.
78    bool match(StringRef String, SmallVectorImpl<StringRef> *Matches = nullptr);
79
80    /// sub - Return the result of replacing the first match of the regex in
81    /// \p String with the \p Repl string. Backreferences like "\0" in the
82    /// replacement string are replaced with the appropriate match substring.
83    ///
84    /// Note that the replacement string has backslash escaping performed on
85    /// it. Invalid backreferences are ignored (replaced by empty strings).
86    ///
87    /// \param Error If non-null, any errors in the substitution (invalid
88    /// backreferences, trailing backslashes) will be recorded as a non-empty
89    /// string.
90    std::string sub(StringRef Repl, StringRef String,
91                    std::string *Error = nullptr);
92
93    /// \brief If this function returns true, ^Str$ is an extended regular
94    /// expression that matches Str and only Str.
95    static bool isLiteralERE(StringRef Str);
96
97    /// \brief Turn String into a regex by escaping its special characters.
98    static std::string escape(StringRef String);
99
100  private:
101    struct llvm_regex *preg;
102    int error;
103  };
104}
105
106#endif // LLVM_SUPPORT_REGEX_H
107