1 //===-- Regex.h - Regular Expression matcher implementation -*- C++ -*-----===//
3 // The LLVM Compiler Infrastructure
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
8 //===----------------------------------------------------------------------===//
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.
15 //===----------------------------------------------------------------------===//
17 #ifndef LLVM_SUPPORT_REGEX_H
18 #define LLVM_SUPPORT_REGEX_H
26 template<typename T> class SmallVectorImpl;
32 /// Compile for matching that ignores upper/lower case distinctions.
34 /// Compile for newline-sensitive matching. With this flag '[^' bracket
35 /// expressions and '.' never match newline. A ^ anchor matches the
36 /// null string after any newline in the string in addition to its normal
37 /// function, and the $ anchor matches the null string before any
38 /// newline in the string in addition to its normal function.
40 /// By default, the POSIX extended regular expression (ERE) syntax is
41 /// assumed. Pass this flag to turn on basic regular expressions (BRE)
46 /// Compiles the given regular expression \p Regex.
47 Regex(StringRef Regex, unsigned Flags = NoFlags);
48 Regex(const Regex &) = delete;
49 Regex &operator=(Regex regex) {
50 std::swap(preg, regex.preg);
51 std::swap(error, regex.error);
54 Regex(Regex &®ex) {
61 /// isValid - returns the error encountered during regex compilation, or
63 bool isValid(std::string &Error);
65 /// getNumMatches - In a valid regex, return the number of parenthesized
66 /// matches it contains. The number filled in by match will include this
67 /// many entries plus one for the whole regex (as element 0).
68 unsigned getNumMatches() const;
70 /// matches - Match the regex against a given \p String.
72 /// \param Matches - If given, on a successful match this will be filled in
73 /// with references to the matched group expressions (inside \p String),
74 /// the first group is always the entire pattern.
76 /// This returns true on a successful match.
77 bool match(StringRef String, SmallVectorImpl<StringRef> *Matches = nullptr);
79 /// sub - Return the result of replacing the first match of the regex in
80 /// \p String with the \p Repl string. Backreferences like "\0" in the
81 /// replacement string are replaced with the appropriate match substring.
83 /// Note that the replacement string has backslash escaping performed on
84 /// it. Invalid backreferences are ignored (replaced by empty strings).
86 /// \param Error If non-null, any errors in the substitution (invalid
87 /// backreferences, trailing backslashes) will be recorded as a non-empty
89 std::string sub(StringRef Repl, StringRef String,
90 std::string *Error = nullptr);
92 /// \brief If this function returns true, ^Str$ is an extended regular
93 /// expression that matches Str and only Str.
94 static bool isLiteralERE(StringRef Str);
96 /// \brief Turn String into a regex by escaping its special characters.
97 static std::string escape(StringRef String);
100 struct llvm_regex *preg;
105 #endif // LLVM_SUPPORT_REGEX_H