blob: 2d19b10fd890762283038e8e0f868e4d4d1f6839 [file] [log] [blame]
Andrew Scull5e1ddfa2018-08-14 10:06:54 +01001//===-- Regex.h - Regular Expression matcher implementation -*- C++ -*-----===//
2//
Andrew Walbran16937d02019-10-22 13:54:20 +01003// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
Andrew Scull5e1ddfa2018-08-14 10:06:54 +01006//
7//===----------------------------------------------------------------------===//
8//
9// This file implements a POSIX regular expression matcher. Both Basic and
10// Extended POSIX regular expressions (ERE) are supported. EREs were extended
11// to support backreferences in matches.
12// This implementation also supports matching strings with embedded NUL chars.
13//
14//===----------------------------------------------------------------------===//
15
16#ifndef LLVM_SUPPORT_REGEX_H
17#define LLVM_SUPPORT_REGEX_H
18
19#include <string>
20
21struct llvm_regex;
22
23namespace llvm {
24 class StringRef;
25 template<typename T> class SmallVectorImpl;
26
27 class Regex {
28 public:
29 enum {
30 NoFlags=0,
31 /// Compile for matching that ignores upper/lower case distinctions.
32 IgnoreCase=1,
33 /// Compile for newline-sensitive matching. With this flag '[^' bracket
34 /// expressions and '.' never match newline. A ^ anchor matches the
35 /// null string after any newline in the string in addition to its normal
36 /// function, and the $ anchor matches the null string before any
37 /// newline in the string in addition to its normal function.
38 Newline=2,
39 /// By default, the POSIX extended regular expression (ERE) syntax is
40 /// assumed. Pass this flag to turn on basic regular expressions (BRE)
41 /// instead.
42 BasicRegex=4
43 };
44
45 Regex();
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);
52 return *this;
53 }
54 Regex(Regex &&regex);
55 ~Regex();
56
57 /// isValid - returns the error encountered during regex compilation, or
58 /// matching, if any.
59 bool isValid(std::string &Error) const;
60
61 /// getNumMatches - In a valid regex, return the number of parenthesized
62 /// matches it contains. The number filled in by match will include this
63 /// many entries plus one for the whole regex (as element 0).
64 unsigned getNumMatches() const;
65
66 /// matches - Match the regex against a given \p String.
67 ///
68 /// \param Matches - If given, on a successful match this will be filled in
69 /// with references to the matched group expressions (inside \p String),
70 /// the first group is always the entire pattern.
71 ///
72 /// This returns true on a successful match.
73 bool match(StringRef String, SmallVectorImpl<StringRef> *Matches = nullptr);
74
75 /// sub - Return the result of replacing the first match of the regex in
76 /// \p String with the \p Repl string. Backreferences like "\0" in the
77 /// replacement string are replaced with the appropriate match substring.
78 ///
79 /// Note that the replacement string has backslash escaping performed on
80 /// it. Invalid backreferences are ignored (replaced by empty strings).
81 ///
82 /// \param Error If non-null, any errors in the substitution (invalid
83 /// backreferences, trailing backslashes) will be recorded as a non-empty
84 /// string.
85 std::string sub(StringRef Repl, StringRef String,
86 std::string *Error = nullptr);
87
Andrew Scullcdfcccc2018-10-05 20:58:37 +010088 /// If this function returns true, ^Str$ is an extended regular
Andrew Scull5e1ddfa2018-08-14 10:06:54 +010089 /// expression that matches Str and only Str.
90 static bool isLiteralERE(StringRef Str);
91
Andrew Scullcdfcccc2018-10-05 20:58:37 +010092 /// Turn String into a regex by escaping its special characters.
Andrew Scull5e1ddfa2018-08-14 10:06:54 +010093 static std::string escape(StringRef String);
94
95 private:
96 struct llvm_regex *preg;
97 int error;
98 };
99}
100
101#endif // LLVM_SUPPORT_REGEX_H