1 //===-- Regex.h - Regular Expression matcher implementation -*- C++ -*-----===// 2 // 3 // 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 6 // 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 "llvm/ADT/BitmaskEnum.h" 20 #include "llvm/Support/Compiler.h" 21 #include <string> 22 23 struct llvm_regex; 24 25 namespace llvm { 26 class StringRef; 27 template<typename T> class SmallVectorImpl; 28 29 class Regex { 30 public: 31 enum RegexFlags : unsigned { 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 LLVM_MARK_AS_BITMASK_ENUM(BasicRegex) 47 }; 48 49 LLVM_ABI Regex(); 50 /// Compiles the given regular expression \p Regex. 51 /// 52 /// \param Regex - referenced string is no longer needed after this 53 /// constructor does finish. Only its compiled form is kept stored. 54 LLVM_ABI Regex(StringRef Regex, RegexFlags Flags = NoFlags); 55 LLVM_ABI Regex(StringRef Regex, unsigned Flags); 56 Regex(const Regex &) = delete; 57 Regex &operator=(Regex regex) { 58 std::swap(preg, regex.preg); 59 std::swap(error, regex.error); 60 return *this; 61 } 62 LLVM_ABI Regex(Regex &®ex); 63 LLVM_ABI ~Regex(); 64 65 /// isValid - returns the error encountered during regex compilation, if 66 /// any. 67 LLVM_ABI bool isValid(std::string &Error) const; isValid()68 bool isValid() const { return !error; } 69 70 /// getNumMatches - In a valid regex, return the number of parenthesized 71 /// matches it contains. The number filled in by match will include this 72 /// many entries plus one for the whole regex (as element 0). 73 LLVM_ABI unsigned getNumMatches() const; 74 75 /// matches - Match the regex against a given \p String. 76 /// 77 /// \param Matches - If given, on a successful match this will be filled in 78 /// with references to the matched group expressions (inside \p String), 79 /// the first group is always the entire pattern. 80 /// 81 /// \param Error - If non-null, any errors in the matching will be recorded 82 /// as a non-empty string. If there is no error, it will be an empty string. 83 /// 84 /// This returns true on a successful match. 85 LLVM_ABI bool match(StringRef String, 86 SmallVectorImpl<StringRef> *Matches = nullptr, 87 std::string *Error = nullptr) const; 88 89 /// sub - Return the result of replacing the first match of the regex in 90 /// \p String with the \p Repl string. Backreferences like "\0" and "\g<1>" 91 /// in the replacement string are replaced with the appropriate match 92 /// substring. 93 /// 94 /// Note that the replacement string has backslash escaping performed on 95 /// it. Invalid backreferences are ignored (replaced by empty strings). 96 /// 97 /// \param Error If non-null, any errors in the substitution (invalid 98 /// backreferences, trailing backslashes) will be recorded as a non-empty 99 /// string. If there is no error, it will be an empty string. 100 LLVM_ABI std::string sub(StringRef Repl, StringRef String, 101 std::string *Error = nullptr) const; 102 103 /// If this function returns true, ^Str$ is an extended regular 104 /// expression that matches Str and only Str. 105 LLVM_ABI static bool isLiteralERE(StringRef Str); 106 107 /// Turn String into a regex by escaping its special characters. 108 LLVM_ABI static std::string escape(StringRef String); 109 110 private: 111 struct llvm_regex *preg; 112 int error; 113 }; 114 } 115 116 #endif // LLVM_SUPPORT_REGEX_H 117