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. 11 // 12 //===----------------------------------------------------------------------===// 13 14 #ifndef LLVM_SUPPORT_REGEX_H 15 #define LLVM_SUPPORT_REGEX_H 16 17 #include <string> 18 19 struct llvm_regex; 20 21 namespace llvm { 22 class StringRef; 23 template<typename T> class SmallVectorImpl; 24 25 class Regex { 26 public: 27 enum { 28 NoFlags=0, 29 /// Compile for matching that ignores upper/lower case distinctions. 30 IgnoreCase=1, 31 /// Compile for newline-sensitive matching. With this flag '[^' bracket 32 /// expressions and '.' never match newline. A ^ anchor matches the 33 /// null string after any newline in the string in addition to its normal 34 /// function, and the $ anchor matches the null string before any 35 /// newline in the string in addition to its normal function. 36 Newline=2 37 }; 38 39 /// Compiles the given POSIX Extended Regular Expression \arg Regex. 40 /// This implementation supports regexes and matching strings with embedded 41 /// NUL characters. 42 Regex(StringRef Regex, unsigned Flags = NoFlags); 43 ~Regex(); 44 45 /// isValid - returns the error encountered during regex compilation, or 46 /// matching, if any. 47 bool isValid(std::string &Error); 48 49 /// getNumMatches - In a valid regex, return the number of parenthesized 50 /// matches it contains. The number filled in by match will include this 51 /// many entries plus one for the whole regex (as element 0). 52 unsigned getNumMatches() const; 53 54 /// matches - Match the regex against a given \arg String. 55 /// 56 /// \param Matches - If given, on a successful match this will be filled in 57 /// with references to the matched group expressions (inside \arg String), 58 /// the first group is always the entire pattern. 59 /// 60 /// This returns true on a successful match. 61 bool match(StringRef String, SmallVectorImpl<StringRef> *Matches = 0); 62 63 /// sub - Return the result of replacing the first match of the regex in 64 /// \arg String with the \arg Repl string. Backreferences like "\0" in the 65 /// replacement string are replaced with the appropriate match substring. 66 /// 67 /// Note that the replacement string has backslash escaping performed on 68 /// it. Invalid backreferences are ignored (replaced by empty strings). 69 /// 70 /// \param Error If non-null, any errors in the substitution (invalid 71 /// backreferences, trailing backslashes) will be recorded as a non-empty 72 /// string. 73 std::string sub(StringRef Repl, StringRef String, std::string *Error = 0); 74 75 private: 76 struct llvm_regex *preg; 77 int error; 78 }; 79 } 80 81 #endif // LLVM_SUPPORT_REGEX_H 82
