Home | History | Annotate | Download | only in re2
      1 // Copyright 2001-2010 The RE2 Authors.  All Rights Reserved.
      2 // Use of this source code is governed by a BSD-style
      3 // license that can be found in the LICENSE file.
      4 
      5 // A string-like object that points to a sized piece of memory.
      6 //
      7 // Functions or methods may use const StringPiece& parameters to accept either
      8 // a "const char*" or a "string" value that will be implicitly converted to
      9 // a StringPiece.  The implicit conversion means that it is often appropriate
     10 // to include this .h file in other files rather than forward-declaring
     11 // StringPiece as would be appropriate for most other Google classes.
     12 //
     13 // Systematic usage of StringPiece is encouraged as it will reduce unnecessary
     14 // conversions from "const char*" to "string" and back again.
     15 //
     16 //
     17 // Arghh!  I wish C++ literals were "string".
     18 
     19 #ifndef STRINGS_STRINGPIECE_H__
     20 #define STRINGS_STRINGPIECE_H__
     21 
     22 #include <string.h>
     23 #include <cstddef>
     24 #include <iosfwd>
     25 #include <string>
     26 #ifdef WIN32
     27 #include <algorithm>
     28 #endif
     29 
     30 namespace re2 {
     31 
     32 class StringPiece {
     33  private:
     34   const char*   ptr_;
     35   int           length_;
     36 
     37  public:
     38   // We provide non-explicit singleton constructors so users can pass
     39   // in a "const char*" or a "string" wherever a "StringPiece" is
     40   // expected.
     41   StringPiece() : ptr_(NULL), length_(0) { }
     42   StringPiece(const char* str)
     43     : ptr_(str), length_((str == NULL) ? 0 : static_cast<int>(strlen(str))) { }
     44   StringPiece(const std::string& str)
     45     : ptr_(str.data()), length_(static_cast<int>(str.size())) { }
     46   StringPiece(const char* offset, int len) : ptr_(offset), length_(len) { }
     47 
     48   // data() may return a pointer to a buffer with embedded NULs, and the
     49   // returned buffer may or may not be null terminated.  Therefore it is
     50   // typically a mistake to pass data() to a routine that expects a NUL
     51   // terminated string.
     52   const char* data() const { return ptr_; }
     53   int size() const { return length_; }
     54   int length() const { return length_; }
     55   bool empty() const { return length_ == 0; }
     56 
     57   void clear() { ptr_ = NULL; length_ = 0; }
     58   void set(const char* data, int len) { ptr_ = data; length_ = len; }
     59   void set(const char* str) {
     60     ptr_ = str;
     61     if (str != NULL)
     62       length_ = static_cast<int>(strlen(str));
     63     else
     64       length_ = 0;
     65   }
     66   void set(const void* data, int len) {
     67     ptr_ = reinterpret_cast<const char*>(data);
     68     length_ = len;
     69   }
     70 
     71   char operator[](int i) const { return ptr_[i]; }
     72 
     73   void remove_prefix(int n) {
     74     ptr_ += n;
     75     length_ -= n;
     76   }
     77 
     78   void remove_suffix(int n) {
     79     length_ -= n;
     80   }
     81 
     82   int compare(const StringPiece& x) const {
     83     int r = memcmp(ptr_, x.ptr_, std::min(length_, x.length_));
     84     if (r == 0) {
     85       if (length_ < x.length_) r = -1;
     86       else if (length_ > x.length_) r = +1;
     87     }
     88     return r;
     89   }
     90 
     91   std::string as_string() const {
     92     return std::string(data(), size());
     93   }
     94   // We also define ToString() here, since many other string-like
     95   // interfaces name the routine that converts to a C++ string
     96   // "ToString", and it's confusing to have the method that does that
     97   // for a StringPiece be called "as_string()".  We also leave the
     98   // "as_string()" method defined here for existing code.
     99   std::string ToString() const {
    100     return std::string(data(), size());
    101   }
    102 
    103   void CopyToString(std::string* target) const;
    104   void AppendToString(std::string* target) const;
    105 
    106   // Does "this" start with "x"
    107   bool starts_with(const StringPiece& x) const {
    108     return ((length_ >= x.length_) &&
    109             (memcmp(ptr_, x.ptr_, x.length_) == 0));
    110   }
    111 
    112   // Does "this" end with "x"
    113   bool ends_with(const StringPiece& x) const {
    114     return ((length_ >= x.length_) &&
    115             (memcmp(ptr_ + (length_-x.length_), x.ptr_, x.length_) == 0));
    116   }
    117 
    118   // standard STL container boilerplate
    119   typedef char value_type;
    120   typedef const char* pointer;
    121   typedef const char& reference;
    122   typedef const char& const_reference;
    123   typedef size_t size_type;
    124   typedef ptrdiff_t difference_type;
    125   static const size_type npos;
    126   typedef const char* const_iterator;
    127   typedef const char* iterator;
    128   typedef std::reverse_iterator<const_iterator> const_reverse_iterator;
    129   typedef std::reverse_iterator<iterator> reverse_iterator;
    130   iterator begin() const { return ptr_; }
    131   iterator end() const { return ptr_ + length_; }
    132   const_reverse_iterator rbegin() const {
    133     return const_reverse_iterator(ptr_ + length_);
    134   }
    135   const_reverse_iterator rend() const {
    136     return const_reverse_iterator(ptr_);
    137   }
    138   // STLS says return size_type, but Google says return int
    139   int max_size() const { return length_; }
    140   int capacity() const { return length_; }
    141 
    142   int copy(char* buf, size_type n, size_type pos = 0) const;
    143 
    144   int find(const StringPiece& s, size_type pos = 0) const;
    145   int find(char c, size_type pos = 0) const;
    146   int rfind(const StringPiece& s, size_type pos = npos) const;
    147   int rfind(char c, size_type pos = npos) const;
    148 
    149   StringPiece substr(size_type pos, size_type n = npos) const;
    150 
    151   static bool _equal(const StringPiece&, const StringPiece&);
    152 };
    153 
    154 inline bool operator==(const StringPiece& x, const StringPiece& y) {
    155   return StringPiece::_equal(x, y);
    156 }
    157 
    158 inline bool operator!=(const StringPiece& x, const StringPiece& y) {
    159   return !(x == y);
    160 }
    161 
    162 inline bool operator<(const StringPiece& x, const StringPiece& y) {
    163   const int r = memcmp(x.data(), y.data(),
    164                        std::min(x.size(), y.size()));
    165   return ((r < 0) || ((r == 0) && (x.size() < y.size())));
    166 }
    167 
    168 inline bool operator>(const StringPiece& x, const StringPiece& y) {
    169   return y < x;
    170 }
    171 
    172 inline bool operator<=(const StringPiece& x, const StringPiece& y) {
    173   return !(x > y);
    174 }
    175 
    176 inline bool operator>=(const StringPiece& x, const StringPiece& y) {
    177   return !(x < y);
    178 }
    179 
    180 }  // namespace re2
    181 
    182 // allow StringPiece to be logged
    183 extern std::ostream& operator<<(std::ostream& o, const re2::StringPiece& piece);
    184 
    185 #endif  // STRINGS_STRINGPIECE_H__
    186