jpayne@69: // © 2016 and later: Unicode, Inc. and others.
jpayne@69: // License & terms of use: http://www.unicode.org/copyright.html
jpayne@69: // Copyright (C) 2009-2013, International Business Machines
jpayne@69: // Corporation and others. All Rights Reserved.
jpayne@69: //
jpayne@69: // Copyright 2001 and onwards Google Inc.
jpayne@69: // Author: Sanjay Ghemawat
jpayne@69: 
jpayne@69: // This code is a contribution of Google code, and the style used here is
jpayne@69: // a compromise between the original Google code and the ICU coding guidelines.
jpayne@69: // For example, data types are ICU-ified (size_t,int->int32_t),
jpayne@69: // and API comments doxygen-ified, but function names and behavior are
jpayne@69: // as in the original, if possible.
jpayne@69: // Assertion-style error handling, not available in ICU, was changed to
jpayne@69: // parameter "pinning" similar to UnicodeString.
jpayne@69: //
jpayne@69: // In addition, this is only a partial port of the original Google code,
jpayne@69: // limited to what was needed so far. The (nearly) complete original code
jpayne@69: // is in the ICU svn repository at icuhtml/trunk/design/strings/contrib
jpayne@69: // (see ICU ticket 6765, r25517).
jpayne@69: 
jpayne@69: #ifndef __STRINGPIECE_H__
jpayne@69: #define __STRINGPIECE_H__
jpayne@69: 
jpayne@69: /**
jpayne@69:  * \file 
jpayne@69:  * \brief C++ API: StringPiece: Read-only byte string wrapper class.
jpayne@69:  */
jpayne@69: 
jpayne@69: #include "unicode/utypes.h"
jpayne@69: 
jpayne@69: #if U_SHOW_CPLUSPLUS_API
jpayne@69: 
jpayne@69: #include <cstddef>
jpayne@69: #include <type_traits>
jpayne@69: 
jpayne@69: #include "unicode/uobject.h"
jpayne@69: #include "unicode/std_string.h"
jpayne@69: 
jpayne@69: // Arghh!  I wish C++ literals were "string".
jpayne@69: 
jpayne@69: U_NAMESPACE_BEGIN
jpayne@69: 
jpayne@69: /**
jpayne@69:  * A string-like object that points to a sized piece of memory.
jpayne@69:  *
jpayne@69:  * We provide non-explicit singleton constructors so users can pass
jpayne@69:  * in a "const char*" or a "string" wherever a "StringPiece" is
jpayne@69:  * expected.
jpayne@69:  *
jpayne@69:  * Functions or methods may use StringPiece parameters to accept either a
jpayne@69:  * "const char*" or a "string" value that will be implicitly converted to a
jpayne@69:  * StringPiece.
jpayne@69:  *
jpayne@69:  * Systematic usage of StringPiece is encouraged as it will reduce unnecessary
jpayne@69:  * conversions from "const char*" to "string" and back again.
jpayne@69:  *
jpayne@69:  * @stable ICU 4.2
jpayne@69:  */
jpayne@69: class U_COMMON_API StringPiece : public UMemory {
jpayne@69:  private:
jpayne@69:   const char*   ptr_;
jpayne@69:   int32_t       length_;
jpayne@69: 
jpayne@69:  public:
jpayne@69:   /**
jpayne@69:    * Default constructor, creates an empty StringPiece.
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   StringPiece() : ptr_(nullptr), length_(0) { }
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Constructs from a NUL-terminated const char * pointer.
jpayne@69:    * @param str a NUL-terminated const char * pointer
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   StringPiece(const char* str);
jpayne@69: #ifndef U_HIDE_DRAFT_API
jpayne@69: #if defined(__cpp_char8_t) || defined(U_IN_DOXYGEN)
jpayne@69:   /**
jpayne@69:    * Constructs from a NUL-terminated const char8_t * pointer.
jpayne@69:    * @param str a NUL-terminated const char8_t * pointer
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   StringPiece(const char8_t* str) : StringPiece(reinterpret_cast<const char*>(str)) {}
jpayne@69: #endif
jpayne@69:   /**
jpayne@69:    * Constructs an empty StringPiece.
jpayne@69:    * Needed for type disambiguation from multiple other overloads.
jpayne@69:    * @param p nullptr
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   StringPiece(std::nullptr_t p) : ptr_(p), length_(0) {}
jpayne@69: #endif  // U_HIDE_DRAFT_API
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Constructs from a std::string.
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   StringPiece(const std::string& str)
jpayne@69:     : ptr_(str.data()), length_(static_cast<int32_t>(str.size())) { }
jpayne@69: #ifndef U_HIDE_DRAFT_API
jpayne@69: #if defined(__cpp_lib_char8_t) || defined(U_IN_DOXYGEN)
jpayne@69:   /**
jpayne@69:    * Constructs from a std::u8string.
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   StringPiece(const std::u8string& str)
jpayne@69:     : ptr_(reinterpret_cast<const char*>(str.data())),
jpayne@69:       length_(static_cast<int32_t>(str.size())) { }
jpayne@69: #endif
jpayne@69: #endif  // U_HIDE_DRAFT_API
jpayne@69: 
jpayne@69: #ifndef U_HIDE_DRAFT_API
jpayne@69:   /**
jpayne@69:    * Constructs from some other implementation of a string piece class, from any
jpayne@69:    * C++ record type that has these two methods:
jpayne@69:    *
jpayne@69:    * \code{.cpp}
jpayne@69:    *
jpayne@69:    *   struct OtherStringPieceClass {
jpayne@69:    *     const char* data();  // or const char8_t*
jpayne@69:    *     size_t size();
jpayne@69:    *   };
jpayne@69:    *
jpayne@69:    * \endcode
jpayne@69:    *
jpayne@69:    * The other string piece class will typically be std::string_view from C++17
jpayne@69:    * or absl::string_view from Abseil.
jpayne@69:    *
jpayne@69:    * Starting with C++20, data() may also return a const char8_t* pointer,
jpayne@69:    * as from std::u8string_view.
jpayne@69:    *
jpayne@69:    * @param str the other string piece
jpayne@69:    * @draft ICU 65
jpayne@69:    */
jpayne@69:   template <typename T,
jpayne@69:             typename = typename std::enable_if<
jpayne@69:                 (std::is_same<decltype(T().data()), const char*>::value
jpayne@69: #if defined(__cpp_char8_t)
jpayne@69:                     || std::is_same<decltype(T().data()), const char8_t*>::value
jpayne@69: #endif
jpayne@69:                 ) &&
jpayne@69:                 std::is_same<decltype(T().size()), size_t>::value>::type>
jpayne@69:   StringPiece(T str)
jpayne@69:       : ptr_(reinterpret_cast<const char*>(str.data())),
jpayne@69:         length_(static_cast<int32_t>(str.size())) {}
jpayne@69: #endif  // U_HIDE_DRAFT_API
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Constructs from a const char * pointer and a specified length.
jpayne@69:    * @param offset a const char * pointer (need not be terminated)
jpayne@69:    * @param len the length of the string; must be non-negative
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   StringPiece(const char* offset, int32_t len) : ptr_(offset), length_(len) { }
jpayne@69: #ifndef U_HIDE_DRAFT_API
jpayne@69: #if defined(__cpp_char8_t) || defined(U_IN_DOXYGEN)
jpayne@69:   /**
jpayne@69:    * Constructs from a const char8_t * pointer and a specified length.
jpayne@69:    * @param str a const char8_t * pointer (need not be terminated)
jpayne@69:    * @param len the length of the string; must be non-negative
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   StringPiece(const char8_t* str, int32_t len) :
jpayne@69:       StringPiece(reinterpret_cast<const char*>(str), len) {}
jpayne@69: #endif
jpayne@69: #endif  // U_HIDE_DRAFT_API
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Substring of another StringPiece.
jpayne@69:    * @param x the other StringPiece
jpayne@69:    * @param pos start position in x; must be non-negative and <= x.length().
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   StringPiece(const StringPiece& x, int32_t pos);
jpayne@69:   /**
jpayne@69:    * Substring of another StringPiece.
jpayne@69:    * @param x the other StringPiece
jpayne@69:    * @param pos start position in x; must be non-negative and <= x.length().
jpayne@69:    * @param len length of the substring;
jpayne@69:    *            must be non-negative and will be pinned to at most x.length() - pos.
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   StringPiece(const StringPiece& x, int32_t pos, int32_t len);
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Returns the string pointer. May be nullptr if it is empty.
jpayne@69:    *
jpayne@69:    * data() may return a pointer to a buffer with embedded NULs, and the
jpayne@69:    * returned buffer may or may not be null terminated.  Therefore it is
jpayne@69:    * typically a mistake to pass data() to a routine that expects a NUL
jpayne@69:    * terminated string.
jpayne@69:    * @return the string pointer
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   const char* data() const { return ptr_; }
jpayne@69:   /**
jpayne@69:    * Returns the string length. Same as length().
jpayne@69:    * @return the string length
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   int32_t size() const { return length_; }
jpayne@69:   /**
jpayne@69:    * Returns the string length. Same as size().
jpayne@69:    * @return the string length
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   int32_t length() const { return length_; }
jpayne@69:   /**
jpayne@69:    * Returns whether the string is empty.
jpayne@69:    * @return TRUE if the string is empty
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   UBool empty() const { return length_ == 0; }
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Sets to an empty string.
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   void clear() { ptr_ = nullptr; length_ = 0; }
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Reset the stringpiece to refer to new data.
jpayne@69:    * @param xdata pointer the new string data.  Need not be nul terminated.
jpayne@69:    * @param len the length of the new data
jpayne@69:    * @stable ICU 4.8
jpayne@69:    */
jpayne@69:   void set(const char* xdata, int32_t len) { ptr_ = xdata; length_ = len; }
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Reset the stringpiece to refer to new data.
jpayne@69:    * @param str a pointer to a NUL-terminated string. 
jpayne@69:    * @stable ICU 4.8
jpayne@69:    */
jpayne@69:   void set(const char* str);
jpayne@69: 
jpayne@69: #ifndef U_HIDE_DRAFT_API
jpayne@69: #if defined(__cpp_char8_t) || defined(U_IN_DOXYGEN)
jpayne@69:   /**
jpayne@69:    * Resets the stringpiece to refer to new data.
jpayne@69:    * @param xdata pointer the new string data. Need not be NUL-terminated.
jpayne@69:    * @param len the length of the new data
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   inline void set(const char8_t* xdata, int32_t len) {
jpayne@69:       set(reinterpret_cast<const char*>(xdata), len);
jpayne@69:   }
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Resets the stringpiece to refer to new data.
jpayne@69:    * @param str a pointer to a NUL-terminated string.
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   inline void set(const char8_t* str) {
jpayne@69:       set(reinterpret_cast<const char*>(str));
jpayne@69:   }
jpayne@69: #endif
jpayne@69: #endif  // U_HIDE_DRAFT_API
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Removes the first n string units.
jpayne@69:    * @param n prefix length, must be non-negative and <=length()
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   void remove_prefix(int32_t n) {
jpayne@69:     if (n >= 0) {
jpayne@69:       if (n > length_) {
jpayne@69:         n = length_;
jpayne@69:       }
jpayne@69:       ptr_ += n;
jpayne@69:       length_ -= n;
jpayne@69:     }
jpayne@69:   }
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Removes the last n string units.
jpayne@69:    * @param n suffix length, must be non-negative and <=length()
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   void remove_suffix(int32_t n) {
jpayne@69:     if (n >= 0) {
jpayne@69:       if (n <= length_) {
jpayne@69:         length_ -= n;
jpayne@69:       } else {
jpayne@69:         length_ = 0;
jpayne@69:       }
jpayne@69:     }
jpayne@69:   }
jpayne@69: 
jpayne@69: #ifndef U_HIDE_DRAFT_API
jpayne@69:   /**
jpayne@69:    * Searches the StringPiece for the given search string (needle);
jpayne@69:    * @param needle The string for which to search.
jpayne@69:    * @param offset Where to start searching within this string (haystack).
jpayne@69:    * @return The offset of needle in haystack, or -1 if not found.
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   int32_t find(StringPiece needle, int32_t offset);
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Compares this StringPiece with the other StringPiece, with semantics
jpayne@69:    * similar to std::string::compare().
jpayne@69:    * @param other The string to compare to.
jpayne@69:    * @return below zero if this < other; above zero if this > other; 0 if this == other.
jpayne@69:    * @draft ICU 67
jpayne@69:    */
jpayne@69:   int32_t compare(StringPiece other);
jpayne@69: #endif  // U_HIDE_DRAFT_API
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Maximum integer, used as a default value for substring methods.
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   static const int32_t npos; // = 0x7fffffff;
jpayne@69: 
jpayne@69:   /**
jpayne@69:    * Returns a substring of this StringPiece.
jpayne@69:    * @param pos start position; must be non-negative and <= length().
jpayne@69:    * @param len length of the substring;
jpayne@69:    *            must be non-negative and will be pinned to at most length() - pos.
jpayne@69:    * @return the substring StringPiece
jpayne@69:    * @stable ICU 4.2
jpayne@69:    */
jpayne@69:   StringPiece substr(int32_t pos, int32_t len = npos) const {
jpayne@69:     return StringPiece(*this, pos, len);
jpayne@69:   }
jpayne@69: };
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Global operator == for StringPiece
jpayne@69:  * @param x The first StringPiece to compare.
jpayne@69:  * @param y The second StringPiece to compare.
jpayne@69:  * @return TRUE if the string data is equal
jpayne@69:  * @stable ICU 4.8
jpayne@69:  */
jpayne@69: U_EXPORT UBool U_EXPORT2 
jpayne@69: operator==(const StringPiece& x, const StringPiece& y);
jpayne@69: 
jpayne@69: /**
jpayne@69:  * Global operator != for StringPiece
jpayne@69:  * @param x The first StringPiece to compare.
jpayne@69:  * @param y The second StringPiece to compare.
jpayne@69:  * @return TRUE if the string data is not equal
jpayne@69:  * @stable ICU 4.8
jpayne@69:  */
jpayne@69: inline UBool operator!=(const StringPiece& x, const StringPiece& y) {
jpayne@69:   return !(x == y);
jpayne@69: }
jpayne@69: 
jpayne@69: U_NAMESPACE_END
jpayne@69: 
jpayne@69: #endif /* U_SHOW_CPLUSPLUS_API */
jpayne@69: 
jpayne@69: #endif  // __STRINGPIECE_H__