string_piece.h

// Copyright (c) 2011 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
// Copied from strings/stringpiece.h with modifications
//
// A string-like object that points to a sized piece of memory.
//
// Functions or methods may use const StringPiece& parameters to accept either
// a "const char*" or a "string" value that will be implicitly converted to
// a StringPiece.  The implicit conversion means that it is often appropriate
// to include this .h file in other files rather than forward-declaring
// StringPiece as would be appropriate for most other Google classes.
//
// Systematic usage of StringPiece is encouraged as it will reduce unnecessary
// conversions from "const char*" to "string" and back again.
//

#ifndef BASE_STRING_PIECE_H_
#define BASE_STRING_PIECE_H_
#pragma once

#include <string>

//#include <base_api.h>
// Copyright (c) 2011 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#ifndef BASE_BASE_API_H_
#define BASE_BASE_API_H_
#pragma once

#if defined(BASE_DLL)
#if defined(WIN32)
#if defined(BASE_IMPLEMENTATION)
#define BASE_API __declspec(dllexport)
#else
#define BASE_API __declspec(dllimport)
#endif  // defined(BASE_IMPLEMENTATION)
#else
#define BASE_API __attribute__((visibility("default")))
#endif  // defined(WIN32)
#else
#define BASE_API
#endif

#endif  // BASE_BASE_API_H_

//#include <basictypes.h>
// Copyright (c) 2011 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#ifndef BASE_BASICTYPES_H_
#define BASE_BASICTYPES_H_
#pragma once

#include <limits.h>         // So we can set the bounds of our types
#include <stddef.h>         // For size_t
#include <string.h>         // for memcpy

//#include <port.h>    // Types that only need exist on certain systems
// Copyright (c) 2006-2008 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#ifndef BASE_PORT_H_
#define BASE_PORT_H_
#pragma once

#include <stdarg.h>
//#include "build/build_config.h"

#ifdef COMPILER_MSVC
#define GG_LONGLONG(x) x##I64
#define GG_ULONGLONG(x) x##UI64
#else
#define GG_LONGLONG(x) x##LL
#define GG_ULONGLONG(x) x##ULL
#endif

// Per C99 7.8.14, define __STDC_CONSTANT_MACROS before including <stdint.h>
// to get the INTn_C and UINTn_C macros for integer constants.  It's difficult
// to guarantee any specific ordering of header includes, so it's difficult to
// guarantee that the INTn_C macros can be defined by including <stdint.h> at
// any specific point.  Provide GG_INTn_C macros instead.

#define GG_INT8_C(x)    (x)
#define GG_INT16_C(x)   (x)
#define GG_INT32_C(x)   (x)
#define GG_INT64_C(x)   GG_LONGLONG(x)

#define GG_UINT8_C(x)   (x ## U)
#define GG_UINT16_C(x)  (x ## U)
#define GG_UINT32_C(x)  (x ## U)
#define GG_UINT64_C(x)  GG_ULONGLONG(x)

// It's possible for functions that use a va_list, such as StringPrintf, to
// invalidate the data in it upon use.  The fix is to make a copy of the
// structure before using it and use that copy instead.  va_copy is provided
// for this purpose.  MSVC does not provide va_copy, so define an
// implementation here.  It is not guaranteed that assignment is a copy, so the
// StringUtil.VariableArgsFunc unit test tests this capability.
#if defined(COMPILER_GCC)
#define GG_VA_COPY(a, b) (va_copy(a, b))
#elif defined(COMPILER_MSVC)
#define GG_VA_COPY(a, b) (a = b)
#endif

// Define an OS-neutral wrapper for shared library entry points
#if defined(OS_WIN)
#define API_CALL __stdcall
#else
#define API_CALL
#endif

#endif  // BASE_PORT_H_

#ifndef COMPILER_MSVC
// stdint.h is part of C99 but MSVC doesn't have it.
#include <stdint.h>         // For intptr_t.
#endif

typedef signed char         schar;
typedef signed char         int8;
typedef short               int16;
// TODO: Remove these type guards.  These are to avoid conflicts with
// obsolete/protypes.h in the Gecko SDK.
#ifndef _INT32
#define _INT32
typedef int                 int32;
#endif

// The NSPR system headers define 64-bit as |long| when possible, except on
// Mac OS X.  In order to not have typedef mismatches, we do the same on LP64.
//
// On Mac OS X, |long long| is used for 64-bit types for compatibility with
// <inttypes.h> format macros even in the LP64 model.
#if defined(__LP64__) && !defined(OS_MACOSX)
typedef long                int64;
#else
typedef long long           int64;
#endif

// NOTE: unsigned types are DANGEROUS in loops and other arithmetical
// places.  Use the signed types unless your variable represents a bit
// pattern (eg a hash value) or you really need the extra bit.  Do NOT
// use 'unsigned' to express "this value should always be positive";
// use assertions for this.

typedef unsigned char      uint8;
typedef unsigned short     uint16;
// TODO: Remove these type guards.  These are to avoid conflicts with
// obsolete/protypes.h in the Gecko SDK.
#ifndef _UINT32
#define _UINT32
typedef unsigned int       uint32;
#endif

// See the comment above about NSPR and 64-bit.
#if defined(__LP64__) && !defined(OS_MACOSX)
typedef unsigned long uint64;
#else
typedef unsigned long long uint64;
#endif

// A type to represent a Unicode code-point value. As of Unicode 4.0,
// such values require up to 21 bits.
// (For type-checking on pointers, make this explicitly signed,
// and it should always be the signed version of whatever int32 is.)
typedef signed int         char32;

const uint8  kuint8max  = (( uint8) 0xFF);
const uint16 kuint16max = ((uint16) 0xFFFF);
const uint32 kuint32max = ((uint32) 0xFFFFFFFF);
const uint64 kuint64max = ((uint64) GG_LONGLONG(0xFFFFFFFFFFFFFFFF));
const  int8  kint8min   = ((  int8) 0x80);
const  int8  kint8max   = ((  int8) 0x7F);
const  int16 kint16min  = (( int16) 0x8000);
const  int16 kint16max  = (( int16) 0x7FFF);
const  int32 kint32min  = (( int32) 0x80000000);
const  int32 kint32max  = (( int32) 0x7FFFFFFF);
const  int64 kint64min  = (( int64) GG_LONGLONG(0x8000000000000000));
const  int64 kint64max  = (( int64) GG_LONGLONG(0x7FFFFFFFFFFFFFFF));

// A macro to disallow the copy constructor and operator= functions
// This should be used in the private: declarations for a class
#define DISALLOW_COPY_AND_ASSIGN(TypeName) \
  TypeName(const TypeName&);               \
  void operator=(const TypeName&)

// An older, deprecated, politically incorrect name for the above.
// NOTE: The usage of this macro was baned from our code base, but some
// third_party libraries are yet using it.
// TODO(tfarina): Figure out how to fix the usage of this macro in the
// third_party libraries and get rid of it.
#define DISALLOW_EVIL_CONSTRUCTORS(TypeName) DISALLOW_COPY_AND_ASSIGN(TypeName)

// A macro to disallow all the implicit constructors, namely the
// default constructor, copy constructor and operator= functions.
//
// This should be used in the private: declarations for a class
// that wants to prevent anyone from instantiating it. This is
// especially useful for classes containing only static methods.
#define DISALLOW_IMPLICIT_CONSTRUCTORS(TypeName) \
  TypeName();                                    \
  DISALLOW_COPY_AND_ASSIGN(TypeName)

// The arraysize(arr) macro returns the # of elements in an array arr.
// The expression is a compile-time constant, and therefore can be
// used in defining new arrays, for example.  If you use arraysize on
// a pointer by mistake, you will get a compile-time error.
//
// One caveat is that arraysize() doesn't accept any array of an
// anonymous type or a type defined inside a function.  In these rare
// cases, you have to use the unsafe ARRAYSIZE_UNSAFE() macro below.  This is
// due to a limitation in C++'s template system.  The limitation might
// eventually be removed, but it hasn't happened yet.

// This template function declaration is used in defining arraysize.
// Note that the function doesn't need an implementation, as we only
// use its type.
template <typename T, size_t N>
char (&ArraySizeHelper(T (&array)[N]))[N];

// That gcc wants both of these prototypes seems mysterious. VC, for
// its part, can't decide which to use (another mystery). Matching of
// template overloads: the final frontier.
#ifndef _MSC_VER
template <typename T, size_t N>
char (&ArraySizeHelper(const T (&array)[N]))[N];
#endif

#define arraysize(array) (sizeof(ArraySizeHelper(array)))

// ARRAYSIZE_UNSAFE performs essentially the same calculation as arraysize,
// but can be used on anonymous types or types defined inside
// functions.  It's less safe than arraysize as it accepts some
// (although not all) pointers.  Therefore, you should use arraysize
// whenever possible.
//
// The expression ARRAYSIZE_UNSAFE(a) is a compile-time constant of type
// size_t.
//
// ARRAYSIZE_UNSAFE catches a few type errors.  If you see a compiler error
//
//   "warning: division by zero in ..."
//
// when using ARRAYSIZE_UNSAFE, you are (wrongfully) giving it a pointer.
// You should only use ARRAYSIZE_UNSAFE on statically allocated arrays.
//
// The following comments are on the implementation details, and can
// be ignored by the users.
//
// ARRAYSIZE_UNSAFE(arr) works by inspecting sizeof(arr) (the # of bytes in
// the array) and sizeof(*(arr)) (the # of bytes in one array
// element).  If the former is divisible by the latter, perhaps arr is
// indeed an array, in which case the division result is the # of
// elements in the array.  Otherwise, arr cannot possibly be an array,
// and we generate a compiler error to prevent the code from
// compiling.
//
// Since the size of bool is implementation-defined, we need to cast
// !(sizeof(a) & sizeof(*(a))) to size_t in order to ensure the final
// result has type size_t.
//
// This macro is not perfect as it wrongfully accepts certain
// pointers, namely where the pointer size is divisible by the pointee
// size.  Since all our code has to go through a 32-bit compiler,
// where a pointer is 4 bytes, this means all pointers to a type whose
// size is 3 or greater than 4 will be (righteously) rejected.

#define ARRAYSIZE_UNSAFE(a) \
  ((sizeof(a) / sizeof(*(a))) / \
   static_cast<size_t>(!(sizeof(a) % sizeof(*(a)))))


// Use implicit_cast as a safe version of static_cast or const_cast
// for upcasting in the type hierarchy (i.e. casting a pointer to Foo
// to a pointer to SuperclassOfFoo or casting a pointer to Foo to
// a const pointer to Foo).
// When you use implicit_cast, the compiler checks that the cast is safe.
// Such explicit implicit_casts are necessary in surprisingly many
// situations where C++ demands an exact type match instead of an
// argument type convertable to a target type.
//
// The From type can be inferred, so the preferred syntax for using
// implicit_cast is the same as for static_cast etc.:
//
//   implicit_cast<ToType>(expr)
//
// implicit_cast would have been part of the C++ standard library,
// but the proposal was submitted too late.  It will probably make
// its way into the language in the future.
template<typename To, typename From>
inline To implicit_cast(From const &f) {
  return f;
}

// The COMPILE_ASSERT macro can be used to verify that a compile time
// expression is true. For example, you could use it to verify the
// size of a static array:
//
//   COMPILE_ASSERT(ARRAYSIZE_UNSAFE(content_type_names) == CONTENT_NUM_TYPES,
//                  content_type_names_incorrect_size);
//
// or to make sure a struct is smaller than a certain size:
//
//   COMPILE_ASSERT(sizeof(foo) < 128, foo_too_large);
//
// The second argument to the macro is the name of the variable. If
// the expression is false, most compilers will issue a warning/error
// containing the name of the variable.

template <bool>
struct CompileAssert {
};

#undef COMPILE_ASSERT
#define COMPILE_ASSERT(expr, msg) \
  typedef CompileAssert<(bool(expr))> msg[bool(expr) ? 1 : -1]

// Implementation details of COMPILE_ASSERT:
//
// - COMPILE_ASSERT works by defining an array type that has -1
//   elements (and thus is invalid) when the expression is false.
//
// - The simpler definition
//
//     #define COMPILE_ASSERT(expr, msg) typedef char msg[(expr) ? 1 : -1]
//
//   does not work, as gcc supports variable-length arrays whose sizes
//   are determined at run-time (this is gcc's extension and not part
//   of the C++ standard).  As a result, gcc fails to reject the
//   following code with the simple definition:
//
//     int foo;
//     COMPILE_ASSERT(foo, msg); // not supposed to compile as foo is
//                               // not a compile-time constant.
//
// - By using the type CompileAssert<(bool(expr))>, we ensures that
//   expr is a compile-time constant.  (Template arguments must be
//   determined at compile-time.)
//
// - The outer parentheses in CompileAssert<(bool(expr))> are necessary
//   to work around a bug in gcc 3.4.4 and 4.0.1.  If we had written
//
//     CompileAssert<bool(expr)>
//
//   instead, these compilers will refuse to compile
//
//     COMPILE_ASSERT(5 > 0, some_message);
//
//   (They seem to think the ">" in "5 > 0" marks the end of the
//   template argument list.)
//
// - The array size is (bool(expr) ? 1 : -1), instead of simply
//
//     ((expr) ? 1 : -1).
//
//   This is to avoid running into a bug in MS VC 7.1, which
//   causes ((0.0) ? 1 : -1) to incorrectly evaluate to 1.


// MetatagId refers to metatag-id that we assign to
// each metatag <name, value> pair..
typedef uint32 MetatagId;

// Argument type used in interfaces that can optionally take ownership
// of a passed in argument.  If TAKE_OWNERSHIP is passed, the called
// object takes ownership of the argument.  Otherwise it does not.
enum Ownership {
  DO_NOT_TAKE_OWNERSHIP,
  TAKE_OWNERSHIP
};

// bit_cast<Dest,Source> is a template function that implements the
// equivalent of "*reinterpret_cast<Dest*>(&source)".  We need this in
// very low-level functions like the protobuf library and fast math
// support.
//
//   float f = 3.14159265358979;
//   int i = bit_cast<int32>(f);
//   // i = 0x40490fdb
//
// The classical address-casting method is:
//
//   // WRONG
//   float f = 3.14159265358979;            // WRONG
//   int i = * reinterpret_cast<int*>(&f);  // WRONG
//
// The address-casting method actually produces undefined behavior
// according to ISO C++ specification section 3.10 -15 -.  Roughly, this
// section says: if an object in memory has one type, and a program
// accesses it with a different type, then the result is undefined
// behavior for most values of "different type".
//
// This is true for any cast syntax, either *(int*)&f or
// *reinterpret_cast<int*>(&f).  And it is particularly true for
// conversions betweeen integral lvalues and floating-point lvalues.
//
// The purpose of 3.10 -15- is to allow optimizing compilers to assume
// that expressions with different types refer to different memory.  gcc
// 4.0.1 has an optimizer that takes advantage of this.  So a
// non-conforming program quietly produces wildly incorrect output.
//
// The problem is not the use of reinterpret_cast.  The problem is type
// punning: holding an object in memory of one type and reading its bits
// back using a different type.
//
// The C++ standard is more subtle and complex than this, but that
// is the basic idea.
//
// Anyways ...
//
// bit_cast<> calls memcpy() which is blessed by the standard,
// especially by the example in section 3.9 .  Also, of course,
// bit_cast<> wraps up the nasty logic in one place.
//
// Fortunately memcpy() is very fast.  In optimized mode, with a
// constant size, gcc 2.95.3, gcc 4.0.1, and msvc 7.1 produce inline
// code with the minimal amount of data movement.  On a 32-bit system,
// memcpy(d,s,4) compiles to one load and one store, and memcpy(d,s,8)
// compiles to two loads and two stores.
//
// I tested this code with gcc 2.95.3, gcc 4.0.1, icc 8.1, and msvc 7.1.
//
// WARNING: if Dest or Source is a non-POD type, the result of the memcpy
// is likely to surprise you.

template <class Dest, class Source>
inline Dest bit_cast(const Source& source) {
  // Compile time assertion: sizeof(Dest) == sizeof(Source)
  // A compile error here means your Dest and Source have different sizes.
  typedef char VerifySizesAreEqual [sizeof(Dest) == sizeof(Source) ? 1 : -1];

  Dest dest;
  memcpy(&dest, &source, sizeof(dest));
  return dest;
}

// Used to explicitly mark the return value of a function as unused. If you are
// really sure you don't want to do anything with the return value of a function
// that has been marked WARN_UNUSED_RESULT, wrap it with this. Example:
//
//   scoped_ptr<MyType> my_var = ...;
//   if (TakeOwnership(my_var.get()) == SUCCESS)
//     ignore_result(my_var.release());
//
template<typename T>
inline void ignore_result(const T& ignored) {
}

// The following enum should be used only as a constructor argument to indicate
// that the variable has static storage class, and that the constructor should
// do nothing to its state.  It indicates to the reader that it is legal to
// declare a static instance of the class, provided the constructor is given
// the base::LINKER_INITIALIZED argument.  Normally, it is unsafe to declare a
// static variable that has a constructor or a destructor because invocation
// order is undefined.  However, IF the type can be initialized by filling with
// zeroes (which the loader does for static variables), AND the destructor also
// does nothing to the storage, AND there are no virtual methods, then a
// constructor declared as
//       explicit MyClass(base::LinkerInitialized x) {}
// and invoked as
//       static MyClass my_variable_name(base::LINKER_INITIALIZED);
namespace base {
enum LinkerInitialized { LINKER_INITIALIZED };
}  // base

#endif  // BASE_BASICTYPES_H_

namespace base {

class BASE_API StringPiece {
 public:
  // standard STL container boilerplate
  typedef size_t size_type;
  typedef char value_type;
  typedef const char* pointer;
  typedef const char& reference;
  typedef const char& const_reference;
  typedef ptrdiff_t difference_type;
  typedef const char* const_iterator;
  typedef const char* iterator;
  typedef std::reverse_iterator<const_iterator> const_reverse_iterator;
  typedef std::reverse_iterator<iterator> reverse_iterator;

  static const size_type npos;

 public:
  // We provide non-explicit singleton constructors so users can pass
  // in a "const char*" or a "string" wherever a "StringPiece" is
  // expected.
  StringPiece() : ptr_(NULL), length_(0) { }
  StringPiece(const char* str)
    : ptr_(str), length_((str == NULL) ? 0 : strlen(str)) { }
  StringPiece(const std::string& str)
    : ptr_(str.data()), length_(str.size()) { }
  StringPiece(const char* offset, size_type len)
    : ptr_(offset), length_(len) { }

  // data() may return a pointer to a buffer with embedded NULs, and the
  // returned buffer may or may not be null terminated.  Therefore it is
  // typically a mistake to pass data() to a routine that expects a NUL
  // terminated string.
  const char* data() const { return ptr_; }
  size_type size() const { return length_; }
  size_type length() const { return length_; }
  bool empty() const { return length_ == 0; }

  void clear() {
    ptr_ = NULL;
    length_ = 0;
  }
  void set(const char* data, size_type len) {
    ptr_ = data;
    length_ = len;
  }
  void set(const char* str) {
    ptr_ = str;
    length_ = str ? strlen(str) : 0;
  }
  void set(const void* data, size_type len) {
    ptr_ = reinterpret_cast<const char*>(data);
    length_ = len;
  }

  char operator[](size_type i) const { return ptr_[i]; }

  void remove_prefix(size_type n) {
    ptr_ += n;
    length_ -= n;
  }

  void remove_suffix(size_type n) {
    length_ -= n;
  }

  int compare(const StringPiece& x) const {
    int r = wordmemcmp(
        ptr_, x.ptr_, (length_ < x.length_ ? length_ : x.length_));
    if (r == 0) {
      if (length_ < x.length_) r = -1;
      else if (length_ > x.length_) r = +1;
    }
    return r;
  }

  std::string as_string() const {
    // std::string doesn't like to take a NULL pointer even with a 0 size.
    return std::string(!empty() ? data() : "", size());
  }

  void CopyToString(std::string* target) const;
  void AppendToString(std::string* target) const;

  // Does "this" start with "x"
  bool starts_with(const StringPiece& x) const {
    return ((length_ >= x.length_) &&
            (wordmemcmp(ptr_, x.ptr_, x.length_) == 0));
  }

  // Does "this" end with "x"
  bool ends_with(const StringPiece& x) const {
    return ((length_ >= x.length_) &&
            (wordmemcmp(ptr_ + (length_-x.length_), x.ptr_, x.length_) == 0));
  }

  iterator begin() const { return ptr_; }
  iterator end() const { return ptr_ + length_; }
  const_reverse_iterator rbegin() const {
    return const_reverse_iterator(ptr_ + length_);
  }
  const_reverse_iterator rend() const {
    return const_reverse_iterator(ptr_);
  }

  size_type max_size() const { return length_; }
  size_type capacity() const { return length_; }

  size_type copy(char* buf, size_type n, size_type pos = 0) const;

  size_type find(const StringPiece& s, size_type pos = 0) const;
  size_type find(char c, size_type pos = 0) const;
  size_type rfind(const StringPiece& s, size_type pos = npos) const;
  size_type rfind(char c, size_type pos = npos) const;

  size_type find_first_of(const StringPiece& s, size_type pos = 0) const;
  size_type find_first_of(char c, size_type pos = 0) const {
    return find(c, pos);
  }
  size_type find_first_not_of(const StringPiece& s, size_type pos = 0) const;
  size_type find_first_not_of(char c, size_type pos = 0) const;
  size_type find_last_of(const StringPiece& s, size_type pos = npos) const;
  size_type find_last_of(char c, size_type pos = npos) const {
    return rfind(c, pos);
  }
  size_type find_last_not_of(const StringPiece& s, size_type pos = npos) const;
  size_type find_last_not_of(char c, size_type pos = npos) const;

  StringPiece substr(size_type pos, size_type n = npos) const;

  static int wordmemcmp(const char* p, const char* p2, size_type N) {
    return memcmp(p, p2, N);
  }

 private:
  const char*   ptr_;
  size_type     length_;
};

BASE_API bool operator==(const StringPiece& x, const StringPiece& y);

inline bool operator!=(const StringPiece& x, const StringPiece& y) {
  return !(x == y);
}

inline bool operator<(const StringPiece& x, const StringPiece& y) {
  const int r = StringPiece::wordmemcmp(
      x.data(), y.data(), (x.size() < y.size() ? x.size() : y.size()));
  return ((r < 0) || ((r == 0) && (x.size() < y.size())));
}

inline bool operator>(const StringPiece& x, const StringPiece& y) {
  return y < x;
}

inline bool operator<=(const StringPiece& x, const StringPiece& y) {
  return !(x > y);
}

inline bool operator>=(const StringPiece& x, const StringPiece& y) {
  return !(x < y);
}

}  // namespace base

#include <ostream>

inline std::ostream& operator << (std::ostream& os, const base::StringPiece& s)
{
    for(size_t i = 0; i < s.size(); ++i) os << s[i];
    return os;
}

#endif  // BASE_STRING_PIECE_H_