1
0
forked from 0ad/0ad
0ad/source/ps/CStr.h
Ykkrosh a265a441fd # Fixed string handling for Windows/Linux compatibility.
* vsnprintf2: Made compatible between GCC and MSVC - it now always
null-terminates the buffer, and returns -1 on overflow. Fixes #158.
Added tests.
 * MeshManager: Use shared_ptr.expired() instead of checking for
bad_weak_ptr exception.
 * Xeromyces: Added tests for GetXMBPath, because it does unusual things
in sscanf which MSVC's /analyze complains about.
 * ConfigDB, ScriptGlue: Replaced some asserts with return-on-failure,
to avoid invalid array accesses when continuing after the assert (as
complained about by /analyze).
 * CStr: Removed "using namespace std". Added tests for handling of
invalid UTF-8.

This was SVN commit r4625.
2006-11-07 21:03:13 +00:00

254 lines
7.1 KiB
C++

/*
CStr.h
by Caecus
Caecus@0ad.wildfiregames.com
Overview:
Contains CStr class which is a versatile class for making string use easy.
Basic functionality implemented via STL string, so performance is limited
based on the STL implementation we use.
Example:
CStr MyString = _T("Hello, World.");
int MyNum = 126;
MyString += _T(" I'm the number ") += CStr(MyNumber);
// Prints "Hello, World. I'm the number 126"
_tcout << (LPCTSTR)MyString << endl;
MyString = _("2341");
MyNum = MyString.ToInt();
// Prints "2341"
_tcout << MyNum << endl;
More Info:
http://wildfiregames.com/0ad/codepit/TDD/CStr.html
[[ This documentation is out of date; CStr is always 8 bits, and CStrW is always
sizeof(wchar_t) (16 or 32). Don't use _T; CStrW constants are just L"string". ]]
*/
// history:
// 19 May 04, Mark Thompson (mot20@cam.ac.uk / mark@wildfiregames.com)
// 2004-06-18 janwas: replaced conversion buffer with stringstream
// 2004-10-31 Philip: Changed to inherit from std::[w]string
#ifndef CSTR_H_FIRST
#define CSTR_H_FIRST
enum PS_TRIM_MODE { PS_TRIM_LEFT, PS_TRIM_RIGHT, PS_TRIM_BOTH };
#ifndef IN_UNIDOUBLER
#define UNIDOUBLER_HEADER "CStr.h"
#include "UniDoubler.h"
#endif
#endif
// Include this section when in unidoubler mode, and when this unicode/ascii
// version has not already been included.
#if defined(IN_UNIDOUBLER) && ( (defined(_UNICODE) && !defined(CSTR_H_U)) || (!defined(_UNICODE) && !defined(CSTR_H_A)) )
#ifdef _UNICODE
#define CSTR_H_U
#else
#define CSTR_H_A
#endif
#include <string>
#include "ps/utf16string.h"
class CStr8;
class CStrW;
// CStr class, the mother of all strings
class CStr: public std::tstring
{
// The two variations must be friends with each other
#ifdef _UNICODE
friend class CStr8;
#else
friend class CStrW;
#endif
public:
// CONSTRUCTORS
CStr() {}
CStr(const CStr& String) : std::tstring(String) {}
CStr(const tchar* String) : std::tstring(String) {}
CStr(const tchar* String, size_t Length)
: std::tstring(String, Length) {}
CStr(const tchar Char) : std::tstring(1, Char) {} // std::string's constructor is (repeats, chr)
CStr(std::tstring String) : std::tstring(String) {}
// Named constructor, to avoid overload overload.
static CStr Repeat(const CStr& String, size_t Reps);
// CStr(8|W) construction from utf16strings.
// allowed on MSVC as of 2006-02-03 because utf16string is
// now distinct from CStrW.
CStr(utf16string String) : std::tstring(String.begin(), String.end()) {}
// Transparent CStrW/8 conversion. Non-ASCII characters are not
// handled correctly.
#ifndef _UNICODE
CStr8(const CStrW& wideStr);
#else
CStrW(const CStr8 &asciiStr);
#endif
// Conversion to/from UTF-8, encoded in a CStr8.
// Common non-ASCII characters are handled correctly.
// Characters outside the BMP (above 0xFFFF) are *not* handled correctly.
// FromUTF8 may fail, if converting from invalid UTF-8 data - the empty
// string will be returned.
#ifdef _UNICODE
CStr8 ToUTF8() const;
#else
CStrW FromUTF8() const;
#endif
CStr(int Number); // Creates CStr from a int
CStr(unsigned int Number); // Creates CStr from a uint
CStr(long Number); // Creates CStr from a long
CStr(unsigned long Number); // Creates CStr from a ulong
CStr(float Number); // Creates CStr from a float
CStr(double Number); // Creates CStr from a double
~CStr() {}; // Destructor
// Conversions
int ToInt() const;
unsigned int ToUInt() const;
long ToLong() const;
unsigned long ToULong() const;
float ToFloat() const;
double ToDouble() const;
// Returns the length of the string in characters
size_t Length() const { return length(); }
// Retrieves the substring within the string
CStr GetSubstring(size_t start, size_t len) const;
// Search the string for another string. Returns the offset of the first
// match, or -1 if no matches are found.
long Find(const CStr& Str) const;
long Find(const tchar &chr) const;
long Find(const int &start, const tchar &chr) const;
// Case-insensitive versions of Find
long FindInsensitive(const CStr& Str) const;
long FindInsensitive(const tchar &chr) const;
long FindInsensitive(const int &start, const tchar &chr) const;
// You can also do a "ReverseFind" - i.e. search starting from the end
long ReverseFind(const CStr& Str) const;
// Lowercase and uppercase
CStr LowerCase() const;
CStr UpperCase() const;
CStr LCase() const;
CStr UCase() const;
// Retrieve the substring of the first n characters
CStr Left(size_t len) const;
// Retrieve the substring of the last n characters
CStr Right(size_t len) const;
// Retrieve the substring following the last occurrence of Str
// (or the whole string if it doesn't contain Str)
CStr AfterLast(const CStr& Str) const;
// Retrieve the substring preceding the last occurrence of Str
// (or the whole string if it doesn't contain Str)
CStr BeforeLast(const CStr& Str) const;
// Retrieve the substring following the first occurrence of Str
// (or the whole string if it doesn't contain Str)
CStr AfterFirst(const CStr& Str) const;
// Retrieve the substring preceding the first occurrence of Str
// (or the whole string if it doesn't contain Str)
CStr BeforeFirst(const CStr& Str) const;
// Remove all occurrences of some character or substring
void Remove(const CStr& Str);
// Replace all occurrences of some substring by another
void Replace(const CStr& StrToReplace, const CStr& ReplaceWith);
// Convert strings like \\\n into <backslash><newline>
CStr UnescapeBackslashes();
// Returns a trimmed string, removes whitespace from the left/right/both
CStr Trim(PS_TRIM_MODE Mode) const;
// Returns a padded string, adding spaces to the left/right/both
CStr Pad(PS_TRIM_MODE Mode, size_t Length) const;
// Overloaded operations
CStr& operator=(int Number);
CStr& operator=(long Number);
CStr& operator=(unsigned int Number);
CStr& operator=(unsigned long Number);
CStr& operator=(float Number);
CStr& operator=(double Number);
CStr operator+(const CStr& Str);
CStr operator+(const tchar* Str);
#ifndef _UNICODE
CStr8 operator+(const CStrW& Str);
#else
CStrW operator+(const CStr8& Str);
#endif
operator const tchar*() const;
// Do some range checking in debug builds
tchar& operator[](size_t n) { debug_assert(n < length()); return this->std::tstring::operator[](n); }
tchar& operator[](int n) { debug_assert((size_t)n < length()); return this->std::tstring::operator[](n); }
// Conversion to utf16string
inline utf16string utf16() const
{ return utf16string(begin(), end()); }
// Calculates a hash of the string's contents
size_t GetHashCode() const;
// Serialization functions
// (These are not virtual or inherited from ISerializable, to avoid
// adding a vtable and making the strings larger than std::string)
uint GetSerializedLength() const;
u8 *Serialize(u8 *buffer) const;
const u8 *Deserialize(const u8 *buffer, const u8 *bufferend);
};
// Hash function (for STL_HASH_MAP, etc)
class CStr_hash_compare
{
public:
static const size_t bucket_size = 1;
static const size_t min_buckets = 16;
size_t operator() (const CStr& Key) const
{
return Key.GetHashCode();
}
bool operator() (const CStr& _Key1, const CStr& _Key2) const
{
return (_Key1 < _Key2);
}
};
#endif