version 1.47, 2002/07/30 16:14:53
|
version 1.68, 2003/04/30 13:50:44
|
|
|
| |
| |
#include <cctype> | #include <cctype> |
|
#include <cstring> |
#include "String.h" | #include "String.h" |
#include "Array.h" | #include "Array.h" |
#include "Exception.h" |
#include "InternalException.h" |
#include <iostream> | #include <iostream> |
|
#include <fstream> |
|
#ifndef PEGASUS_REMOVE_DEPRECATED |
|
#include "System.h" // for strcasecmp |
|
#endif |
| |
PEGASUS_USING_STD; | PEGASUS_USING_STD; |
| |
|
|
| |
/////////////////////////////////////////////////////////////////////////////// | /////////////////////////////////////////////////////////////////////////////// |
// | // |
|
// CString |
|
// |
|
/////////////////////////////////////////////////////////////////////////////// |
|
|
|
CString::CString() |
|
: _rep(0) |
|
{ |
|
} |
|
|
|
CString::CString(const CString& cstr) |
|
{ |
|
_rep = (void*)new char[strlen((char*)cstr._rep)+1]; |
|
strcpy((char*)_rep, (char*)cstr._rep); |
|
} |
|
|
|
CString::CString(char* cstr) |
|
: _rep(cstr) |
|
{ |
|
} |
|
|
|
CString::~CString() |
|
{ |
|
if (_rep) |
|
delete [] (char*)_rep; |
|
} |
|
|
|
CString& CString::operator=(const CString& cstr) |
|
{ |
|
_rep = (char*)new char[strlen((char*)cstr._rep)+1]; |
|
strcpy((char*)_rep, (char*)cstr._rep); |
|
return *this; |
|
} |
|
|
|
CString::operator const char*() const |
|
{ |
|
return (char*)_rep; |
|
} |
|
|
|
/////////////////////////////////////////////////////////////////////////////// |
|
// |
// String | // String |
// | // |
/////////////////////////////////////////////////////////////////////////////// | /////////////////////////////////////////////////////////////////////////////// |
|
|
| |
String::String(const String& str) | String::String(const String& str) |
{ | { |
|
if (str._rep != NULL) |
|
{ |
_rep = new StringRep(*str._rep); | _rep = new StringRep(*str._rep); |
} | } |
|
else |
|
{ |
|
_rep = new StringRep(); |
|
} |
|
} |
|
|
| |
String::String(const String& str, Uint32 n) | String::String(const String& str, Uint32 n) |
{ | { |
_rep = new StringRep; | _rep = new StringRep; |
assign(str.getData(), n); |
assign(str.getChar16Data(), n); |
} | } |
| |
String::String(const Char16* str) | String::String(const Char16* str) |
|
|
_rep->c16a.reserveCapacity(n); | _rep->c16a.reserveCapacity(n); |
| |
while (n--) | while (n--) |
_rep->c16a.append(*str++); |
_rep->c16a.append(Uint8(*str++)); |
| |
return *this; | return *this; |
} | } |
|
|
_rep->c16a.reserveCapacity(_n + 1); | _rep->c16a.reserveCapacity(_n + 1); |
| |
while (_n--) | while (_n--) |
_rep->c16a.append(*str++); |
_rep->c16a.append(Uint8(*str++)); |
| |
_rep->c16a.append('\0'); | _rep->c16a.append('\0'); |
| |
|
|
return _rep->c16a.size() - 1; | return _rep->c16a.size() - 1; |
} | } |
| |
const Char16* String::getData() const |
const Char16* String::getChar16Data() const |
{ | { |
return _rep->c16a.getData(); | return _rep->c16a.getData(); |
} | } |
| |
char* String::allocateCString(Uint32 extraBytes, Boolean noThrow) const |
CString String::getCString() const |
{ | { |
Uint32 n = size() + 1; | Uint32 n = size() + 1; |
char* str = new char[n + extraBytes]; |
char* str = new char[n]; |
char* p = str; | char* p = str; |
const Char16* q = getData(); |
const Char16* q = getChar16Data(); |
| |
for (Uint32 i = 0; i < n; i++) | for (Uint32 i = 0; i < n; i++) |
{ | { |
Uint16 c = *q++; | Uint16 c = *q++; |
*p++ = char(c); | *p++ = char(c); |
| |
if ((c & 0xff00) && !noThrow) |
//if (c & 0xff00) |
throw TruncatedCharacter(); |
// truncatedCharacters = true; |
} | } |
| |
return str; |
return CString(str); |
} | } |
| |
Char16& String::operator[](Uint32 i) |
Char16& String::operator[](Uint32 index) |
{ | { |
if (i > size()) |
if (index > size()) |
throw OutOfBounds(); |
throw IndexOutOfBoundsException(); |
| |
return _rep->c16a[i]; |
return _rep->c16a[index]; |
} | } |
| |
const Char16 String::operator[](Uint32 i) const |
const Char16 String::operator[](Uint32 index) const |
{ | { |
if (i > size()) |
if (index > size()) |
throw OutOfBounds(); |
throw IndexOutOfBoundsException(); |
| |
return _rep->c16a[i]; |
return _rep->c16a[index]; |
} | } |
| |
String& String::append(const Char16& c) | String& String::append(const Char16& c) |
|
|
| |
String& String::append(const String& str) | String& String::append(const String& str) |
{ | { |
return append(str.getData(), str.size()); |
return append(str.getChar16Data(), str.size()); |
} |
|
|
|
String& String::operator+=(const String& str) |
|
{ |
|
return append(str); |
|
} |
|
|
|
String& String::operator+=(Char16 c) |
|
{ |
|
return append(c); |
|
} |
|
|
|
String& String::operator+=(char c) |
|
{ |
|
return append(Char16(c)); |
|
} | } |
| |
void String::remove(Uint32 pos, Uint32 size) |
void String::remove(Uint32 index, Uint32 size) |
{ | { |
if (size == PEG_NOT_FOUND) | if (size == PEG_NOT_FOUND) |
size = this->size() - pos; |
size = this->size() - index; |
| |
if (pos + size > this->size()) |
if (index + size > this->size()) |
throw OutOfBounds(); |
throw IndexOutOfBoundsException(); |
| |
if (size) | if (size) |
_rep->c16a.remove(pos, size); |
_rep->c16a.remove(index, size); |
} | } |
| |
String String::subString(Uint32 pos, Uint32 length) const |
String String::subString(Uint32 index, Uint32 length) const |
{ | { |
if (pos < size()) |
if (index < size()) |
{ | { |
if (length == PEG_NOT_FOUND) |
if ((length == PEG_NOT_FOUND) || (length > size() - index)) |
length = size() - pos; |
length = size() - index; |
| |
return String(getData() + pos, length); |
return String(getChar16Data() + index, length); |
} | } |
else | else |
return String(); | return String(); |
|
|
| |
Uint32 String::find(Char16 c) const | Uint32 String::find(Char16 c) const |
{ | { |
const Char16* first = getData(); |
const Char16* first = getChar16Data(); |
| |
for (const Char16* p = first; *p; p++) | for (const Char16* p = first; *p; p++) |
{ | { |
|
|
return PEG_NOT_FOUND; | return PEG_NOT_FOUND; |
} | } |
| |
Uint32 String::find(Uint32 pos, Char16 c) const |
Uint32 String::find(Uint32 index, Char16 c) const |
{ | { |
const Char16* data = getData(); |
const Char16* data = getChar16Data(); |
| |
for (Uint32 i = pos, n = size(); i < n; i++) |
for (Uint32 i = index, n = size(); i < n; i++) |
{ | { |
if (data[i] == c) | if (data[i] == c) |
return i; | return i; |
|
|
| |
Uint32 String::find(const String& s) const | Uint32 String::find(const String& s) const |
{ | { |
const Char16* pSubStr = s.getData(); |
const Char16* pSubStr = s.getChar16Data(); |
const Char16* pStr = getData(); |
const Char16* pStr = getChar16Data(); |
Uint32 subStrLen = s.size(); | Uint32 subStrLen = s.size(); |
Uint32 strLen = size(); | Uint32 strLen = size(); |
| |
|
|
| |
Uint32 String::reverseFind(Char16 c) const | Uint32 String::reverseFind(Char16 c) const |
{ | { |
const Char16* first = getData(); |
const Char16* first = getChar16Data(); |
const Char16* last = getData() + size(); |
const Char16* last = getChar16Data() + size(); |
| |
while (last != first) | while (last != first) |
{ | { |
|
|
return PEG_NOT_FOUND; | return PEG_NOT_FOUND; |
} | } |
| |
|
// ATTN-RK-P3-20020509: Define case-sensitivity for non-English characters |
void String::toLower() | void String::toLower() |
{ | { |
for (Char16* p = &_rep->c16a[0]; *p; p++) | for (Char16* p = &_rep->c16a[0]; *p; p++) |
|
|
| |
int String::compare(const String& s1, const String& s2, Uint32 n) | int String::compare(const String& s1, const String& s2, Uint32 n) |
{ | { |
const Char16* s1c16 = s1.getData(); |
const Char16* s1c16 = s1.getChar16Data(); |
const Char16* s2c16 = s2.getData(); |
const Char16* s2c16 = s2.getChar16Data(); |
| |
while (n--) | while (n--) |
{ | { |
|
|
| |
int String::compare(const String& s1, const String& s2) | int String::compare(const String& s1, const String& s2) |
{ | { |
const Char16* s1c16 = s1.getData(); |
const Char16* s1c16 = s1.getChar16Data(); |
const Char16* s2c16 = s2.getData(); |
const Char16* s2c16 = s2.getChar16Data(); |
| |
while (*s1c16 && *s2c16) | while (*s1c16 && *s2c16) |
{ | { |
|
|
return 0; | return 0; |
} | } |
| |
int String::compareNoCase(const char* s1, const char* s2, Uint32 n) |
|
{ |
|
while (n--) |
|
{ |
|
int r = tolower(*s1++) - tolower(*s2++); |
|
|
|
if (r) |
|
return r; |
|
} |
|
|
|
return 0; |
|
} |
|
|
|
int String::compareNoCase(const char* s1, const char* s2) |
|
{ |
|
while (*s1 && *s2) |
|
{ |
|
int r = tolower(*s1++) - tolower(*s2++); |
|
|
|
if (r) |
|
return r; |
|
} |
|
|
|
if (*s2) |
|
return -1; |
|
else if (*s1) |
|
return 1; |
|
|
|
return 0; |
|
} |
|
|
|
int String::compareNoCase(const String& s1, const String& s2) | int String::compareNoCase(const String& s1, const String& s2) |
{ | { |
const Char16* _s1 = s1.getData(); |
const Char16* _s1 = s1.getChar16Data(); |
const Char16* _s2 = s2.getData(); |
const Char16* _s2 = s2.getChar16Data(); |
| |
while (*_s1 && *_s2) | while (*_s1 && *_s2) |
{ | { |
|
|
if (str1.size() != str2.size()) | if (str1.size() != str2.size()) |
return false; | return false; |
| |
const Char16* p = str1.getData(); |
const Char16* p = str1.getChar16Data(); |
const Char16* q = str2.getData(); |
const Char16* q = str2.getChar16Data(); |
| |
Uint32 n = str1.size(); | Uint32 n = str1.size(); |
| |
|
|
} | } |
| |
| |
|
#if 0 |
// ATTN-RK-P3-20020603: This code is not completely correct | // ATTN-RK-P3-20020603: This code is not completely correct |
// Wildcard String matching function that may be useful in the future | // Wildcard String matching function that may be useful in the future |
// The following code was provided by Bob Blair. | // The following code was provided by Bob Blair. |
|
|
} | } |
| |
| |
|
/** match matches a string against a GLOB style pattern. |
|
Return trues if the String parameter matches the pattern. C-Shell style |
|
glob matching is used. |
|
@param str String to be matched against the pattern |
|
@param pattern Pattern to use in the match |
|
@return Boolean true if str matches pattern |
|
The pattern definition is as follows: |
|
<pre> |
|
* Matches any number of any characters |
|
? Match exactly one character |
|
[chars] Match any character in chars |
|
[chara-charb] Match any character in the range between chara and charb |
|
</pre> |
|
The literal characters *, ?, [, ] can be included in a string by |
|
escaping them with backslash "\". Ranges of characters can be concatenated. |
|
<pre> |
|
examples: |
|
Boolean result = String::match("This is a test", "*is*"); |
|
Boolean works = String::match("abcdef123", "*[0-9]"); |
|
</pre> |
|
*/ |
Boolean String::match(const String& str, const String& pattern) | Boolean String::match(const String& str, const String& pattern) |
{ | { |
return _StringMatch( | return _StringMatch( |
(Uint16*)str.getData(), (Uint16*)pattern.getData(), 0) != 0; |
(Uint16*)str.getChar16Data(), (Uint16*)pattern.getChar16Data(), 0) != 0; |
} | } |
| |
|
/** matchNoCase Matches a String against a GLOB style pattern independent |
|
of case. |
|
Returns true if the str parameter matches the pattern. C-Shell style |
|
glob matching is used. Ignore case in all comparisons. Case is |
|
ignored in the match. |
|
@parm str String containing the string to be matched\ |
|
@parm pattern GLOB style patterh to use in the match. |
|
@return Boolean true if str matches patterh |
|
@SeeAlso match |
|
*/ |
Boolean String::matchNoCase(const String& str, const String& pattern) | Boolean String::matchNoCase(const String& str, const String& pattern) |
{ | { |
return _StringMatch( | return _StringMatch( |
(Uint16*)str.getData(), (Uint16*)pattern.getData(), 1) != 0; |
(Uint16*)str.getChar16Data(), (Uint16*)pattern.getChar16Data(), 1) != 0; |
} | } |
|
#endif |
| |
| |
/////////////////////////////////////////////////////////////////////////////// | /////////////////////////////////////////////////////////////////////////////// |
|
|
| |
PEGASUS_STD(ostream)& operator<<(PEGASUS_STD(ostream)& os, const String& str) | PEGASUS_STD(ostream)& operator<<(PEGASUS_STD(ostream)& os, const String& str) |
{ | { |
|
#ifdef PEGASUS_OS_OS400 |
|
int inc = 0; |
|
int newbuf = 0; |
|
char *buffer = NULL; |
|
char buffer1[201]; |
|
char temp[2]; |
|
if (str.size() > 200) |
|
{ |
|
buffer = new char[str.size()+1]; |
|
newbuf = 1; |
|
} |
|
else |
|
buffer = buffer1; |
|
#endif |
|
|
for (Uint32 i = 0, n = str.size(); i < n; i++) | for (Uint32 i = 0, n = str.size(); i < n; i++) |
os << str[i]; |
{ |
|
Uint16 code = str[i]; |
|
|
|
if (code > 0 && code <= PEGASUS_MAX_PRINTABLE_CHAR) |
|
{ |
|
#ifdef PEGASUS_OS_OS400 |
|
// process so messages don't get displayed as one char per line on OS/400. |
|
// Uint16 is a 2 byte character where byte 1 is '00' and byte 2 is |
|
// the character. Also, the entire string needs to be sent to os instead |
|
// of one "byte/Unit16" at a time. Sending one "byte/Uint16" at a time also |
|
// causes one character per line. On OS/400 use of os << char(code) is a |
|
// restriction and no available c/cpp alternative was available. The |
|
// following was created to compensate for this restriction. |
|
memcpy(temp, &code, 2); |
|
memcpy(buffer+inc, &temp[1], 1); // do not include the '00' |
|
if ((i+1) == n) // last character |
|
{ |
|
memset(buffer+n, 0x00, 1); // add null terminator |
|
os << buffer; // return 1-byte per character string |
|
if (buffer && newbuf != 0) |
|
delete [] buffer; // okay; this is the end of the loop |
|
} |
|
inc++; |
|
#else |
|
os << char(code); |
|
#endif |
|
} |
|
else |
|
{ |
|
// Print in hex format: |
|
char buffer[8]; |
|
sprintf(buffer, "\\x%04X", code); |
|
os << buffer; |
|
} |
|
} |
| |
return os; | return os; |
} | } |
|
|
return String::compare(str1, str2) >= 0; | return String::compare(str1, str2) >= 0; |
} | } |
| |
|
#ifndef PEGASUS_REMOVE_DEPRECATED |
int CompareNoCase(const char* s1, const char* s2) | int CompareNoCase(const char* s1, const char* s2) |
{ | { |
while (*s1 && *s2) |
return System::strcasecmp(s1, s2); |
{ |
|
int r = tolower(*s1++) - tolower(*s2++); |
|
|
|
if (r) |
|
return r; |
|
} |
|
|
|
if (*s2) |
|
return -1; |
|
else if (*s1) |
|
return 1; |
|
|
|
return 0; |
|
} |
|
|
|
int EqualNoCase(const char* s1, const char* s2) |
|
{ |
|
return CompareNoCase(s1, s2) == 0; |
|
} | } |
|
#endif |
| |
PEGASUS_NAMESPACE_END | PEGASUS_NAMESPACE_END |