King_DuckZ/utfcpp
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Fix for the bug [name clash with std::next - ID: 3215839]

Browse source 
git-svn-id: http://svn.code.sf.net/p/utfcpp/code@122 a809a056-fc17-0410-9590-b4f493f8b08e

Conflicts:
	v2_0/test_drivers/performance/iconvtest.cpp
This commit is contained in:
ntrifunovic 2011-06-24 23:21:41 +00:00 committed by King_DuckZ
parent 7ed2b4457b
commit a5c5cc8019
4 changed files with 122 additions and 121 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

132
source/utf8/checked.h
View file

@ -34,7 +34,7 @@ DEALINGS IN THE SOFTWARE.
namespace utf8
{
// Base for the exceptions that may be thrown from the library
class exception : public std::exception {
class exception : public ::std::exception {
};
// Exceptions that may be thrown from the library functions.
@ -69,48 +69,10 @@ namespace utf8
/// The library API - functions intended to be called by the users
template <typename octet_iterator, typename output_iterator>
output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_t replacement)
{
while (start != end) {
octet_iterator sequence_start = start;
internal::utf_error err_code = internal::validate_next(start, end);
switch (err_code) {
case internal::UTF8_OK :
for (octet_iterator it = sequence_start; it != start; ++it)
*out++ = *it;
break;
case internal::NOT_ENOUGH_ROOM:
throw not_enough_room();
case internal::INVALID_LEAD:
append (replacement, out);
++start;
break;
case internal::INCOMPLETE_SEQUENCE:
case internal::OVERLONG_SEQUENCE:
case internal::INVALID_CODE_POINT:
append (replacement, out);
++start;
// just one replacement mark for the sequence
while (internal::is_trail(*start) && start != end)
++start;
break;
}
}
return out;
}
template <typename octet_iterator, typename output_iterator>
inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out)
{
static const uint32_t replacement_marker = internal::mask16(0xfffd);
return replace_invalid(start, end, out, replacement_marker);
}
template <typename octet_iterator>
octet_iterator append(uint32_t cp, octet_iterator result)
{
if (!internal::is_code_point_valid(cp))
if (!utf8::internal::is_code_point_valid(cp))
throw invalid_code_point(cp);
if (cp < 0x80) // one octet
@ -124,7 +86,7 @@ namespace utf8
*(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f) | 0x80);
*(result++) = static_cast<uint8_t>((cp & 0x3f) | 0x80);
}
else { // four octets
else { // four octets
*(result++) = static_cast<uint8_t>((cp >> 18) | 0xf0);
*(result++) = static_cast<uint8_t>(((cp >> 12) & 0x3f) | 0x80);
*(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f) | 0x80);
@ -133,11 +95,49 @@ namespace utf8
return result;
}
template <typename octet_iterator, typename output_iterator>
output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_t replacement)
{
while (start != end) {
octet_iterator sequence_start = start;
internal::utf_error err_code = utf8::internal::validate_next(start, end);
switch (err_code) {
case internal::UTF8_OK :
for (octet_iterator it = sequence_start; it != start; ++it)
*out++ = *it;
break;
case internal::NOT_ENOUGH_ROOM:
throw not_enough_room();
case internal::INVALID_LEAD:
utf8::append (replacement, out);
++start;
break;
case internal::INCOMPLETE_SEQUENCE:
case internal::OVERLONG_SEQUENCE:
case internal::INVALID_CODE_POINT:
utf8::append (replacement, out);
++start;
// just one replacement mark for the sequence
while (utf8::internal::is_trail(*start) && start != end)
++start;
break;
}
}
return out;
}
template <typename octet_iterator, typename output_iterator>
inline output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out)
{
static const uint32_t replacement_marker = utf8::internal::mask16(0xfffd);
return utf8::replace_invalid(start, end, out, replacement_marker);
}
template <typename octet_iterator>
uint32_t next(octet_iterator& it, octet_iterator end)
{
uint32_t cp = 0;
internal::utf_error err_code = internal::validate_next(it, end, &cp);
internal::utf_error err_code = utf8::internal::validate_next(it, end, &cp);
switch (err_code) {
case internal::UTF8_OK :
break;
@ -156,22 +156,22 @@ namespace utf8
template <typename octet_iterator>
uint32_t peek_next(octet_iterator it, octet_iterator end)
{
return next(it, end);
return utf8::next(it, end);
}
template <typename octet_iterator>
uint32_t prior(octet_iterator& it, octet_iterator start)
{
{
// can't do much if it == start
if (it == start)
if (it == start)
throw not_enough_room();
octet_iterator end = it;
octet_iterator end = it;
// Go back until we hit either a lead octet or start
while (internal::is_trail(*(--it)))
while (utf8::internal::is_trail(*(--it)))
if (it == start)
throw invalid_utf8(*it); // error - no lead byte in the sequence
return peek_next(it, end);
return utf8::peek_next(it, end);
}
/// Deprecated in versions that include "prior"
@ -179,18 +179,18 @@ namespace utf8
uint32_t previous(octet_iterator& it, octet_iterator pass_start)
{
octet_iterator end = it;
while (internal::is_trail(*(--it)))
while (utf8::internal::is_trail(*(--it)))
if (it == pass_start)
throw invalid_utf8(*it); // error - no lead byte in the sequence
octet_iterator temp = it;
return next(temp, end);
return utf8::next(temp, end);
}
template <typename octet_iterator, typename distance_type>
void advance (octet_iterator& it, distance_type n, octet_iterator end)
{
for (distance_type i = 0; i < n; ++i)
next(it, end);
utf8::next(it, end);
}
template <typename octet_iterator>
@ -199,7 +199,7 @@ namespace utf8
{
typename std::iterator_traits<octet_iterator>::difference_type dist;
for (dist = 0; first < last; ++dist)
next(first, last);
utf8::next(first, last);
return dist;
}
@ -207,12 +207,12 @@ namespace utf8
octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result)
{
while (start != end) {
uint32_t cp = internal::mask16(*start++);
uint32_t cp = utf8::internal::mask16(*start++);
// Take care of surrogate pairs first
if (internal::is_lead_surrogate(cp)) {
if (utf8::internal::is_lead_surrogate(cp)) {
if (start != end) {
uint32_t trail_surrogate = internal::mask16(*start++);
if (internal::is_trail_surrogate(trail_surrogate))
uint32_t trail_surrogate = utf8::internal::mask16(*start++);
if (utf8::internal::is_trail_surrogate(trail_surrogate))
cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET;
else
throw invalid_utf16(static_cast<uint16_t>(trail_surrogate));
@ -222,10 +222,10 @@ namespace utf8
}
// Lone trail surrogate
else if (internal::is_trail_surrogate(cp))
else if (utf8::internal::is_trail_surrogate(cp))
throw invalid_utf16(static_cast<uint16_t>(cp));
result = append(cp, result);
result = utf8::append(cp, result);
}
return result;
}
@ -234,7 +234,7 @@ namespace utf8
u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result)
{
while (start != end) {
uint32_t cp = next(start, end);
uint32_t cp = utf8::next(start, end);
if (cp > 0xffff) { //make a surrogate pair
*result++ = static_cast<uint16_t>((cp >> 10) + internal::LEAD_OFFSET);
*result++ = static_cast<uint16_t>((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN);
@ -249,7 +249,7 @@ namespace utf8
octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result)
{
while (start != end)
result = append(*(start++), result);
result = utf8::append(*(start++), result);
return result;
}
@ -258,7 +258,7 @@ namespace utf8
u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result)
{
while (start != end)
(*result++) = next(start, end);
(*result++) = utf8::next(start, end);
return result;
}
@ -284,7 +284,7 @@ namespace utf8
uint32_t operator * () const
{
octet_iterator temp = it;
return next(temp, range_end);
return utf8::next(temp, range_end);
}
bool operator == (const iterator& rhs) const
{
@ -298,24 +298,24 @@ namespace utf8
}
iterator& operator ++ ()
{
next(it, range_end);
utf8::next(it, range_end);
return *this;
}
iterator operator ++ (int)
{
iterator temp = *this;
next(it, range_end);
utf8::next(it, range_end);
return temp;
}
iterator& operator -- ()
{
prior(it, range_start);
utf8::prior(it, range_start);
return *this;
}
iterator operator -- (int)
{
iterator temp = *this;
prior(it, range_start);
utf8::prior(it, range_start);
return temp;
}
}; // class iterator

66
source/utf8/core.h
View file

@ -68,7 +68,7 @@ namespace internal
template<typename octet_type>
inline bool is_trail(octet_type oc)
{
return ((mask8(oc) >> 6) == 0x2);
return ((utf8::internal::mask8(oc) >> 6) == 0x2);
}
template <typename u16>
@ -92,14 +92,14 @@ namespace internal
template <typename u32>
inline bool is_code_point_valid(u32 cp)
{
return (cp <= CODE_POINT_MAX && !is_surrogate(cp));
return (cp <= CODE_POINT_MAX && !utf8::internal::is_surrogate(cp));
}
template <typename octet_iterator>
inline typename std::iterator_traits<octet_iterator>::difference_type
sequence_length(octet_iterator lead_it)
{
uint8_t lead = mask8(*lead_it);
uint8_t lead = utf8::internal::mask8(*lead_it);
if (lead < 0x80)
return 1;
else if ((lead >> 5) == 0x6)
@ -140,7 +140,7 @@ namespace internal
{
if (it != end) {
if (code_point)
*code_point = mask8(*it);
*code_point = utf8::internal::mask8(*it);
return UTF8_OK;
}
return NOT_ENOUGH_ROOM;
@ -152,9 +152,9 @@ namespace internal
utf_error ret_code = NOT_ENOUGH_ROOM;
if (it != end) {
uint32_t cp = mask8(*it);
uint32_t cp = utf8::internal::mask8(*it);
if (++it != end) {
if (is_trail(*it)) {
if (utf8::internal::is_trail(*it)) {
cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f);
if (code_point)
@ -177,12 +177,12 @@ namespace internal
utf_error ret_code = NOT_ENOUGH_ROOM;
if (it != end) {
uint32_t cp = mask8(*it);
uint32_t cp = utf8::internal::mask8(*it);
if (++it != end) {
if (is_trail(*it)) {
cp = ((cp << 12) & 0xffff) + ((mask8(*it) << 6) & 0xfff);
if (utf8::internal::is_trail(*it)) {
cp = ((cp << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff);
if (++it != end) {
if (is_trail(*it)) {
if (utf8::internal::is_trail(*it)) {
cp += (*it) & 0x3f;
if (code_point)
@ -211,15 +211,15 @@ namespace internal
utf_error ret_code = NOT_ENOUGH_ROOM;
if (it != end) {
uint32_t cp = mask8(*it);
uint32_t cp = utf8::internal::mask8(*it);
if (++it != end) {
if (is_trail(*it)) {
cp = ((cp << 18) & 0x1fffff) + ((mask8(*it) << 12) & 0x3ffff);
if (utf8::internal::is_trail(*it)) {
cp = ((cp << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff);
if (++it != end) {
if (is_trail(*it)) {
cp += (mask8(*it) << 6) & 0xfff;
if (utf8::internal::is_trail(*it)) {
cp += (utf8::internal::mask8(*it) << 6) & 0xfff;
if (++it != end) {
if (is_trail(*it)) {
if (utf8::internal::is_trail(*it)) {
cp += (*it) & 0x3f;
if (code_point)
@ -258,7 +258,7 @@ namespace internal
uint32_t cp = 0;
// Determine the sequence length based on the lead octet
typedef typename std::iterator_traits<octet_iterator>::difference_type octet_difference_type;
octet_difference_type length = sequence_length(it);
octet_difference_type length = utf8::internal::sequence_length(it);
if (length == 0)
return INVALID_LEAD;
@ -266,23 +266,23 @@ namespace internal
utf_error err = UTF8_OK;
switch (length) {
case 1:
err = get_sequence_1(it, end, &cp);
err = utf8::internal::get_sequence_1(it, end, &cp);
break;
case 2:
err = get_sequence_2(it, end, &cp);
err = utf8::internal::get_sequence_2(it, end, &cp);
break;
case 3:
err = get_sequence_3(it, end, &cp);
err = utf8::internal::get_sequence_3(it, end, &cp);
break;
case 4:
err = get_sequence_4(it, end, &cp);
err = utf8::internal::get_sequence_4(it, end, &cp);
break;
}
if (err == UTF8_OK) {
// Decoding succeeded. Now, security checks...
if (is_code_point_valid(cp)) {
if (!is_overlong_sequence(cp, length)){
if (utf8::internal::is_code_point_valid(cp)) {
if (!utf8::internal::is_overlong_sequence(cp, length)){
// Passed! Return here.
if (code_point)
*code_point = cp;
@ -303,7 +303,7 @@ namespace internal
template <typename octet_iterator>
inline utf_error validate_next(octet_iterator& it, octet_iterator end) {
return validate_next(it, end, 0);
return utf8::internal::validate_next(it, end, 0);
}
} // namespace internal
@ -318,7 +318,7 @@ namespace internal
{
octet_iterator result = start;
while (result != end) {
internal::utf_error err_code = internal::validate_next(result, end);
utf8::internal::utf_error err_code = utf8::internal::validate_next(result, end);
if (err_code != internal::UTF8_OK)
return result;
}
@ -328,27 +328,27 @@ namespace internal
template <typename octet_iterator>
inline bool is_valid(octet_iterator start, octet_iterator end)
{
return (find_invalid(start, end) == end);
return (utf8::find_invalid(start, end) == end);
}
template <typename octet_iterator>
inline bool starts_with_bom (octet_iterator it, octet_iterator end)
{
return (
((it != end) && (internal::mask8(*it++)) == bom[0]) &&
((it != end) && (internal::mask8(*it++)) == bom[1]) &&
((it != end) && (internal::mask8(*it)) == bom[2])
((it != end) && (utf8::internal::mask8(*it++)) == bom[0]) &&
((it != end) && (utf8::internal::mask8(*it++)) == bom[1]) &&
((it != end) && (utf8::internal::mask8(*it)) == bom[2])
);
}
//Deprecated in release 2.3
//Deprecated in release 2.3
template <typename octet_iterator>
inline bool is_bom (octet_iterator it)
{
return (
(internal::mask8(*it++)) == bom[0] &&
(internal::mask8(*it++)) == bom[1] &&
(internal::mask8(*it)) == bom[2]
(utf8::internal::mask8(*it++)) == bom[0] &&
(utf8::internal::mask8(*it++)) == bom[1] &&
(utf8::internal::mask8(*it)) == bom[2]
);
}
} // namespace utf8

44
source/utf8/unchecked.h
View file

@ -60,7 +60,7 @@ namespace utf8
template <typename octet_iterator>
uint32_t next(octet_iterator& it)
{
uint32_t cp = internal::mask8(*it);
uint32_t cp = utf8::internal::mask8(*it);
typename std::iterator_traits<octet_iterator>::difference_type length = utf8::internal::sequence_length(it);
switch (length) {
case 1:
@ -71,15 +71,15 @@ namespace utf8
break;
case 3:
++it;
cp = ((cp << 12) & 0xffff) + ((internal::mask8(*it) << 6) & 0xfff);
cp = ((cp << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff);
++it;
cp += (*it) & 0x3f;
break;
case 4:
++it;
cp = ((cp << 18) & 0x1fffff) + ((internal::mask8(*it) << 12) & 0x3ffff);
cp = ((cp << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff);
++it;
cp += (internal::mask8(*it) << 6) & 0xfff;
cp += (utf8::internal::mask8(*it) << 6) & 0xfff;
++it;
cp += (*it) & 0x3f;
break;
@ -91,29 +91,29 @@ namespace utf8
template <typename octet_iterator>
uint32_t peek_next(octet_iterator it)
{
return next(it);
return utf8::unchecked::next(it);
}
template <typename octet_iterator>
uint32_t prior(octet_iterator& it)
{
while (internal::is_trail(*(--it))) ;
while (utf8::internal::is_trail(*(--it))) ;
octet_iterator temp = it;
return next(temp);
return utf8::unchecked::next(temp);
}
// Deprecated in versions that include prior, but only for the sake of consistency (see utf8::previous)
template <typename octet_iterator>
inline uint32_t previous(octet_iterator& it)
{
return prior(it);
return utf8::unchecked::prior(it);
}
template <typename octet_iterator, typename distance_type>
void advance (octet_iterator& it, distance_type n)
{
for (distance_type i = 0; i < n; ++i)
next(it);
utf8::unchecked::next(it);
}
template <typename octet_iterator>
@ -122,7 +122,7 @@ namespace utf8
{
typename std::iterator_traits<octet_iterator>::difference_type dist;
for (dist = 0; first < last; ++dist)
next(first);
utf8::unchecked::next(first);
return dist;
}
@ -130,13 +130,13 @@ namespace utf8
octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result)
{
while (start != end) {
uint32_t cp = internal::mask16(*start++);
uint32_t cp = utf8::internal::mask16(*start++);
// Take care of surrogate pairs first
if (internal::is_lead_surrogate(cp)) {
uint32_t trail_surrogate = internal::mask16(*start++);
if (utf8::internal::is_lead_surrogate(cp)) {
uint32_t trail_surrogate = utf8::internal::mask16(*start++);
cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET;
}
result = append(cp, result);
result = utf8::unchecked::append(cp, result);
}
return result;
}
@ -145,7 +145,7 @@ namespace utf8
u16bit_iterator utf8to16 (octet_iterator start, octet_iterator end, u16bit_iterator result)
{
while (start < end) {
uint32_t cp = next(start);
uint32_t cp = utf8::unchecked::next(start);
if (cp > 0xffff) { //make a surrogate pair
*result++ = static_cast<uint16_t>((cp >> 10) + internal::LEAD_OFFSET);
*result++ = static_cast<uint16_t>((cp & 0x3ff) + internal::TRAIL_SURROGATE_MIN);
@ -160,7 +160,7 @@ namespace utf8
octet_iterator utf32to8 (u32bit_iterator start, u32bit_iterator end, octet_iterator result)
{
while (start != end)
result = append(*(start++), result);
result = utf8::unchecked::append(*(start++), result);
return result;
}
@ -169,7 +169,7 @@ namespace utf8
u32bit_iterator utf8to32 (octet_iterator start, octet_iterator end, u32bit_iterator result)
{
while (start < end)
(*result++) = next(start);
(*result++) = utf8::unchecked::next(start);
return result;
}
@ -186,7 +186,7 @@ namespace utf8
uint32_t operator * () const
{
octet_iterator temp = it;
return next(temp);
return utf8::unchecked::next(temp);
}
bool operator == (const iterator& rhs) const
{
@ -198,24 +198,24 @@ namespace utf8
}
iterator& operator ++ ()
{
std::advance(it, internal::sequence_length(it));
::std::advance(it, utf8::internal::sequence_length(it));
return *this;
}
iterator operator ++ (int)
{
iterator temp = *this;
std::advance(it, internal::sequence_length(it));
::std::advance(it, utf8::internal::sequence_length(it));
return temp;
}
iterator& operator -- ()
{
prior(it);
utf8::unchecked::prior(it);
return *this;
}
iterator operator -- (int)
{
iterator temp = *this;
prior(it);
utf8::unchecked::prior(it);
return temp;
}
}; // class iterator

1
test_drivers/performance/iconvtest.cpp
View file

@ -1,4 +1,5 @@
#include <iconv.h>
#include <string.h>
#include "../../source/utf8.h"
#include "timer.h"
#include <fstream>