mirror of
https://github.com/php/php-src.git
synced 2025-08-18 15:08:55 +02:00
whitespaces -> tabs
no other changes done
This commit is contained in:
Antony Dovgal
parent
9cee8be28e
commit
792a2f6e65
1 changed files with 385 additions and 379 deletions
|
|
@ -35,383 +35,383 @@ ZEND_API zend_class_entry *unicodeConversionException;
|
||||||
/* {{{ zend_set_converter_error_mode */
|
/* {{{ zend_set_converter_error_mode */
|
||||||
void zend_set_converter_error_mode(UConverter *conv, uint8_t error_mode)
|
void zend_set_converter_error_mode(UConverter *conv, uint8_t error_mode)
|
||||||
{
|
{
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
|
|
||||||
switch (error_mode) {
|
switch (error_mode) {
|
||||||
case ZEND_CONV_ERROR_STOP:
|
case ZEND_CONV_ERROR_STOP:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_STOP, NULL, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_STOP, NULL, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
case ZEND_CONV_ERROR_SKIP:
|
case ZEND_CONV_ERROR_SKIP:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_SKIP, UCNV_SKIP_STOP_ON_ILLEGAL, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_SKIP, UCNV_SKIP_STOP_ON_ILLEGAL, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
case ZEND_CONV_ERROR_SUBST:
|
case ZEND_CONV_ERROR_SUBST:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_SUBSTITUTE, UCNV_SUB_STOP_ON_ILLEGAL, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_SUBSTITUTE, UCNV_SUB_STOP_ON_ILLEGAL, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
case ZEND_CONV_ERROR_ESCAPE_UNICODE:
|
case ZEND_CONV_ERROR_ESCAPE_UNICODE:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_UNICODE, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_UNICODE, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
case ZEND_CONV_ERROR_ESCAPE_ICU:
|
case ZEND_CONV_ERROR_ESCAPE_ICU:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_ICU, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_ICU, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
case ZEND_CONV_ERROR_ESCAPE_JAVA:
|
case ZEND_CONV_ERROR_ESCAPE_JAVA:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_JAVA, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_JAVA, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
case ZEND_CONV_ERROR_ESCAPE_XML_DEC:
|
case ZEND_CONV_ERROR_ESCAPE_XML_DEC:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_XML_DEC, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_XML_DEC, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
case ZEND_CONV_ERROR_ESCAPE_XML_HEX:
|
case ZEND_CONV_ERROR_ESCAPE_XML_HEX:
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_XML_HEX, NULL, NULL, &status);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_ESCAPE, UCNV_ESCAPE_XML_HEX, NULL, NULL, &status);
|
||||||
break;
|
break;
|
||||||
|
|
||||||
default:
|
default:
|
||||||
assert(0);
|
assert(0);
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_set_converter_subst_char */
|
/* {{{ zend_set_converter_subst_char */
|
||||||
void zend_set_converter_subst_char(UConverter *conv, UChar *subst_char, int8_t subst_char_len)
|
void zend_set_converter_subst_char(UConverter *conv, UChar *subst_char, int8_t subst_char_len)
|
||||||
{
|
{
|
||||||
char dest[8];
|
char dest[8];
|
||||||
int8_t dest_len = 8;
|
int8_t dest_len = 8;
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
UErrorCode temp = U_ZERO_ERROR;
|
UErrorCode temp = U_ZERO_ERROR;
|
||||||
const void *old_context;
|
const void *old_context;
|
||||||
UConverterFromUCallback old_cb;
|
UConverterFromUCallback old_cb;
|
||||||
|
|
||||||
if (!subst_char_len)
|
if (!subst_char_len)
|
||||||
return;
|
return;
|
||||||
|
|
||||||
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_STOP, NULL, &old_cb, &old_context, &temp);
|
ucnv_setFromUCallBack(conv, UCNV_FROM_U_CALLBACK_STOP, NULL, &old_cb, &old_context, &temp);
|
||||||
dest_len = ucnv_fromUChars(conv, dest, dest_len, subst_char, subst_char_len, &status);
|
dest_len = ucnv_fromUChars(conv, dest, dest_len, subst_char, subst_char_len, &status);
|
||||||
ucnv_setFromUCallBack(conv, old_cb, old_context, NULL, NULL, &temp);
|
ucnv_setFromUCallBack(conv, old_cb, old_context, NULL, NULL, &temp);
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
zend_error(E_WARNING, "Could not set substitution character for the converter");
|
zend_error(E_WARNING, "Could not set substitution character for the converter");
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
ucnv_setSubstChars(conv, dest, dest_len, &status);
|
ucnv_setSubstChars(conv, dest, dest_len, &status);
|
||||||
if (status == U_ILLEGAL_ARGUMENT_ERROR) {
|
if (status == U_ILLEGAL_ARGUMENT_ERROR) {
|
||||||
zend_error(E_WARNING, "Substitution character byte sequence is too short or long for this converter");
|
zend_error(E_WARNING, "Substitution character byte sequence is too short or long for this converter");
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_set_converter_encoding */
|
/* {{{ zend_set_converter_encoding */
|
||||||
int zend_set_converter_encoding(UConverter **converter, const char *encoding)
|
int zend_set_converter_encoding(UConverter **converter, const char *encoding)
|
||||||
{
|
{
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
UConverter *new_converter = NULL;
|
UConverter *new_converter = NULL;
|
||||||
|
|
||||||
if (!converter) {
|
if (!converter) {
|
||||||
return FAILURE;
|
return FAILURE;
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* The specified encoding might be the same as converter's existing one,
|
* The specified encoding might be the same as converter's existing one,
|
||||||
* which results in a no-op.
|
* which results in a no-op.
|
||||||
*/
|
*/
|
||||||
if (*converter && encoding && encoding[0]) {
|
if (*converter && encoding && encoding[0]) {
|
||||||
const char *current = ucnv_getName(*converter, &status);
|
const char *current = ucnv_getName(*converter, &status);
|
||||||
status = U_ZERO_ERROR; /* reset error */
|
status = U_ZERO_ERROR; /* reset error */
|
||||||
if (!ucnv_compareNames(current, encoding)) {
|
if (!ucnv_compareNames(current, encoding)) {
|
||||||
return SUCCESS;
|
return SUCCESS;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* If encoding is NULL, ucnv_open() will return a converter based on
|
* If encoding is NULL, ucnv_open() will return a converter based on
|
||||||
* the default platform encoding as determined by ucnv_getDefaultName().
|
* the default platform encoding as determined by ucnv_getDefaultName().
|
||||||
*/
|
*/
|
||||||
new_converter = ucnv_open(encoding, &status);
|
new_converter = ucnv_open(encoding, &status);
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
return FAILURE;
|
return FAILURE;
|
||||||
}
|
}
|
||||||
|
|
||||||
if (*converter) {
|
if (*converter) {
|
||||||
ucnv_close(*converter);
|
ucnv_close(*converter);
|
||||||
}
|
}
|
||||||
*converter = new_converter;
|
*converter = new_converter;
|
||||||
|
|
||||||
return SUCCESS;
|
return SUCCESS;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_copy_converter */
|
/* {{{ zend_copy_converter */
|
||||||
int zend_copy_converter(UConverter **target, UConverter *source)
|
int zend_copy_converter(UConverter **target, UConverter *source)
|
||||||
{
|
{
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
const char *encoding;
|
const char *encoding;
|
||||||
|
|
||||||
assert(source != NULL);
|
assert(source != NULL);
|
||||||
|
|
||||||
encoding = ucnv_getName(source, &status);
|
encoding = ucnv_getName(source, &status);
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
return FAILURE;
|
return FAILURE;
|
||||||
}
|
}
|
||||||
|
|
||||||
return zend_set_converter_encoding(target, encoding);
|
return zend_set_converter_encoding(target, encoding);
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_convert_to_unicode */
|
/* {{{ zend_convert_to_unicode */
|
||||||
ZEND_API void zend_convert_to_unicode(UConverter *conv, UChar **target, int *target_len, const char *source, int source_len, UErrorCode *status)
|
ZEND_API void zend_convert_to_unicode(UConverter *conv, UChar **target, int *target_len, const char *source, int source_len, UErrorCode *status)
|
||||||
{
|
{
|
||||||
UChar *buffer = NULL;
|
UChar *buffer = NULL;
|
||||||
UChar *output;
|
UChar *output;
|
||||||
int32_t buffer_len = 0;
|
int32_t buffer_len = 0;
|
||||||
int32_t converted = 0;
|
int32_t converted = 0;
|
||||||
const char *input = source;
|
const char *input = source;
|
||||||
UConverterType conv_type;
|
UConverterType conv_type;
|
||||||
|
|
||||||
if (U_FAILURE(*status)) {
|
if (U_FAILURE(*status)) {
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
|
|
||||||
ucnv_resetToUnicode(conv);
|
ucnv_resetToUnicode(conv);
|
||||||
conv_type = ucnv_getType(conv);
|
conv_type = ucnv_getType(conv);
|
||||||
|
|
||||||
switch (conv_type) {
|
switch (conv_type) {
|
||||||
case UCNV_SBCS:
|
case UCNV_SBCS:
|
||||||
case UCNV_LATIN_1:
|
case UCNV_LATIN_1:
|
||||||
case UCNV_US_ASCII:
|
case UCNV_US_ASCII:
|
||||||
/*
|
/*
|
||||||
* For single-byte charsets, 1 input byte = 1 output UChar
|
* For single-byte charsets, 1 input byte = 1 output UChar
|
||||||
*/
|
*/
|
||||||
buffer_len = source_len;
|
buffer_len = source_len;
|
||||||
break;
|
break;
|
||||||
|
|
||||||
default:
|
default:
|
||||||
/*
|
/*
|
||||||
* Initial estimate: 1.25 UChar's for every 2 source bytes + 2 (past a
|
* Initial estimate: 1.25 UChar's for every 2 source bytes + 2 (past a
|
||||||
* certain limit (2)). The rationale behind this is that (atleast
|
* certain limit (2)). The rationale behind this is that (atleast
|
||||||
* in the case of GB2312) it is possible that there are single byte
|
* in the case of GB2312) it is possible that there are single byte
|
||||||
* characters in the input string. By using an GD2312 text as
|
* characters in the input string. By using an GD2312 text as
|
||||||
* example it seemed that a value of 1.25 allowed for as little
|
* example it seemed that a value of 1.25 allowed for as little
|
||||||
* re-allocations as possible without over estimating the buffer
|
* re-allocations as possible without over estimating the buffer
|
||||||
* too much. In case there is a lot of single-byte characters
|
* too much. In case there is a lot of single-byte characters
|
||||||
* around a single multi-byte character this estimation is too low,
|
* around a single multi-byte character this estimation is too low,
|
||||||
* and then the re-allocation routines in the loop below kick in.
|
* and then the re-allocation routines in the loop below kick in.
|
||||||
* There we multiply by 1.33 and add 1 so that it's quite efficient
|
* There we multiply by 1.33 and add 1 so that it's quite efficient
|
||||||
* for smaller input strings without causing too many iterations of
|
* for smaller input strings without causing too many iterations of
|
||||||
* this loop.
|
* this loop.
|
||||||
*/
|
*/
|
||||||
buffer_len = (source_len > 2) ? ((source_len >> 1) + (source_len >> 3) + 2) : source_len;
|
buffer_len = (source_len > 2) ? ((source_len >> 1) + (source_len >> 3) + 2) : source_len;
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
|
|
||||||
while (1) {
|
while (1) {
|
||||||
buffer = eurealloc(buffer, buffer_len + 1);
|
buffer = eurealloc(buffer, buffer_len + 1);
|
||||||
output = buffer + converted;
|
output = buffer + converted;
|
||||||
ucnv_toUnicode(conv, &output, buffer + buffer_len, &input, source + source_len, NULL, TRUE, status);
|
ucnv_toUnicode(conv, &output, buffer + buffer_len, &input, source + source_len, NULL, TRUE, status);
|
||||||
converted = (int32_t) (output - buffer);
|
converted = (int32_t) (output - buffer);
|
||||||
if (*status == U_BUFFER_OVERFLOW_ERROR) {
|
if (*status == U_BUFFER_OVERFLOW_ERROR) {
|
||||||
buffer_len = (buffer_len * 1.33) + 1;
|
buffer_len = (buffer_len * 1.33) + 1;
|
||||||
*status = U_ZERO_ERROR;
|
*status = U_ZERO_ERROR;
|
||||||
} else {
|
} else {
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* We return the buffer in case of failure anyway. The caller may want to
|
* We return the buffer in case of failure anyway. The caller may want to
|
||||||
* use partially converted string for something.
|
* use partially converted string for something.
|
||||||
*/
|
*/
|
||||||
|
|
||||||
buffer[converted] = 0;
|
buffer[converted] = 0;
|
||||||
*target = buffer;
|
*target = buffer;
|
||||||
*target_len = converted;
|
*target_len = converted;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_convert_from_unicode */
|
/* {{{ zend_convert_from_unicode */
|
||||||
ZEND_API int zend_convert_from_unicode(UConverter *conv, char **target, int *target_len, const UChar *source, int source_len, UErrorCode *status)
|
ZEND_API int zend_convert_from_unicode(UConverter *conv, char **target, int *target_len, const UChar *source, int source_len, UErrorCode *status)
|
||||||
{
|
{
|
||||||
char *buffer = NULL;
|
char *buffer = NULL;
|
||||||
char *output;
|
char *output;
|
||||||
int32_t buffer_len = 0;
|
int32_t buffer_len = 0;
|
||||||
int32_t converted = 0;
|
int32_t converted = 0;
|
||||||
const UChar *input = source;
|
const UChar *input = source;
|
||||||
|
|
||||||
if (U_FAILURE(*status)) {
|
if (U_FAILURE(*status)) {
|
||||||
return 0;
|
return 0;
|
||||||
}
|
}
|
||||||
|
|
||||||
ucnv_resetFromUnicode(conv);
|
ucnv_resetFromUnicode(conv);
|
||||||
|
|
||||||
buffer_len = ucnv_getMaxCharSize(conv) * source_len;
|
buffer_len = ucnv_getMaxCharSize(conv) * source_len;
|
||||||
|
|
||||||
while (1) {
|
while (1) {
|
||||||
buffer = erealloc(buffer, buffer_len + 1);
|
buffer = erealloc(buffer, buffer_len + 1);
|
||||||
output = buffer + converted;
|
output = buffer + converted;
|
||||||
ucnv_fromUnicode(conv, &output, buffer + buffer_len, &input, source + source_len, NULL, TRUE, status);
|
ucnv_fromUnicode(conv, &output, buffer + buffer_len, &input, source + source_len, NULL, TRUE, status);
|
||||||
converted = (int32_t) (output - buffer);
|
converted = (int32_t) (output - buffer);
|
||||||
if (*status == U_BUFFER_OVERFLOW_ERROR) {
|
if (*status == U_BUFFER_OVERFLOW_ERROR) {
|
||||||
buffer_len += 64;
|
buffer_len += 64;
|
||||||
*status = U_ZERO_ERROR;
|
*status = U_ZERO_ERROR;
|
||||||
} else {
|
} else {
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* We return the buffer in case of failure anyway. The caller may want to
|
* We return the buffer in case of failure anyway. The caller may want to
|
||||||
* use partially converted string for something.
|
* use partially converted string for something.
|
||||||
*/
|
*/
|
||||||
|
|
||||||
buffer[converted] = 0; /* NULL-terminate the output string */
|
buffer[converted] = 0; /* NULL-terminate the output string */
|
||||||
*target = buffer;
|
*target = buffer;
|
||||||
*target_len = converted;
|
*target_len = converted;
|
||||||
|
|
||||||
return input - source;
|
return input - source;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_convert_encodings */
|
/* {{{ zend_convert_encodings */
|
||||||
ZEND_API void zend_convert_encodings(UConverter *target_conv, UConverter *source_conv,
|
ZEND_API void zend_convert_encodings(UConverter *target_conv, UConverter *source_conv,
|
||||||
char **target, int *target_len,
|
char **target, int *target_len,
|
||||||
const char *source, int source_len, UErrorCode *status)
|
const char *source, int source_len, UErrorCode *status)
|
||||||
{
|
{
|
||||||
char *buffer = NULL;
|
char *buffer = NULL;
|
||||||
char *output;
|
char *output;
|
||||||
const char *input = source;
|
const char *input = source;
|
||||||
int32_t allocated = 0;
|
int32_t allocated = 0;
|
||||||
int32_t converted = 0;
|
int32_t converted = 0;
|
||||||
int8_t null_size;
|
int8_t null_size;
|
||||||
UChar pivot_buf[1024], *pivot, *pivot2;
|
UChar pivot_buf[1024], *pivot, *pivot2;
|
||||||
|
|
||||||
if (U_FAILURE(*status)) {
|
if (U_FAILURE(*status)) {
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
|
|
||||||
null_size = ucnv_getMinCharSize(target_conv);
|
null_size = ucnv_getMinCharSize(target_conv);
|
||||||
allocated = source_len + null_size;
|
allocated = source_len + null_size;
|
||||||
|
|
||||||
ucnv_resetToUnicode(source_conv);
|
ucnv_resetToUnicode(source_conv);
|
||||||
ucnv_resetFromUnicode(target_conv);
|
ucnv_resetFromUnicode(target_conv);
|
||||||
pivot = pivot2 = pivot_buf;
|
pivot = pivot2 = pivot_buf;
|
||||||
|
|
||||||
while (1) {
|
while (1) {
|
||||||
buffer = (char *) erealloc(buffer, allocated);
|
buffer = (char *) erealloc(buffer, allocated);
|
||||||
output = buffer + converted;
|
output = buffer + converted;
|
||||||
ucnv_convertEx(target_conv, source_conv, &output, buffer + allocated - null_size,
|
ucnv_convertEx(target_conv, source_conv, &output, buffer + allocated - null_size,
|
||||||
&input, source + source_len, pivot_buf, &pivot, &pivot2, pivot_buf + 1024, FALSE, TRUE, status);
|
&input, source + source_len, pivot_buf, &pivot, &pivot2, pivot_buf + 1024, FALSE, TRUE, status);
|
||||||
converted = (int32_t) (output - buffer);
|
converted = (int32_t) (output - buffer);
|
||||||
if (*status == U_BUFFER_OVERFLOW_ERROR) {
|
if (*status == U_BUFFER_OVERFLOW_ERROR) {
|
||||||
allocated += 1024;
|
allocated += 1024;
|
||||||
*status = U_ZERO_ERROR;
|
*status = U_ZERO_ERROR;
|
||||||
} else {
|
} else {
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
memset(buffer + converted, 0, null_size); /* NULL-terminate the output string */
|
memset(buffer + converted, 0, null_size); /* NULL-terminate the output string */
|
||||||
*target = buffer;
|
*target = buffer;
|
||||||
*target_len = converted;
|
*target_len = converted;
|
||||||
|
|
||||||
/* Report the conversion error */
|
/* Report the conversion error */
|
||||||
if (U_FAILURE(*status)) {
|
if (U_FAILURE(*status)) {
|
||||||
zend_error(E_NOTICE, "Error converting from codepage string to Unicode: %s", u_errorName(*status));
|
zend_error(E_NOTICE, "Error converting from codepage string to Unicode: %s", u_errorName(*status));
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_raise_conversion_error_ex */
|
/* {{{ zend_raise_conversion_error_ex */
|
||||||
ZEND_API void zend_raise_conversion_error_ex(char *message, UConverter *conv, int error_char_offset, int use_exception TSRMLS_DC)
|
ZEND_API void zend_raise_conversion_error_ex(char *message, UConverter *conv, int error_char_offset, int use_exception TSRMLS_DC)
|
||||||
{
|
{
|
||||||
UChar err_char[U16_MAX_LENGTH];
|
UChar err_char[U16_MAX_LENGTH];
|
||||||
int8_t err_char_len = sizeof(err_char);
|
int8_t err_char_len = sizeof(err_char);
|
||||||
UChar32 codepoint;
|
UChar32 codepoint;
|
||||||
const char *conv_name;
|
const char *conv_name;
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
char *reason_fmt = "%s (converter %s failed on character {U+%04X} at offset %d)";
|
char *reason_fmt = "%s (converter %s failed on character {U+%04X} at offset %d)";
|
||||||
char *no_reason_fmt = "%s";
|
char *no_reason_fmt = "%s";
|
||||||
char *message_fmt;
|
char *message_fmt;
|
||||||
|
|
||||||
if (!message)
|
if (!message)
|
||||||
return;
|
return;
|
||||||
|
|
||||||
ucnv_getInvalidUChars(conv, err_char, &err_char_len, &status);
|
ucnv_getInvalidUChars(conv, err_char, &err_char_len, &status);
|
||||||
conv_name = ucnv_getName(conv, &status);
|
conv_name = ucnv_getName(conv, &status);
|
||||||
/*
|
/*
|
||||||
* UTODO
|
* UTODO
|
||||||
* use some other standard than MIME? or fallback onto IANA? or use
|
* use some other standard than MIME? or fallback onto IANA? or use
|
||||||
* internal converter name? ponder
|
* internal converter name? ponder
|
||||||
*/
|
*/
|
||||||
conv_name = ucnv_getStandardName(conv_name, "MIME", &status);
|
conv_name = ucnv_getStandardName(conv_name, "MIME", &status);
|
||||||
codepoint = (err_char_len < 2) ? err_char[0] : U16_GET_SUPPLEMENTARY(err_char[0], err_char[1]);
|
codepoint = (err_char_len < 2) ? err_char[0] : U16_GET_SUPPLEMENTARY(err_char[0], err_char[1]);
|
||||||
;
|
;
|
||||||
|
|
||||||
message_fmt = conv ? reason_fmt : no_reason_fmt;
|
message_fmt = conv ? reason_fmt : no_reason_fmt;
|
||||||
|
|
||||||
if (use_exception) {
|
if (use_exception) {
|
||||||
zend_throw_exception_ex(unicodeConversionException, 0 TSRMLS_CC, message_fmt, message, conv_name?conv_name:"<unknown>", codepoint, error_char_offset);
|
zend_throw_exception_ex(unicodeConversionException, 0 TSRMLS_CC, message_fmt, message, conv_name?conv_name:"<unknown>", codepoint, error_char_offset);
|
||||||
} else {
|
} else {
|
||||||
zend_error(E_WARNING, message_fmt, message, conv_name?conv_name:"", codepoint, error_char_offset);
|
zend_error(E_WARNING, message_fmt, message, conv_name?conv_name:"", codepoint, error_char_offset);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zval_unicode_to_string */
|
/* {{{ zval_unicode_to_string */
|
||||||
ZEND_API int zval_unicode_to_string(zval *string, UConverter *conv TSRMLS_DC)
|
ZEND_API int zval_unicode_to_string(zval *string, UConverter *conv TSRMLS_DC)
|
||||||
{
|
{
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
int retval = TRUE;
|
int retval = TRUE;
|
||||||
char *s = NULL;
|
char *s = NULL;
|
||||||
int s_len;
|
int s_len;
|
||||||
int num_conv;
|
int num_conv;
|
||||||
|
|
||||||
UChar *u = Z_USTRVAL_P(string);
|
UChar *u = Z_USTRVAL_P(string);
|
||||||
int u_len = Z_USTRLEN_P(string);
|
int u_len = Z_USTRLEN_P(string);
|
||||||
|
|
||||||
num_conv = zend_convert_from_unicode(conv, &s, &s_len, u, u_len, &status);
|
num_conv = zend_convert_from_unicode(conv, &s, &s_len, u, u_len, &status);
|
||||||
|
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
int32_t offset = u_countChar32(u, num_conv)-1;
|
int32_t offset = u_countChar32(u, num_conv)-1;
|
||||||
|
|
||||||
if (s) {
|
if (s) {
|
||||||
efree(s);
|
efree(s);
|
||||||
}
|
}
|
||||||
zend_raise_conversion_error_ex("Could not convert Unicode string to binary string", conv, offset, (UG(from_u_error_mode) & ZEND_CONV_ERROR_EXCEPTION) TSRMLS_CC);
|
zend_raise_conversion_error_ex("Could not convert Unicode string to binary string", conv, offset, (UG(from_u_error_mode) & ZEND_CONV_ERROR_EXCEPTION) TSRMLS_CC);
|
||||||
retval = FAILURE;
|
retval = FAILURE;
|
||||||
}
|
}
|
||||||
|
|
||||||
ZVAL_STRINGL(string, s, s_len, 0);
|
ZVAL_STRINGL(string, s, s_len, 0);
|
||||||
|
|
||||||
efree((UChar*)u);
|
efree((UChar*)u);
|
||||||
return retval;
|
return retval;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zval_string_to_unicode_ex */
|
/* {{{ zval_string_to_unicode_ex */
|
||||||
ZEND_API int zval_string_to_unicode_ex(zval *string, UConverter *conv)
|
ZEND_API int zval_string_to_unicode_ex(zval *string, UConverter *conv)
|
||||||
{
|
{
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
int retval = TRUE;
|
int retval = TRUE;
|
||||||
UChar *u = NULL;
|
UChar *u = NULL;
|
||||||
int u_len;
|
int u_len;
|
||||||
|
|
||||||
char *s = Z_STRVAL_P(string);
|
char *s = Z_STRVAL_P(string);
|
||||||
int s_len = Z_STRLEN_P(string);
|
int s_len = Z_STRLEN_P(string);
|
||||||
|
|
||||||
Z_TYPE_P(string) = IS_UNICODE;
|
Z_TYPE_P(string) = IS_UNICODE;
|
||||||
zend_convert_to_unicode(conv, &u, &u_len, s, s_len, &status);
|
zend_convert_to_unicode(conv, &u, &u_len, s, s_len, &status);
|
||||||
ZVAL_UNICODEL(string, u, u_len, 0);
|
ZVAL_UNICODEL(string, u, u_len, 0);
|
||||||
|
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
retval = FALSE;
|
retval = FALSE;
|
||||||
}
|
}
|
||||||
|
|
||||||
efree(s);
|
efree(s);
|
||||||
return retval;
|
return retval;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
|
|
@ -425,20 +425,20 @@ ZEND_API int zval_string_to_unicode(zval *string TSRMLS_DC)
|
||||||
/* {{{ zend_cmp_unicode_and_string */
|
/* {{{ zend_cmp_unicode_and_string */
|
||||||
ZEND_API int zend_cmp_unicode_and_string(UChar *ustr, char* str, uint len)
|
ZEND_API int zend_cmp_unicode_and_string(UChar *ustr, char* str, uint len)
|
||||||
{
|
{
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
UChar *u = NULL;
|
UChar *u = NULL;
|
||||||
int u_len;
|
int u_len;
|
||||||
int retval = TRUE;
|
int retval = TRUE;
|
||||||
TSRMLS_FETCH();
|
TSRMLS_FETCH();
|
||||||
|
|
||||||
zend_convert_to_unicode(ZEND_U_CONVERTER(UG(runtime_encoding_conv)), &u, &u_len, str, len, &status);
|
zend_convert_to_unicode(ZEND_U_CONVERTER(UG(runtime_encoding_conv)), &u, &u_len, str, len, &status);
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
efree(u);
|
efree(u);
|
||||||
return FALSE;
|
return FALSE;
|
||||||
}
|
}
|
||||||
retval = u_memcmp(ustr, u, u_len);
|
retval = u_memcmp(ustr, u, u_len);
|
||||||
efree(u);
|
efree(u);
|
||||||
return retval;
|
return retval;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
|
|
@ -450,145 +450,151 @@ ZEND_API int zend_cmp_unicode_and_string(UChar *ustr, char* str, uint len)
|
||||||
*/
|
*/
|
||||||
ZEND_API int zend_cmp_unicode_and_literal(UChar *ustr, int ulen, char *str, int slen)
|
ZEND_API int zend_cmp_unicode_and_literal(UChar *ustr, int ulen, char *str, int slen)
|
||||||
{
|
{
|
||||||
int result;
|
int result;
|
||||||
uint len = MIN(ulen, slen);
|
uint len = MIN(ulen, slen);
|
||||||
|
|
||||||
while (len--) {
|
while (len--) {
|
||||||
result = (int)(uint16_t)*ustr - (int)(uint16_t)*str;
|
result = (int)(uint16_t)*ustr - (int)(uint16_t)*str;
|
||||||
if (result != 0)
|
if (result != 0)
|
||||||
return result;
|
return result;
|
||||||
ustr++;
|
ustr++;
|
||||||
str++;
|
str++;
|
||||||
}
|
}
|
||||||
|
|
||||||
return ulen - slen;
|
return ulen - slen;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_is_valid_identifier */
|
/* {{{ zend_is_valid_identifier */
|
||||||
ZEND_API int zend_is_valid_identifier(UChar *ident, int len)
|
ZEND_API int zend_is_valid_identifier(UChar *ident, int len)
|
||||||
{
|
{
|
||||||
UChar32 codepoint;
|
UChar32 codepoint;
|
||||||
int32_t i;
|
int32_t i;
|
||||||
int32_t ident_len = len;
|
int32_t ident_len = len;
|
||||||
UProperty id_prop = UCHAR_XID_START;
|
UProperty id_prop = UCHAR_XID_START;
|
||||||
|
|
||||||
for (i = 0; i < ident_len; ) {
|
for (i = 0; i < ident_len; ) {
|
||||||
U16_NEXT(ident, i, ident_len, codepoint);
|
U16_NEXT(ident, i, ident_len, codepoint);
|
||||||
if (!u_hasBinaryProperty(codepoint, id_prop) &&
|
if (!u_hasBinaryProperty(codepoint, id_prop) &&
|
||||||
codepoint != 0x5f) { /* special case for starting '_' */
|
codepoint != 0x5f) { /* special case for starting '_' */
|
||||||
return 0;
|
return 0;
|
||||||
}
|
}
|
||||||
id_prop = UCHAR_XID_CONTINUE;
|
id_prop = UCHAR_XID_CONTINUE;
|
||||||
}
|
}
|
||||||
|
|
||||||
return 1;
|
return 1;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_normalize_string */
|
/* {{{ zend_normalize_string */
|
||||||
static inline void zend_normalize_string(UChar **dest, int32_t *dest_len, UChar *src, int src_len, UErrorCode *status)
|
static inline void zend_normalize_string(UChar **dest, int32_t *dest_len, UChar *src, int src_len, UErrorCode *status)
|
||||||
{
|
{
|
||||||
UChar *buffer = NULL;
|
UChar *buffer = NULL;
|
||||||
int32_t buffer_len;
|
int32_t buffer_len;
|
||||||
|
|
||||||
buffer_len = src_len;
|
buffer_len = src_len;
|
||||||
while (1) {
|
while (1) {
|
||||||
*status = U_ZERO_ERROR;
|
*status = U_ZERO_ERROR;
|
||||||
buffer = eurealloc(buffer, buffer_len+1);
|
buffer = eurealloc(buffer, buffer_len+1);
|
||||||
buffer_len = unorm_normalize(src, src_len, UNORM_NFKC, 0, buffer, buffer_len, status);
|
buffer_len = unorm_normalize(src, src_len, UNORM_NFKC, 0, buffer, buffer_len, status);
|
||||||
if (*status != U_BUFFER_OVERFLOW_ERROR) {
|
if (*status != U_BUFFER_OVERFLOW_ERROR) {
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
if (U_SUCCESS(*status)) {
|
if (U_SUCCESS(*status)) {
|
||||||
buffer[buffer_len] = 0;
|
buffer[buffer_len] = 0;
|
||||||
*dest = buffer;
|
*dest = buffer;
|
||||||
*dest_len = buffer_len;
|
*dest_len = buffer_len;
|
||||||
} else {
|
} else {
|
||||||
efree(buffer);
|
efree(buffer);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_case_fold_string */
|
/* {{{ zend_case_fold_string */
|
||||||
ZEND_API void zend_case_fold_string(UChar **dest, int *dest_len, UChar *src, int src_len, uint32_t options, UErrorCode *status)
|
ZEND_API void zend_case_fold_string(UChar **dest, int *dest_len, UChar *src, int src_len, uint32_t options, UErrorCode *status)
|
||||||
{
|
{
|
||||||
UChar *buffer = NULL;
|
UChar *buffer = NULL;
|
||||||
int32_t buffer_len;
|
int32_t buffer_len;
|
||||||
|
|
||||||
buffer_len = src_len;
|
buffer_len = src_len;
|
||||||
while (1) {
|
while (1) {
|
||||||
*status = U_ZERO_ERROR;
|
*status = U_ZERO_ERROR;
|
||||||
buffer = eurealloc(buffer, buffer_len+1);
|
buffer = eurealloc(buffer, buffer_len+1);
|
||||||
buffer_len = u_strFoldCase(buffer, buffer_len, src, src_len, options, status);
|
buffer_len = u_strFoldCase(buffer, buffer_len, src, src_len, options, status);
|
||||||
if (*status != U_BUFFER_OVERFLOW_ERROR) {
|
if (*status != U_BUFFER_OVERFLOW_ERROR) {
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
if (U_SUCCESS(*status)) {
|
if (U_SUCCESS(*status)) {
|
||||||
buffer[buffer_len] = 0;
|
buffer[buffer_len] = 0;
|
||||||
*dest = buffer;
|
*dest = buffer;
|
||||||
*dest_len = buffer_len;
|
*dest_len = buffer_len;
|
||||||
} else {
|
} else {
|
||||||
efree(buffer);
|
efree(buffer);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
/* {{{ zend_normalize_identifier */
|
/* {{{ zend_normalize_identifier */
|
||||||
ZEND_API int zend_normalize_identifier(UChar **dest, int *dest_len, UChar *ident, int ident_len, zend_bool fold_case)
|
ZEND_API int zend_normalize_identifier(UChar **dest, int *dest_len, UChar *ident, int ident_len, zend_bool fold_case)
|
||||||
{
|
{
|
||||||
UChar *buffer = NULL;
|
UChar *buffer = NULL;
|
||||||
UChar *orig_ident = ident;
|
UChar *orig_ident = ident;
|
||||||
int32_t buffer_len;
|
int32_t buffer_len;
|
||||||
UErrorCode status = U_ZERO_ERROR;
|
UErrorCode status = U_ZERO_ERROR;
|
||||||
|
|
||||||
if (unorm_quickCheck(ident, ident_len, UNORM_NFKC, &status) != UNORM_YES) {
|
if (unorm_quickCheck(ident, ident_len, UNORM_NFKC, &status) != UNORM_YES) {
|
||||||
zend_normalize_string(&buffer, &buffer_len, ident, ident_len, &status);
|
zend_normalize_string(&buffer, &buffer_len, ident, ident_len, &status);
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
return 0;
|
return 0;
|
||||||
}
|
}
|
||||||
ident = buffer;
|
ident = buffer;
|
||||||
ident_len = buffer_len;
|
ident_len = buffer_len;
|
||||||
}
|
}
|
||||||
|
|
||||||
if (fold_case) {
|
if (fold_case) {
|
||||||
zend_case_fold_string(&buffer, &buffer_len, ident, ident_len, U_FOLD_CASE_DEFAULT, &status);
|
zend_case_fold_string(&buffer, &buffer_len, ident, ident_len, U_FOLD_CASE_DEFAULT, &status);
|
||||||
if (ident != orig_ident) {
|
if (ident != orig_ident) {
|
||||||
efree(ident);
|
efree(ident);
|
||||||
}
|
}
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
return 0;
|
return 0;
|
||||||
}
|
}
|
||||||
ident = buffer;
|
ident = buffer;
|
||||||
ident_len = buffer_len;
|
ident_len = buffer_len;
|
||||||
|
|
||||||
if (unorm_quickCheck(ident, ident_len, UNORM_NFKC, &status) != UNORM_YES) {
|
if (unorm_quickCheck(ident, ident_len, UNORM_NFKC, &status) != UNORM_YES) {
|
||||||
zend_normalize_string(&buffer, &buffer_len, ident, ident_len, &status);
|
zend_normalize_string(&buffer, &buffer_len, ident, ident_len, &status);
|
||||||
if (ident != orig_ident) {
|
if (ident != orig_ident) {
|
||||||
efree(ident);
|
efree(ident);
|
||||||
}
|
}
|
||||||
if (U_FAILURE(status)) {
|
if (U_FAILURE(status)) {
|
||||||
return 0;
|
return 0;
|
||||||
}
|
}
|
||||||
ident = buffer;
|
ident = buffer;
|
||||||
ident_len = buffer_len;
|
ident_len = buffer_len;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
*dest = ident;
|
*dest = ident;
|
||||||
*dest_len = ident_len;
|
*dest_len = ident_len;
|
||||||
return 1;
|
return 1;
|
||||||
}
|
}
|
||||||
/* }}} */
|
/* }}} */
|
||||||
|
|
||||||
void zend_register_unicode_exceptions(TSRMLS_D)
|
void zend_register_unicode_exceptions(TSRMLS_D)
|
||||||
{
|
{
|
||||||
zend_class_entry ce;
|
zend_class_entry ce;
|
||||||
|
|
||||||
INIT_CLASS_ENTRY(ce, "UnicodeConversionException", NULL);
|
INIT_CLASS_ENTRY(ce, "UnicodeConversionException", NULL);
|
||||||
unicodeConversionException = zend_register_internal_class_ex(&ce, zend_exception_get_default(TSRMLS_C), NULL TSRMLS_CC);
|
unicodeConversionException = zend_register_internal_class_ex(&ce, zend_exception_get_default(TSRMLS_C), NULL TSRMLS_CC);
|
||||||
}
|
}
|
||||||
|
|
||||||
/* vim: set fdm=marker et sts=4: */
|
/*
|
||||||
|
* Local variables:
|
||||||
|
* tab-width: 4
|
||||||
|
* c-basic-offset: 4
|
||||||
|
* indent-tabs-mode: t
|
||||||
|
* End:
|
||||||
|
*/
|
||||||
|
|
|
||||||
Loading…
Add table
Add a link
Reference in a new issue