errno_t sss_utf8_case_eq(const uint8_t *s1, const uint8_t *s2)
{
/* Do a case-insensitive comparison.
* The input must be encoded in UTF8.
* We have no way of knowing the language,
* so we'll pass NULL for the language and
* hope for the best.
*/
int ret;
int resultp;
size_t n1, n2;
errno = 0;
n1 = u8_strlen(s1);
n2 = u8_strlen(s2);
ret = u8_casecmp(s1, n1,
s2, n2,
NULL, NULL,
&resultp);
if (ret < 0) {
/* An error occurred */
return errno;
}
if (resultp == 0) {
return EOK;
}
return ENOMATCH;
}
int
main ()
{
/* Empty string. */
{
static const uint8_t input[] = { 0 };
ASSERT (u8_strlen (input) == 0);
}
/* Simple string. */
{ /* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a) 日本語,中文,한글" */
static const uint8_t input[] =
{ 'G', 'r', 0xC3, 0xBC, 0xC3, 0x9F, ' ', 'G', 'o', 't', 't', '.', ' ',
0xD0, 0x97, 0xD0, 0xB4, 0xD1, 0x80, 0xD0, 0xB0, 0xD0, 0xB2, 0xD1, 0x81,
0xD1, 0x82, 0xD0, 0xB2, 0xD1, 0x83, 0xD0, 0xB9, 0xD1, 0x82, 0xD0, 0xB5,
'!', ' ', 'x', '=', '(', '-', 'b', 0xC2, 0xB1, 's', 'q', 'r', 't', '(',
'b', 0xC2, 0xB2, '-', '4', 'a', 'c', ')', ')', '/', '(', '2', 'a', ')',
' ', ' ', 0xE6, 0x97, 0xA5, 0xE6, 0x9C, 0xAC, 0xE8, 0xAA, 0x9E, ',',
0xE4, 0xB8, 0xAD, 0xE6, 0x96, 0x87, ',',
0xED, 0x95, 0x9C, 0xEA, 0xB8, 0x80, '\0'
};
ASSERT (u8_strlen (input) == SIZEOF (input) - 1);
}
return 0;
}
U8_EXPORT int u8_has_prefix(u8_string string,u8_string prefix,int casefold)
{
if ((casefold>0)&&(strncasecmp(string,prefix,u8_strlen(prefix))==0))
return 1;
else if ((casefold<=0)&&(strncmp(string,prefix,u8_strlen(prefix))==0))
return 1;
else return 0;
}
U8_EXPORT int u8_has_suffix(u8_string string,u8_string suffix,int casefold)
{
int stringlen=u8_strlen(string), suffixlen=u8_strlen(suffix);
if (suffixlen>stringlen) return 0;
else if ((casefold>0)&&(strcasecmp(string+(stringlen-suffixlen),suffix)==0))
return 1;
else if ((casefold<=0)&&(strcmp(string+(stringlen-suffixlen),suffix)==0))
return 1;
else return 0;
}
static void
test_function (uint8_t * (*my_asnprintf) (uint8_t *, size_t *, const char *, ...))
{
/* Test the support of the 's' conversion specifier for strings. */
{
const char *locale_string = "\303\204rger"; /* Ärger */
{
size_t length;
uint8_t *result =
my_asnprintf (NULL, &length, "%s %d", locale_string, 33, 44, 55);
static const uint8_t expected[] = "\303\204rger 33";
ASSERT (result != NULL);
ASSERT (u8_strcmp (result, expected) == 0);
ASSERT (length == u8_strlen (result));
free (result);
}
{ /* Width. */
size_t length;
uint8_t *result =
my_asnprintf (NULL, &length, "%10s %d", locale_string, 33, 44, 55);
static const uint8_t expected[] = " \303\204rger 33";
ASSERT (result != NULL);
ASSERT (u8_strcmp (result, expected) == 0);
ASSERT (length == u8_strlen (result));
free (result);
}
{ /* FLAG_LEFT. */
size_t length;
uint8_t *result =
my_asnprintf (NULL, &length, "%-10s %d", locale_string, 33, 44, 55);
static const uint8_t expected[] = "\303\204rger 33";
ASSERT (result != NULL);
ASSERT (u8_strcmp (result, expected) == 0);
ASSERT (length == u8_strlen (result));
free (result);
}
{ /* FLAG_ZERO: no effect. */
size_t length;
uint8_t *result =
my_asnprintf (NULL, &length, "%010s %d", locale_string, 33, 44, 55);
static const uint8_t expected[] = " \303\204rger 33";
ASSERT (result != NULL);
ASSERT (u8_strcmp (result, expected) == 0);
ASSERT (length == u8_strlen (result));
free (result);
}
}
}
char *
u8_strconv_to_encoding (const uint8_t *string,
const char *tocode,
enum iconv_ilseq_handler handler)
{
char *result;
size_t length;
if (STRCASEEQ (tocode, "UTF-8", 'U','T','F','-','8',0,0,0,0))
{
/* Conversion from UTF-8 to UTF-8. No need to go through iconv(). */
length = u8_strlen (string) + 1;
#if CONFIG_UNICODE_SAFETY
if (u8_check (string, length))
{
errno = EILSEQ;
return NULL;
}
#endif
result = (char *) malloc (length);
if (result == NULL)
{
errno = ENOMEM;
return NULL;
}
memcpy (result, (const char *) string, length);
return result;
}
else
{
result = NULL;
length = 0;
if (mem_iconveha ((const char *) string, u8_strlen (string) + 1,
"UTF-8", tocode,
handler == iconveh_question_mark, handler,
NULL, &result, &length) < 0)
return NULL;
/* Verify the result has exactly one NUL byte, at the end. */
if (!(length > 0 && result[length-1] == '\0'
&& strlen (result) == length-1))
{
free (result);
errno = EILSEQ;
return NULL;
}
return result;
}
}
bool isInFont (char * theText) {
if (! fontTableSize) return 0;
if (! theText[0]) return 0;
// We don't want to compare strings. Only single characters allowed!
if (u8_strlen (theText) > 1) return false;
int i=0;
uint32_t c = u8_nextchar(theText, &i);
return u8_strchr(fontOrderString, c, &i);
}
static uint8_t *
my_xasprintf (const char *format, ...)
{
va_list args;
size_t length;
uint8_t *ret;
va_start (args, format);
ret = u8_vasnprintf (NULL, &length, format, args);
va_end (args);
if (ret != NULL)
ASSERT (length == u8_strlen (ret));
return ret;
}
static uint8_t *
my_xasprintf (const char *format, ...)
{
va_list args;
uint8_t buf[1000];
int retval;
size_t length;
uint8_t *result;
va_start (args, format);
retval = u8_vsnprintf (buf, sizeof (buf), format, args);
va_end (args);
if (retval < 0 || retval >= (int) sizeof (buf))
return NULL;
length = u8_strlen (buf);
result = XNMALLOC (length + 1, uint8_t);
u8_cpy (result, buf, length + 1);
return result;
}
static void _add_punycode_if_needed(_psl_vector_t *v, _psl_entry_t *e)
{
char *lookupname = NULL;
int rc;
uint8_t *lower, resbuf[256];
size_t len = sizeof(resbuf) - 1; /* leave space for additional \0 byte */
if (_str_is_ascii(e->label_buf))
return;
/* we need a conversion to lowercase */
lower = u8_tolower((uint8_t *)e->label_buf, u8_strlen((uint8_t *)e->label_buf), 0, UNINORM_NFKC, resbuf, &len);
if (!lower) {
/* fprintf(stderr, "u8_tolower(%s) failed (%d)\n", e->label_buf, errno); */
return;
}
/* u8_tolower() does not terminate the result string */
if (lower == resbuf) {
lower[len]=0;
} else {
uint8_t *tmp = lower;
lower = (uint8_t *)strndup((char *)lower, len);
free(tmp);
}
if ((rc = idn2_lookup_u8(lower, (uint8_t **)&lookupname, 0)) == IDN2_OK) {
if (strcmp(e->label_buf, lookupname)) {
_psl_entry_t suffix, *suffixp;
/* fprintf(stderr, "libidn '%s' -> '%s'\n", e->label_buf, lookupname); */
_suffix_init(&suffix, lookupname, strlen(lookupname));
suffix.wildcard = e->wildcard;
suffixp = _vector_get(v, _vector_add(v, &suffix));
suffixp->label = suffixp->label_buf; /* set label to changed address */
} /* else ignore */
} /* else
fprintf(stderr, "toASCII(%s) failed (%d): %s\n", lower, rc, idn2_strerror(rc)); */
if (lower != resbuf)
free(lower);
}
int
main ()
{
static enum iconv_ilseq_handler handlers[] =
{ iconveh_error, iconveh_question_mark, iconveh_escape_sequence };
size_t h;
size_t o;
size_t i;
#if HAVE_ICONV
/* Assume that iconv() supports at least the encodings ASCII, ISO-8859-1,
ISO-8859-2, and UTF-8. */
/* Test conversion from ISO-8859-1 to UTF-8 with no errors. */
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const char input[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337";
static const uint8_t expected[] = "\303\204rger mit b\303\266sen B\303\274bchen ohne Augenma\303\237";
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
size_t length;
uint8_t *result = u8_conv_from_encoding ("ISO-8859-1", handler,
input, strlen (input),
offsets,
NULL, &length);
ASSERT (result != NULL);
ASSERT (length == u8_strlen (expected));
ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
if (o)
{
for (i = 0; i < 37; i++)
ASSERT (offsets[i] == (i < 1 ? i :
i < 12 ? i + 1 :
i < 18 ? i + 2 :
i + 3));
ASSERT (offsets[37] == MAGIC);
free (offsets);
}
free (result);
}
}
/* Test conversion from ISO-8859-2 to UTF-8 with no errors. */
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */
static const uint8_t expected[] = "Rafa\305\202 Maszkowski";
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
size_t length;
uint8_t *result = u8_conv_from_encoding ("ISO-8859-2", handler,
input, strlen (input),
offsets,
NULL, &length);
ASSERT (result != NULL);
ASSERT (length == u8_strlen (expected));
ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
if (o)
{
for (i = 0; i < 16; i++)
ASSERT (offsets[i] == (i < 5 ? i :
i + 1));
ASSERT (offsets[16] == MAGIC);
free (offsets);
}
free (result);
}
}
/* autodetect_jp is only supported when iconv() support ISO-2022-JP-2. */
# if defined _LIBICONV_VERSION || !(defined _AIX || defined __sgi || defined __hpux || defined __osf__ || defined __sun)
/* Test conversions from autodetect_jp to UTF-8. */
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const char input[] = "\244\263\244\363\244\313\244\301\244\317"; /* こんにちは in EUC-JP */
static const uint8_t expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
size_t length;
uint8_t *result = u8_conv_from_encoding ("autodetect_jp", handler,
input, strlen (input),
offsets,
NULL, &length);
ASSERT (result != NULL);
ASSERT (length == u8_strlen (expected));
ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
if (o)
{
for (i = 0; i < 10; i++)
ASSERT (offsets[i] == ((i % 2) == 0 ? (i / 2) * 3 : (size_t)(-1)));
ASSERT (offsets[10] == MAGIC);
free (offsets);
}
free (result);
}
}
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const char input[] = "\202\261\202\361\202\311\202\277\202\315"; /* こんにちは in Shift_JIS */
static const uint8_t expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
size_t length;
uint8_t *result = u8_conv_from_encoding ("autodetect_jp", handler,
input, strlen (input),
offsets,
NULL, &length);
ASSERT (result != NULL);
ASSERT (length == u8_strlen (expected));
ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
if (o)
{
for (i = 0; i < 10; i++)
ASSERT (offsets[i] == ((i % 2) == 0 ? (i / 2) * 3 : (size_t)(-1)));
ASSERT (offsets[10] == MAGIC);
free (offsets);
}
free (result);
}
}
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const char input[] = "\033$B$3$s$K$A$O\033(B"; /* こんにちは in ISO-2022-JP-2 */
static const uint8_t expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
size_t length;
uint8_t *result = u8_conv_from_encoding ("autodetect_jp", handler,
input, strlen (input),
offsets,
NULL, &length);
ASSERT (result != NULL);
ASSERT (length == u8_strlen (expected));
ASSERT (u8_cmp (result, expected, u8_strlen (expected)) == 0);
if (o)
{
for (i = 0; i < 16; i++)
ASSERT (offsets[i] == (i == 0 ? 0 :
i == 5 ? 3 :
i == 7 ? 6 :
i == 9 ? 9 :
i == 11 ? 12 :
i == 13 ? 15 :
(size_t)(-1)));
ASSERT (offsets[16] == MAGIC);
free (offsets);
}
free (result);
}
}
# endif
#endif
return 0;
}
const char *wget_str_to_ascii(const char *src)
{
#ifdef WITH_LIBIDN2
if (wget_str_needs_encoding(src)) {
char *asc = NULL;
int rc;
#ifdef WITH_LIBUNISTRING
uint8_t *lower, resbuf[256];
size_t len = sizeof(resbuf) - 1; // leave space for additional \0 byte
// we need a conversion to lowercase
lower = u8_tolower((uint8_t *)src, u8_strlen((uint8_t *)src), 0, UNINORM_NFKC, resbuf, &len);
if (!lower) {
error_printf("u8_tolower(%s) failed (%d)\n", src, errno);
return src;
}
// u8_tolower() does not terminate the result string
if (lower == resbuf) {
lower[len]=0;
} else {
uint8_t *tmp = lower;
lower = (uint8_t *)wget_strmemdup((char *)lower, len);
xfree(tmp);
}
if ((rc = idn2_lookup_u8(lower, (uint8_t **)&asc, 0)) == IDN2_OK) {
debug_printf("idn2 '%s' -> '%s'\n", src, asc);
src = asc;
} else
error_printf(_("toASCII(%s) failed (%d): %s\n"), lower, rc, idn2_strerror(rc));
if (lower != resbuf)
xfree(lower);
#else
if ((rc = idn2_lookup_u8((uint8_t *)src, (uint8_t **)&asc, 0)) == IDN2_OK) {
debug_printf("idn2 '%s' -> '%s'\n", src, asc);
src = asc;
} else
error_printf(_("toASCII(%s) failed (%d): %s\n"), src, rc, idn2_strerror(rc));
#endif
}
#elif WITH_LIBIDN
if (wget_str_needs_encoding(src)) {
char *asc = NULL;
int rc;
if (_utf8_is_valid(src)) {
// idna_to_ascii_8z() automatically converts UTF-8 to lowercase
if ((rc = idna_to_ascii_8z(src, &asc, IDNA_USE_STD3_ASCII_RULES)) == IDNA_SUCCESS) {
// debug_printf("toASCII '%s' -> '%s'\n", src, asc);
src = asc;
} else
error_printf(_("toASCII failed (%d): %s\n"), rc, idna_strerror(rc));
}
else
error_printf(_("Invalid UTF-8 sequence not converted: '%s'\n"), src);
}
#else
if (wget_str_needs_encoding(src)) {
error_printf(_("toASCII not available: '%s'\n"), src);
}
#endif
return src;
}
/**
* psl_str_to_utf8lower:
* @str: string to convert
* @encoding: charset encoding of @str, e.g. 'iso-8859-1' or %NULL
* @locale: locale of @str for to lowercase conversion, e.g. 'de' or %NULL
* @lower: return value containing the converted string
*
* This helper function converts a string to lowercase UTF-8 representation.
* Lowercase UTF-8 is needed as input to the domain checking functions.
*
* @lower is set to %NULL on error.
*
* The return value 'lower' must be freed after usage.
*
* Returns: psl_error_t value.
* PSL_SUCCESS: Success
* PSL_ERR_INVALID_ARG: @str is a %NULL value.
* PSL_ERR_CONVERTER: Failed to open the unicode converter with name @encoding
* PSL_ERR_TO_UTF16: Failed to convert @str to unicode
* PSL_ERR_TO_LOWER: Failed to convert unicode to lowercase
* PSL_ERR_TO_UTF8: Failed to convert unicode to UTF-8
*
* Since: 0.4
*/
psl_error_t psl_str_to_utf8lower(const char *str, const char *encoding, const char *locale, char **lower)
{
int ret = PSL_ERR_INVALID_ARG;
if (lower)
*lower = NULL;
if (!str)
return PSL_ERR_INVALID_ARG;
/* shortcut to avoid costly conversion */
if (_str_is_ascii(str)) {
if (lower) {
char *p;
*lower = strdup(str);
/* convert ASCII string to lowercase */
for (p = *lower; *p; p++)
if (isupper(*p))
*p = tolower(*p);
}
return PSL_SUCCESS;
}
#ifdef WITH_LIBICU
do {
size_t str_length = strlen(str);
UErrorCode status = 0;
UChar *utf16_dst, *utf16_lower;
int32_t utf16_dst_length;
char *utf8_lower;
UConverter *uconv;
/* C89 allocation */
utf16_dst = alloca(sizeof(UChar) * (str_length * 2 + 1));
utf16_lower = alloca(sizeof(UChar) * (str_length * 2 + 1));
utf8_lower = alloca(str_length * 2 + 1);
uconv = ucnv_open(encoding, &status);
if (U_SUCCESS(status)) {
utf16_dst_length = ucnv_toUChars(uconv, utf16_dst, str_length * 2 + 1, str, str_length, &status);
ucnv_close(uconv);
if (U_SUCCESS(status)) {
int32_t utf16_lower_length = u_strToLower(utf16_lower, str_length * 2 + 1, utf16_dst, utf16_dst_length, locale, &status);
if (U_SUCCESS(status)) {
u_strToUTF8(utf8_lower, str_length * 8 + 1, NULL, utf16_lower, utf16_lower_length, &status);
if (U_SUCCESS(status)) {
if (lower)
*lower = strdup(utf8_lower);
ret = PSL_SUCCESS;
} else {
ret = PSL_ERR_TO_UTF8;
/* fprintf(stderr, "Failed to convert UTF-16 to UTF-8 (status %d)\n", status); */
}
} else {
ret = PSL_ERR_TO_LOWER;
/* fprintf(stderr, "Failed to convert UTF-16 to lowercase (status %d)\n", status); */
}
} else {
ret = PSL_ERR_TO_UTF16;
/* fprintf(stderr, "Failed to convert string to UTF-16 (status %d)\n", status); */
}
} else {
ret = PSL_ERR_CONVERTER;
/* fprintf(stderr, "Failed to open converter for '%s' (status %d)\n", encoding, status); */
}
} while (0);
#elif defined(WITH_LIBIDN2) || defined(WITH_LIBIDN)
do {
/* find out local charset encoding */
if (!encoding) {
encoding = nl_langinfo(CODESET);
if (!encoding || !*encoding)
encoding = "ASCII";
}
/* convert to UTF-8 */
if (strcasecmp(encoding, "utf-8")) {
iconv_t cd = iconv_open("utf-8", encoding);
if (cd != (iconv_t)-1) {
char *tmp = (char *)str; /* iconv won't change where str points to, but changes tmp itself */
size_t tmp_len = strlen(str);
size_t dst_len = tmp_len * 6, dst_len_tmp = dst_len;
char *dst = malloc(dst_len + 1), *dst_tmp = dst;
if (iconv(cd, &tmp, &tmp_len, &dst_tmp, &dst_len_tmp) != (size_t)-1) {
uint8_t *resbuf = malloc(dst_len * 2 + 1);
size_t len = dst_len * 2; /* leave space for additional \0 byte */
if ((dst = (char *)u8_tolower((uint8_t *)dst, dst_len - dst_len_tmp, 0, UNINORM_NFKC, resbuf, &len))) {
/* u8_tolower() does not terminate the result string */
if (lower)
*lower = strndup((char *)dst, len);
} else {
ret = PSL_ERR_TO_LOWER;
/* fprintf(stderr, "Failed to convert UTF-8 to lowercase (errno %d)\n", errno); */
}
if (lower)
*lower = strndup(dst, dst_len - dst_len_tmp);
ret = PSL_SUCCESS;
} else {
ret = PSL_ERR_TO_UTF8;
/* fprintf(stderr, "Failed to convert '%s' string into '%s' (%d)\n", src_encoding, dst_encoding, errno); */
}
free(dst);
iconv_close(cd);
} else {
ret = PSL_ERR_TO_UTF8;
/* fprintf(stderr, "Failed to prepare encoding '%s' into '%s' (%d)\n", src_encoding, dst_encoding, errno); */
}
} else
ret = PSL_SUCCESS;
/* convert to lowercase */
if (ret == PSL_SUCCESS) {
uint8_t *dst, resbuf[256];
size_t len = sizeof(resbuf) - 1; /* leave space for additional \0 byte */
/* we need a conversion to lowercase */
if ((dst = u8_tolower((uint8_t *)str, u8_strlen((uint8_t *)str), 0, UNINORM_NFKC, resbuf, &len))) {
/* u8_tolower() does not terminate the result string */
if (lower)
*lower = strndup((char *)dst, len);
} else {
ret = PSL_ERR_TO_LOWER;
/* fprintf(stderr, "Failed to convert UTF-8 to lowercase (errno %d)\n", errno); */
}
}
} while (0);
#endif
return ret;
}
int
u8_strwidth (const uint8_t *s, const char *encoding)
{
return u8_width (s, u8_strlen (s), encoding);
}
int stringLength (char * theText) {
return u8_strlen (theText);
}
int
main ()
{
static enum iconv_ilseq_handler handlers[] =
{ iconveh_error, iconveh_question_mark, iconveh_escape_sequence };
size_t h;
size_t o;
size_t i;
#if HAVE_ICONV
/* Assume that iconv() supports at least the encodings ASCII, ISO-8859-1,
ISO-8859-2, and UTF-8. */
/* Test conversion from UTF-8 to ISO-8859-1 with no errors. */
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const uint8_t input[] = "\303\204rger mit b\303\266sen B\303\274bchen ohne Augenma\303\237";
static const char expected[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337";
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (u8_strlen (input)) : NULL);
size_t length;
char *result = u8_conv_to_encoding ("ISO-8859-1", handler,
input, u8_strlen (input),
offsets,
NULL, &length);
ASSERT (result != NULL);
ASSERT (length == strlen (expected));
ASSERT (memcmp (result, expected, length) == 0);
if (o)
{
for (i = 0; i < 41; i++)
ASSERT (offsets[i] == (i < 1 ? i :
i == 1 ? (size_t)(-1) :
i < 13 ? i - 1 :
i == 13 ? (size_t)(-1) :
i < 20 ? i - 2 :
i == 20 ? (size_t)(-1) :
i < 40 ? i - 3 :
i == 40 ? (size_t)(-1) :
i - 4));
ASSERT (offsets[41] == MAGIC);
free (offsets);
}
free (result);
}
}
/* Test conversion from UTF-8 to ISO-8859-1 with EILSEQ. */
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const uint8_t input[] = "Rafa\305\202 Maszkowski"; /* Rafał Maszkowski */
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (u8_strlen (input)) : NULL);
size_t length = 0xdead;
char *result = u8_conv_to_encoding ("ISO-8859-1", handler,
input, u8_strlen (input),
offsets,
NULL, &length);
switch (handler)
{
case iconveh_error:
ASSERT (result == NULL);
ASSERT (errno == EILSEQ);
ASSERT (length == 0xdead);
break;
case iconveh_question_mark:
{
static const char expected[] = "Rafa? Maszkowski";
static const char expected_translit[] = "Rafal Maszkowski";
ASSERT (result != NULL);
ASSERT (length == strlen (expected));
ASSERT (memcmp (result, expected, length) == 0
|| memcmp (result, expected_translit, length) == 0);
if (o)
{
for (i = 0; i < 17; i++)
ASSERT (offsets[i] == (i < 5 ? i :
i == 5 ? (size_t)(-1) :
i - 1));
ASSERT (offsets[17] == MAGIC);
free (offsets);
}
free (result);
}
break;
case iconveh_escape_sequence:
{
static const char expected[] = "Rafa\\u0142 Maszkowski";
ASSERT (result != NULL);
ASSERT (length == strlen (expected));
ASSERT (memcmp (result, expected, length) == 0);
if (o)
{
for (i = 0; i < 17; i++)
ASSERT (offsets[i] == (i < 5 ? i :
i == 5 ? (size_t)(-1) :
i + 4));
ASSERT (offsets[17] == MAGIC);
free (offsets);
}
free (result);
}
break;
}
}
}
/* Test conversion from UTF-8 to ISO-8859-1 with EINVAL. */
for (h = 0; h < SIZEOF (handlers); h++)
{
enum iconv_ilseq_handler handler = handlers[h];
static const uint8_t input[] = "\342";
for (o = 0; o < 2; o++)
{
size_t *offsets = (o ? new_offsets (u8_strlen (input)) : NULL);
size_t length;
char *result = u8_conv_to_encoding ("ISO-8859-1", handler,
input, u8_strlen (input),
offsets,
NULL, &length);
ASSERT (result != NULL);
ASSERT (length == strlen (""));
if (o)
{
ASSERT (offsets[0] == 0);
ASSERT (offsets[1] == MAGIC);
free (offsets);
}
free (result);
}
}
#endif
return 0;
}
static void
sort_tag_create(char **sort_tag, char *src_tag)
{
const uint8_t *i_ptr;
const uint8_t *n_ptr;
const uint8_t *number;
uint8_t out[1024];
uint8_t *o_ptr;
int append_number;
ucs4_t puc;
int numlen;
size_t len;
int charlen;
/* Note: include terminating NUL in string length for u8_normalize */
if (*sort_tag)
{
DPRINTF(E_DBG, L_LIB, "Existing sort tag will be normalized: %s\n", *sort_tag);
o_ptr = u8_normalize(UNINORM_NFD, (uint8_t *)*sort_tag, strlen(*sort_tag) + 1, NULL, &len);
free(*sort_tag);
*sort_tag = (char *)o_ptr;
return;
}
if (!src_tag || ((len = strlen(src_tag)) == 0))
{
*sort_tag = NULL;
return;
}
// Set input pointer past article if present
if ((strncasecmp(src_tag, "a ", 2) == 0) && (len > 2))
i_ptr = (uint8_t *)(src_tag + 2);
else if ((strncasecmp(src_tag, "an ", 3) == 0) && (len > 3))
i_ptr = (uint8_t *)(src_tag + 3);
else if ((strncasecmp(src_tag, "the ", 4) == 0) && (len > 4))
i_ptr = (uint8_t *)(src_tag + 4);
else
i_ptr = (uint8_t *)src_tag;
// Poor man's natural sort. Makes sure we sort like this: a1, a2, a10, a11, a21, a111
// We do this by padding zeroes to (short) numbers. As an alternative we could have
// made a proper natural sort algorithm in sqlext.c, but we don't, since we don't
// want any risk of hurting response times
memset(&out, 0, sizeof(out));
o_ptr = (uint8_t *)&out;
number = NULL;
append_number = 0;
do
{
n_ptr = u8_next(&puc, i_ptr);
if (uc_is_digit(puc))
{
if (!number) // We have encountered the beginning of a number
number = i_ptr;
append_number = (n_ptr == NULL); // If last char in string append number now
}
else
{
if (number)
append_number = 1; // A number has ended so time to append it
else
{
charlen = u8_strmblen(i_ptr);
if (charlen >= 0)
o_ptr = u8_stpncpy(o_ptr, i_ptr, charlen); // No numbers in sight, just append char
}
}
// Break if less than 100 bytes remain (prevent buffer overflow)
if (sizeof(out) - u8_strlen(out) < 100)
break;
// Break if number is very large (prevent buffer overflow)
if (number && (i_ptr - number > 50))
break;
if (append_number)
{
numlen = i_ptr - number;
if (numlen < 5) // Max pad width
{
u8_strcpy(o_ptr, (uint8_t *)"00000");
o_ptr += (5 - numlen);
}
o_ptr = u8_stpncpy(o_ptr, number, numlen + u8_strmblen(i_ptr));
number = NULL;
append_number = 0;
}
i_ptr = n_ptr;
}
while (n_ptr);
*sort_tag = (char *)u8_normalize(UNINORM_NFD, (uint8_t *)&out, u8_strlen(out) + 1, NULL, &len);
}
