/**
* Convert the len characters long character sequence
* given in input that is in the given input charset
* to a string in given output charset.
*
* @param input input string
* @param len number of bytes in @a input
* @param input_charset character set used for @a input
* @param output_charset desired character set for the return value
* @return the converted string (0-terminated),
* if conversion fails, a copy of the orignal
* string is returned.
*/
char *
GNUNET_STRINGS_conv (const char *input,
size_t len,
const char *input_charset,
const char *output_charset)
{
char *ret;
uint8_t *u8_string;
char *encoded_string;
size_t u8_string_length;
size_t encoded_string_length;
u8_string = u8_conv_from_encoding (input_charset,
iconveh_error,
input, len,
NULL, NULL,
&u8_string_length);
if (NULL == u8_string)
{
LOG_STRERROR (GNUNET_ERROR_TYPE_WARNING, "u8_conv_from_encoding");
goto fail;
}
if (0 == strcmp (output_charset, "UTF-8"))
{
ret = GNUNET_malloc (u8_string_length + 1);
memcpy (ret, u8_string, u8_string_length);
ret[u8_string_length] = '\0';
free (u8_string);
return ret;
}
encoded_string = u8_conv_to_encoding (output_charset, iconveh_error,
u8_string, u8_string_length,
NULL, NULL,
&encoded_string_length);
free (u8_string);
if (NULL == encoded_string)
{
LOG_STRERROR (GNUNET_ERROR_TYPE_WARNING, "u8_conv_to_encoding");
goto fail;
}
ret = GNUNET_malloc (encoded_string_length + 1);
memcpy (ret, encoded_string, encoded_string_length);
ret[encoded_string_length] = '\0';
free (encoded_string);
return ret;
fail:
LOG (GNUNET_ERROR_TYPE_WARNING, _("Character sets requested were `%s'->`%s'\n"),
"UTF-8", output_charset);
ret = GNUNET_malloc (len + 1);
memcpy (ret, input, len);
ret[len] = '\0';
return ret;
}
static uint8_t *
ulc_u8_casefold (const char *s, size_t n, const char *iso639_language,
uninorm_t nf,
uint8_t *resultbuf, size_t *lengthp)
{
uint8_t convbuf[2048 / sizeof (uint8_t)];
uint8_t *conv;
size_t conv_length;
uint8_t *result;
/* Convert the string to UTF-8. */
conv_length = sizeof (convbuf) / sizeof (uint8_t);
conv =
u8_conv_from_encoding (locale_charset (), iconveh_error, s, n, NULL,
convbuf, &conv_length);
if (conv == NULL)
/* errno is set here. */
return NULL;
/* Case-fold and normalize. */
result = u8_casefold (conv, conv_length, iso639_language, nf,
resultbuf, lengthp);
if (result == NULL)
{
if (conv != convbuf)
{
int saved_errno = errno;
free (conv);
errno = saved_errno;
}
return NULL;
}
if (conv != convbuf)
free (conv);
return result;
}
void
ulc_wordbreaks (const char *s, size_t n, char *p)
{
if (n > 0)
{
const char *encoding = locale_charset ();
if (is_utf8_encoding (encoding))
u8_wordbreaks ((const uint8_t *) s, n, p);
else
{
/* Convert the string to UTF-8 and build a translation table
from offsets into s to offsets into the translated string. */
size_t *offsets = (size_t *) malloc (n * sizeof (size_t));
if (offsets != NULL)
{
uint8_t *t;
size_t m;
t = u8_conv_from_encoding (encoding, iconveh_question_mark,
s, n, offsets, NULL, &m);
if (t != NULL)
{
char *q = (char *) (m > 0 ? malloc (m) : NULL);
if (m == 0 || q != NULL)
{
size_t i;
/* Determine the word breaks of the UTF-8 string. */
u8_wordbreaks (t, m, q);
/* Translate the result back to the original string. */
memset (p, 0, n);
for (i = 0; i < n; i++)
if (offsets[i] != (size_t)(-1))
p[i] = q[offsets[i]];
free (q);
free (t);
free (offsets);
return;
}
free (t);
}
free (offsets);
}
/* Impossible to convert. */
#if C_CTYPE_ASCII
if (is_all_ascii (s, n))
{
/* ASCII is a subset of UTF-8. */
u8_wordbreaks ((const uint8_t *) s, n, p);
return;
}
#endif
/* We have a non-ASCII string and cannot convert it.
Don't produce any word breaks. */
memset (p, 0, n);
}
}
}
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;
}
void
ulc_possible_linebreaks (const char *s, size_t n, const char *encoding,
char *p)
{
if (n > 0)
{
if (is_utf8_encoding (encoding))
u8_possible_linebreaks ((const uint8_t *) s, n, encoding, p);
else
{
/* Convert the string to UTF-8 and build a translation table
from offsets into s to offsets into the translated string. */
size_t *offsets = (size_t *) malloc (n * sizeof (size_t));
if (offsets != NULL)
{
uint8_t *t;
size_t m;
t = u8_conv_from_encoding (encoding, iconveh_question_mark,
s, n, offsets, NULL, &m);
if (t != NULL)
{
char *q = (char *) (m > 0 ? malloc (m) : NULL);
if (m == 0 || q != NULL)
{
size_t i;
/* Determine the possible line breaks of the UTF-8
string. */
u8_possible_linebreaks (t, m, encoding, q);
/* Translate the result back to the original string. */
memset (p, UC_BREAK_PROHIBITED, n);
for (i = 0; i < n; i++)
if (offsets[i] != (size_t)(-1))
p[i] = q[offsets[i]];
free (q);
free (t);
free (offsets);
return;
}
free (t);
}
free (offsets);
}
/* Impossible to convert. */
#if C_CTYPE_ASCII
if (is_all_ascii (s, n))
{
/* ASCII is a subset of UTF-8. */
u8_possible_linebreaks ((const uint8_t *) s, n, encoding, p);
return;
}
#endif
/* We have a non-ASCII string and cannot convert it.
Don't produce line breaks except those already present in the
input string. All we assume here is that the encoding is
minimally ASCII compatible. */
{
const char *s_end = s + n;
while (s < s_end)
{
*p = (*s == '\n' ? UC_BREAK_MANDATORY : UC_BREAK_PROHIBITED);
s++;
p++;
}
}
}
}
}
int
ulc_width_linebreaks (const char *s, size_t n,
int width, int start_column, int at_end_columns,
const char *o, const char *encoding,
char *p)
{
if (n > 0)
{
if (is_utf8_encoding (encoding))
return u8_width_linebreaks ((const uint8_t *) s, n, width, start_column, at_end_columns, o, encoding, p);
else
{
/* Convert the string to UTF-8 and build a translation table
from offsets into s to offsets into the translated string. */
size_t *offsets = (size_t *) malloc (n * sizeof (size_t));
if (offsets != NULL)
{
uint8_t *t;
size_t m;
t = u8_conv_from_encoding (encoding, iconveh_question_mark,
s, n, offsets, NULL, &m);
if (t != NULL)
{
char *memory =
(char *) (m > 0 ? malloc (m + (o != NULL ? m : 0)) : NULL);
if (m == 0 || memory != NULL)
{
char *q = (char *) memory;
char *o8 = (o != NULL ? (char *) (q + m) : NULL);
int res_column;
size_t i;
/* Translate the overrides to the UTF-8 string. */
if (o != NULL)
{
memset (o8, UC_BREAK_UNDEFINED, m);
for (i = 0; i < n; i++)
if (offsets[i] != (size_t)(-1))
o8[offsets[i]] = o[i];
}
/* Determine the line breaks of the UTF-8 string. */
res_column =
u8_width_linebreaks (t, m, width, start_column, at_end_columns, o8, encoding, q);
/* Translate the result back to the original string. */
memset (p, UC_BREAK_PROHIBITED, n);
for (i = 0; i < n; i++)
if (offsets[i] != (size_t)(-1))
p[i] = q[offsets[i]];
free (memory);
free (t);
free (offsets);
return res_column;
}
free (t);
}
free (offsets);
}
/* Impossible to convert. */
#if C_CTYPE_ASCII
if (is_all_ascii (s, n))
{
/* ASCII is a subset of UTF-8. */
return u8_width_linebreaks ((const uint8_t *) s, n, width, start_column, at_end_columns, o, encoding, p);
}
#endif
/* We have a non-ASCII string and cannot convert it.
Don't produce line breaks except those already present in the
input string. All we assume here is that the encoding is
minimally ASCII compatible. */
{
const char *s_end = s + n;
while (s < s_end)
{
*p = ((o != NULL && *o == UC_BREAK_MANDATORY) || *s == '\n'
? UC_BREAK_MANDATORY
: UC_BREAK_PROHIBITED);
s++;
p++;
if (o != NULL)
o++;
}
/* We cannot compute widths in this case. */
}
}
}
return start_column;
}
void
ulc_grapheme_breaks (const char *s, size_t n, char *p)
{
if (n > 0)
{
const char *encoding = locale_charset ();
if (is_utf8_encoding (encoding))
u8_grapheme_breaks ((const uint8_t *) s, n, p);
else
{
/* Convert the string to UTF-8 and build a translation table
from offsets into s to offsets into the translated string. */
size_t *offsets = (size_t *) malloc (n * sizeof (size_t));
if (offsets != NULL)
{
uint8_t *t;
size_t m;
t = u8_conv_from_encoding (encoding, iconveh_question_mark,
s, n, offsets, NULL, &m);
if (t != NULL)
{
char *q = (char *) (m > 0 ? malloc (m) : NULL);
if (m == 0 || q != NULL)
{
size_t i;
/* Determine the grapheme breaks of the UTF-8 string. */
u8_grapheme_breaks (t, m, q);
/* Translate the result back to the original string. */
memset (p, 0, n);
for (i = 0; i < n; i++)
if (offsets[i] != (size_t)(-1))
p[i] = q[offsets[i]];
free (q);
free (t);
free (offsets);
return;
}
free (t);
}
free (offsets);
}
/* Impossible to convert. */
#if C_CTYPE_ASCII
/* Fall back to ASCII as best we can. */
ascii_grapheme_breaks (s, n, p);
#else
/* We cannot make any assumptions. */
p[0] = 1;
memset (p + 1, 0, n - 1);
#endif
}
}
}
char *
unicode_fixup_string(char *str)
{
uint8_t *ret;
size_t len;
if (!str)
return NULL;
len = strlen(str);
/* String is valid UTF-8 */
if (!u8_check((uint8_t *)str, len))
return str;
ret = u8_conv_from_encoding("ascii", iconveh_question_mark, str, len, NULL, NULL, &len);
if (!ret)
{
DPRINTF(E_LOG, L_MISC, "Could not convert string '%s' to UTF-8: %s\n", str, strerror(errno));
return NULL;
}
return (char *)ret;
}
char *
unicode_fixup_string(char *str)
{
uint8_t *ret;
size_t len;
if (!str)
return NULL;
len = strlen(str);
/* String is valid UTF-8 */
if (!u8_check((uint8_t *)str, len))
{
if (len >= 3)
{
/* Check for and strip byte-order mark */
if (memcmp("\xef\xbb\xbf", str, 3) == 0)
memmove(str, str + 3, len - 3 + 1);
}
return str;
}
ret = u8_conv_from_encoding("ascii", iconveh_question_mark, str, len, NULL, NULL, &len);
if (!ret)
{
DPRINTF(E_LOG, L_MISC, "Could not convert string '%s' to UTF-8: %s\n", str, strerror(errno));
return NULL;
}
return (char *)ret;
}
