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; } }
char * u8_conv_to_encoding (const char *tocode, enum iconv_ilseq_handler handler, const uint8_t *src, size_t srclen, size_t *offsets, char *resultbuf, size_t *lengthp) { if (STRCASEEQ (tocode, "UTF-8", 'U','T','F','-','8',0,0,0,0)) { char *result; /* Conversion from UTF-8 to UTF-8. No need to go through iconv(). */ #if CONFIG_UNICODE_SAFETY if (u8_check (src, srclen)) { errno = EILSEQ; return NULL; } #endif /* Memory allocation. */ if (resultbuf != NULL && *lengthp >= srclen) result = resultbuf; else { result = (char *) malloc (srclen > 0 ? srclen : 1); if (result == NULL) { errno = ENOMEM; return NULL; } } memcpy (result, (const char *) src, srclen); *lengthp = srclen; return result; } else { char *result = resultbuf; size_t length = *lengthp; if (mem_iconveha ((const char *) src, srclen, "UTF-8", tocode, handler == iconveh_question_mark, handler, offsets, &result, &length) < 0) return NULL; if (result == NULL) /* when (resultbuf == NULL && length == 0) */ { result = (char *) malloc (1); if (result == NULL) { errno = ENOMEM; return NULL; } } *lengthp = length; return result; } }
uint8_t * u8_conv_from_encoding (const char *fromcode, enum iconv_ilseq_handler handler, const char *src, size_t srclen, size_t *offsets, uint8_t *resultbuf, size_t *lengthp) { if (STRCASEEQ (fromcode, "UTF-8", 'U','T','F','-','8',0,0,0,0)) { /* Conversion from UTF-8 to UTF-8. No need to go through iconv(). */ uint8_t *result; if (u8_check ((const uint8_t *) src, srclen)) { errno = EILSEQ; return NULL; } if (offsets != NULL) { size_t i; for (i = 0; i < srclen; ) { int count = u8_mblen ((const uint8_t *) src + i, srclen - i); /* We can rely on count > 0 because of the previous u8_check. */ if (count <= 0) abort (); offsets[i] = i; i++; while (--count > 0) offsets[i++] = (size_t)(-1); } } /* Memory allocation. */ if (resultbuf != NULL && *lengthp >= srclen) result = resultbuf; else { result = (uint8_t *) malloc (srclen > 0 ? srclen : 1); if (result == NULL) { errno = ENOMEM; return NULL; } } memcpy ((char *) result, src, srclen); *lengthp = srclen; return result; } else { char *result = (char *) resultbuf; size_t length = *lengthp; if (mem_iconveha (src, srclen, fromcode, "UTF-8", true, handler, offsets, &result, &length) < 0) return NULL; if (result == NULL) /* when (resultbuf == NULL && length == 0) */ { result = (char *) malloc (1); if (result == NULL) { errno = ENOMEM; return NULL; } } *lengthp = length; return (uint8_t *) result; } }
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 mem_iconveha() ------------------------- */ /* Test conversion from ISO-8859-2 to ISO-8859-1 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 char expected[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337"; for (o = 0; o < 2; o++) { size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "ISO-8859-2", "ISO-8859-1", false, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 0); if (o) { for (i = 0; i < 37; i++) ASSERT (offsets[i] == i); ASSERT (offsets[37] == MAGIC); free (offsets); } free (result); } } /* Test conversion from ISO-8859-2 to ISO-8859-1 with EILSEQ. */ for (h = 0; h < SIZEOF (handlers); h++) { enum iconv_ilseq_handler handler = handlers[h]; static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */ for (o = 0; o < 2; o++) { size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "ISO-8859-2", "ISO-8859-1", false, handler, offsets, &result, &length); switch (handler) { case iconveh_error: ASSERT (retval == -1 && errno == EILSEQ); ASSERT (result == NULL); if (o) free (offsets); break; case iconveh_question_mark: { static const char expected[] = "Rafa? Maszkowski"; ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 0); if (o) { for (i = 0; i < 16; i++) ASSERT (offsets[i] == i); ASSERT (offsets[16] == MAGIC); free (offsets); } free (result); } break; case iconveh_escape_sequence: { static const char expected[] = "Rafa\\u0142 Maszkowski"; ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 0); if (o) { for (i = 0; i < 16; i++) ASSERT (offsets[i] == (i < 5 ? i : i + 5)); ASSERT (offsets[16] == MAGIC); free (offsets); } free (result); } break; } } } /* 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 char 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); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "ISO-8859-1", "UTF-8", false, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, 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 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 char 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 (strlen (input)) : NULL); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "UTF-8", "ISO-8859-1", false, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 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 : (size_t)(-1))); 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 char input[] = "Rafa\305\202 Maszkowski"; /* Rafał Maszkowski */ for (o = 0; o < 2; o++) { size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "UTF-8", "ISO-8859-1", false, handler, offsets, &result, &length); switch (handler) { case iconveh_error: ASSERT (retval == -1 && errno == EILSEQ); ASSERT (result == NULL); if (o) free (offsets); break; case iconveh_question_mark: { static const char expected[] = "Rafa? Maszkowski"; ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 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 (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 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 char input[] = "\342"; for (o = 0; o < 2; o++) { size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "UTF-8", "ISO-8859-1", false, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == 0); if (o) { ASSERT (offsets[0] == 0); ASSERT (offsets[1] == 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 char 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); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "autodetect_jp", "UTF-8", false, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, 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 char 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); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "autodetect_jp", "UTF-8", false, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, 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 char 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); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "autodetect_jp", "UTF-8", false, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, 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 # if (((__GLIBC__ == 2 && __GLIBC_MINOR__ >= 2) || __GLIBC__ > 2) && !defined __UCLIBC__) || _LIBICONV_VERSION >= 0x0105 /* Test conversion from UTF-8 to ISO-8859-1 with transliteration. */ for (h = 0; h < SIZEOF (handlers); h++) { enum iconv_ilseq_handler handler = handlers[h]; static const char input[] = "Costs: 27 \342\202\254"; /* EURO SIGN */ static const char expected[] = "Costs: 27 EUR"; for (o = 0; o < 2; o++) { size_t *offsets = (o ? new_offsets (strlen (input)) : NULL); char *result = NULL; size_t length = 0; int retval = mem_iconveha (input, strlen (input), "UTF-8", "ISO-8859-1", true, handler, offsets, &result, &length); ASSERT (retval == 0); ASSERT (length == strlen (expected)); ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 0); if (o) { for (i = 0; i < 13; i++) ASSERT (offsets[i] == (i < 11 ? i : (size_t)(-1))); ASSERT (offsets[13] == MAGIC); free (offsets); } free (result); } } # endif /* ------------------------- Test str_iconveha() ------------------------- */ /* Test conversion from ISO-8859-2 to ISO-8859-1 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 char expected[] = "\304rger mit b\366sen B\374bchen ohne Augenma\337"; char *result = str_iconveha (input, "ISO-8859-2", "ISO-8859-1", false, handler); ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); free (result); } /* Test conversion from ISO-8859-2 to ISO-8859-1 with EILSEQ. */ for (h = 0; h < SIZEOF (handlers); h++) { enum iconv_ilseq_handler handler = handlers[h]; static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */ char *result = str_iconveha (input, "ISO-8859-2", "ISO-8859-1", false, handler); switch (handler) { case iconveh_error: ASSERT (result == NULL && errno == EILSEQ); break; case iconveh_question_mark: { static const char expected[] = "Rafa? Maszkowski"; ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); free (result); } break; case iconveh_escape_sequence: { static const char expected[] = "Rafa\\u0142 Maszkowski"; ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); free (result); } break; } } /* 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 char expected[] = "\303\204rger mit b\303\266sen B\303\274bchen ohne Augenma\303\237"; char *result = str_iconveha (input, "ISO-8859-1", "UTF-8", false, handler); ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); free (result); } /* 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 char 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"; char *result = str_iconveha (input, "UTF-8", "ISO-8859-1", false, handler); ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); 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 char input[] = "Costs: 27 \342\202\254"; /* EURO SIGN */ char *result = str_iconveha (input, "UTF-8", "ISO-8859-1", false, handler); switch (handler) { case iconveh_error: ASSERT (result == NULL && errno == EILSEQ); break; case iconveh_question_mark: { static const char expected[] = "Costs: 27 ?"; ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); free (result); } break; case iconveh_escape_sequence: { static const char expected[] = "Costs: 27 \\u20AC"; ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); 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 char input[] = "\342"; char *result = str_iconveha (input, "UTF-8", "ISO-8859-1", false, handler); ASSERT (result != NULL); ASSERT (strcmp (result, "") == 0); 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 char expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */ char *result = str_iconveha (input, "autodetect_jp", "UTF-8", false, handler); ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); 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 char expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */ char *result = str_iconveha (input, "autodetect_jp", "UTF-8", false, handler); ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); 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 char expected[] = "\343\201\223\343\202\223\343\201\253\343\201\241\343\201\257"; /* こんにちは */ char *result = str_iconveha (input, "autodetect_jp", "UTF-8", false, handler); ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); free (result); } # endif # if (((__GLIBC__ == 2 && __GLIBC_MINOR__ >= 2) || __GLIBC__ > 2) && !defined __UCLIBC__) || _LIBICONV_VERSION >= 0x0105 /* Test conversion from UTF-8 to ISO-8859-1 with transliteration. */ for (h = 0; h < SIZEOF (handlers); h++) { enum iconv_ilseq_handler handler = handlers[h]; static const char input[] = "Costs: 27 \342\202\254"; /* EURO SIGN */ static const char expected[] = "Costs: 27 EUR"; char *result = str_iconveha (input, "UTF-8", "ISO-8859-1", true, handler); ASSERT (result != NULL); ASSERT (strcmp (result, expected) == 0); free (result); } # endif #endif return 0; }