/* Test of character set conversion with error handling.
- Copyright (C) 2007-2008 Free Software Foundation, Inc.
+ Copyright (C) 2007-2009 Free Software Foundation, Inc.
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
{
static enum iconv_ilseq_handler handlers[] =
{ iconveh_error, iconveh_question_mark, iconveh_escape_sequence };
+ size_t indirect;
size_t h;
size_t o;
size_t i;
iconv_t cd_utf8_to_88591 = iconv_open ("ISO-8859-1", "UTF-8");
iconv_t cd_88592_to_utf8 = iconv_open ("UTF-8", "ISO-8859-2");
iconv_t cd_utf8_to_88592 = iconv_open ("ISO-8859-2", "UTF-8");
+ iconv_t cd_utf7_to_utf8 = iconv_open ("UTF-8", "UTF-7");
+ iconveh_t cdeh_88592_to_88591;
+ iconveh_t cdeh_88592_to_88591_indirectly;
+ iconveh_t cdeh_88591_to_utf8;
+ iconveh_t cdeh_utf8_to_88591;
+ iconveh_t cdeh_utf7_to_utf8;
ASSERT (cd_88591_to_utf8 != (iconv_t)(-1));
ASSERT (cd_utf8_to_88591 != (iconv_t)(-1));
ASSERT (cd_88592_to_utf8 != (iconv_t)(-1));
ASSERT (cd_utf8_to_88592 != (iconv_t)(-1));
+ cdeh_88592_to_88591.cd = cd_88592_to_88591;
+ cdeh_88592_to_88591.cd1 = cd_88592_to_utf8;
+ cdeh_88592_to_88591.cd2 = cd_utf8_to_88591;
+
+ cdeh_88592_to_88591_indirectly.cd = (iconv_t)(-1);
+ cdeh_88592_to_88591_indirectly.cd1 = cd_88592_to_utf8;
+ cdeh_88592_to_88591_indirectly.cd2 = cd_utf8_to_88591;
+
+ cdeh_88591_to_utf8.cd = cd_88591_to_utf8;
+ cdeh_88591_to_utf8.cd1 = cd_88591_to_utf8;
+ cdeh_88591_to_utf8.cd2 = (iconv_t)(-1);
+
+ cdeh_utf8_to_88591.cd = cd_utf8_to_88591;
+ cdeh_utf8_to_88591.cd1 = (iconv_t)(-1);
+ cdeh_utf8_to_88591.cd2 = cd_utf8_to_88591;
+
+ cdeh_utf7_to_utf8.cd = cd_utf7_to_utf8;
+ cdeh_utf7_to_utf8.cd1 = cd_utf7_to_utf8;
+ cdeh_utf7_to_utf8.cd2 = (iconv_t)(-1);
+
/* ------------------------ Test mem_cd_iconveh() ------------------------ */
/* Test conversion from ISO-8859-2 to ISO-8859-1 with no errors. */
- for (h = 0; h < SIZEOF (handlers); h++)
+ for (indirect = 0; indirect <= 1; indirect++)
{
- 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++)
+ for (h = 0; h < SIZEOF (handlers); h++)
{
- size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
- char *result = NULL;
- size_t length = 0;
- int retval = mem_cd_iconveh (input, strlen (input),
- cd_88592_to_88591,
- cd_88592_to_utf8, cd_utf8_to_88591,
- handler,
- offsets,
- &result, &length);
- ASSERT (retval == 0);
- ASSERT (length == strlen (expected));
- ASSERT (result != NULL && memcmp (result, expected, strlen (expected)) == 0);
- if (o)
+ 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++)
{
- for (i = 0; i < 37; i++)
- ASSERT (offsets[i] == i);
- ASSERT (offsets[37] == MAGIC);
- free (offsets);
+ size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
+ char *result = NULL;
+ size_t length = 0;
+ int retval = mem_cd_iconveh (input, strlen (input),
+ (indirect
+ ? &cdeh_88592_to_88591_indirectly
+ : &cdeh_88592_to_88591),
+ 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);
}
- free (result);
}
}
/* Test conversion from ISO-8859-2 to ISO-8859-1 with EILSEQ. */
- for (h = 0; h < SIZEOF (handlers); h++)
+ for (indirect = 0; indirect <= 1; indirect++)
{
- enum iconv_ilseq_handler handler = handlers[h];
- static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */
- for (o = 0; o < 2; o++)
+ for (h = 0; h < SIZEOF (handlers); h++)
{
- size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
- char *result = NULL;
- size_t length = 0;
- int retval = mem_cd_iconveh (input, strlen (input),
- cd_88592_to_88591,
- cd_88592_to_utf8, cd_utf8_to_88591,
- handler,
- offsets,
- &result, &length);
- switch (handler)
+ enum iconv_ilseq_handler handler = handlers[h];
+ static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */
+ for (o = 0; o < 2; o++)
{
- 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);
+ size_t *offsets = (o ? new_offsets (strlen (input)) : NULL);
+ char *result = NULL;
+ size_t length = 0;
+ int retval = mem_cd_iconveh (input, strlen (input),
+ (indirect
+ ? &cdeh_88592_to_88591_indirectly
+ : &cdeh_88592_to_88591),
+ 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);
}
- 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)
+ break;
+ case iconveh_escape_sequence:
{
- for (i = 0; i < 16; i++)
- ASSERT (offsets[i] == (i < 5 ? i :
- i + 5));
- ASSERT (offsets[16] == MAGIC);
- free (offsets);
+ 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);
}
- free (result);
- }
- break;
+ break;
+ }
}
}
}
char *result = NULL;
size_t length = 0;
int retval = mem_cd_iconveh (input, strlen (input),
- cd_88591_to_utf8,
- cd_88591_to_utf8, (iconv_t)(-1),
+ &cdeh_88591_to_utf8,
handler,
offsets,
&result, &length);
char *result = NULL;
size_t length = 0;
int retval = mem_cd_iconveh (input, strlen (input),
- cd_utf8_to_88591,
- (iconv_t)(-1), cd_utf8_to_88591,
+ &cdeh_utf8_to_88591,
handler,
offsets,
&result, &length);
char *result = NULL;
size_t length = 0;
int retval = mem_cd_iconveh (input, strlen (input),
- cd_utf8_to_88591,
- (iconv_t)(-1), cd_utf8_to_88591,
+ &cdeh_utf8_to_88591,
handler,
offsets,
&result, &length);
char *result = NULL;
size_t length = 0;
int retval = mem_cd_iconveh (input, strlen (input),
- cd_utf8_to_88591,
- (iconv_t)(-1), cd_utf8_to_88591,
+ &cdeh_utf8_to_88591,
handler,
offsets,
&result, &length);
}
}
+ if (cd_utf7_to_utf8 != (iconv_t)(-1))
+ {
+ /* Disabled on Solaris, because Solaris 9 iconv() is buggy: it returns
+ -1 / EILSEQ when converting the 7th byte of the input "+VDLYP9hA". */
+# if !(defined __sun && !defined _LIBICONV_VERSION)
+ /* Test conversion from UTF-7 to UTF-8 with EINVAL. */
+ for (h = 0; h < SIZEOF (handlers); h++)
+ {
+ enum iconv_ilseq_handler handler = handlers[h];
+ /* This is base64 encoded 0x54 0x32 0xD8 0x3F 0xD8 0x40. It would
+ convert to U+5432 U+D83F U+D840 but these are Unicode surrogates. */
+ static const char input[] = "+VDLYP9hA";
+ static const char expected1[] = "\345\220\262"; /* 吲 glibc */
+ static const char expected2[] = ""; /* libiconv */
+ char *result = NULL;
+ size_t length = 0;
+ int retval = mem_cd_iconveh (input, 7,
+ &cdeh_utf7_to_utf8,
+ handler,
+ NULL,
+ &result, &length);
+ ASSERT (retval == 0);
+ ASSERT (length == strlen (expected1) || length == strlen (expected2));
+ ASSERT (result != NULL);
+ if (length == strlen (expected1))
+ ASSERT (memcmp (result, expected1, strlen (expected1)) == 0);
+ else
+ ASSERT (memcmp (result, expected2, strlen (expected2)) == 0);
+ free (result);
+ }
+
+ /* Test conversion from UTF-7 to UTF-8 with EILSEQ. */
+ for (h = 0; h < SIZEOF (handlers); h++)
+ {
+ enum iconv_ilseq_handler handler = handlers[h];
+ /* This is base64 encoded 0xD8 0x3F 0xD8 0x40 0xD8 0x41. It would
+ convert to U+D83F U+D840 U+D841 but these are Unicode surrogates. */
+ static const char input[] = "+2D/YQNhB";
+ char *result = NULL;
+ size_t length = 0;
+ int retval = mem_cd_iconveh (input, strlen (input),
+ &cdeh_utf7_to_utf8,
+ handler,
+ NULL,
+ &result, &length);
+ switch (handler)
+ {
+ case iconveh_error:
+ ASSERT (retval == -1 && errno == EILSEQ);
+ ASSERT (result == NULL);
+ break;
+ case iconveh_question_mark:
+ case iconveh_escape_sequence:
+ {
+ /* glibc result */
+ static const char expected1[] = "?????";
+ /* libiconv <= 1.12 result */
+ static const char expected2[] = "?2D/YQNhB";
+ /* libiconv behaviour changed in version 1.13: the result is
+ '?' U+0FF6 U+1036; this is U+D83F U+D840 U+D841 shifted left
+ by 6 bits. */
+ static const char expected3[] = "?\340\277\266\341\200\266";
+ ASSERT (retval == 0);
+ ASSERT (length == strlen (expected1)
+ || length == strlen (expected2)
+ || length == strlen (expected3));
+ ASSERT (result != NULL);
+ if (length == strlen (expected1))
+ ASSERT (memcmp (result, expected1, strlen (expected1)) == 0);
+ else if (length == strlen (expected2))
+ ASSERT (memcmp (result, expected2, strlen (expected2)) == 0);
+ else
+ ASSERT (memcmp (result, expected3, strlen (expected3)) == 0);
+ free (result);
+ }
+ break;
+ }
+ }
+# endif
+ }
+
/* ------------------------ Test str_cd_iconveh() ------------------------ */
/* Test conversion from ISO-8859-2 to ISO-8859-1 with no errors. */
- for (h = 0; h < SIZEOF (handlers); h++)
+ for (indirect = 0; indirect <= 1; indirect++)
{
- 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_cd_iconveh (input,
- cd_88592_to_88591,
- cd_88592_to_utf8, cd_utf8_to_88591,
- 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[] = "\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_cd_iconveh (input,
+ (indirect
+ ? &cdeh_88592_to_88591_indirectly
+ : &cdeh_88592_to_88591),
+ 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++)
+ for (indirect = 0; indirect <= 1; indirect++)
{
- enum iconv_ilseq_handler handler = handlers[h];
- static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */
- char *result = str_cd_iconveh (input,
- cd_88592_to_88591,
- cd_88592_to_utf8, cd_utf8_to_88591,
- handler);
- switch (handler)
+ for (h = 0; h < SIZEOF (handlers); h++)
{
- 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;
+ enum iconv_ilseq_handler handler = handlers[h];
+ static const char input[] = "Rafa\263 Maszkowski"; /* Rafał Maszkowski */
+ char *result = str_cd_iconveh (input,
+ (indirect
+ ? &cdeh_88592_to_88591_indirectly
+ : &cdeh_88592_to_88591),
+ 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;
+ }
}
}
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_cd_iconveh (input,
- cd_88591_to_utf8,
- cd_88591_to_utf8, (iconv_t)(-1),
+ &cdeh_88591_to_utf8,
handler);
ASSERT (result != NULL);
ASSERT (strcmp (result, expected) == 0);
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_cd_iconveh (input,
- cd_utf8_to_88591,
- (iconv_t)(-1), cd_utf8_to_88591,
+ &cdeh_utf8_to_88591,
handler);
ASSERT (result != NULL);
ASSERT (strcmp (result, expected) == 0);
enum iconv_ilseq_handler handler = handlers[h];
static const char input[] = "Costs: 27 \342\202\254"; /* EURO SIGN */
char *result = str_cd_iconveh (input,
- cd_utf8_to_88591,
- (iconv_t)(-1), cd_utf8_to_88591,
+ &cdeh_utf8_to_88591,
handler);
switch (handler)
{
enum iconv_ilseq_handler handler = handlers[h];
static const char input[] = "\342";
char *result = str_cd_iconveh (input,
- cd_utf8_to_88591,
- (iconv_t)(-1), cd_utf8_to_88591,
+ &cdeh_utf8_to_88591,
handler);
ASSERT (result != NULL);
ASSERT (strcmp (result, "") == 0);
projects / gnulib.git / blobdiff