int
main ()
{
/* Test case n = 0. */
u8_width_linebreaks (NULL, 0, 80, 0, 0, NULL, "GB18030", NULL);
{
static const uint8_t input[91] =
/* "Grüß Gott. Здравствуйте! x=(-b±sqrt(b²-4ac))/(2a) 日本語,中文,한글" */
"Gr\303\274\303\237 Gott. \320\227\320\264\321\200\320\260\320\262\321\201\321\202\320\262\321\203\320\271\321\202\320\265! x=(-b\302\261sqrt(b\302\262-4ac))/(2a) \346\227\245\346\234\254\350\252\236,\344\270\255\346\226\207,\355\225\234\352\270\200\n";
{
char *p = (char *) malloc (SIZEOF (input));
size_t i;
u8_width_linebreaks (input, SIZEOF (input), 25, 0, 0, NULL, "GB18030", p);
for (i = 0; i < 91; i++)
{
ASSERT (p[i] == (i == 90 ? UC_BREAK_MANDATORY :
i == 39 || i == 61 ? UC_BREAK_POSSIBLE :
UC_BREAK_PROHIBITED));
}
free (p);
}
{
char *p = (char *) malloc (SIZEOF (input));
size_t i;
u8_width_linebreaks (input, SIZEOF (input), 25, 0, 0, NULL, "GB2312", p);
for (i = 0; i < 91; i++)
{
ASSERT (p[i] == (i == 90 ? UC_BREAK_MANDATORY :
i == 13 || i == 39 || i == 61 ? UC_BREAK_POSSIBLE :
UC_BREAK_PROHIBITED));
}
free (p);
}
}
return 0;
}
int
main (int argc, char * argv[])
{
if (argc == 2)
{
/* Insert line breaks for a given width. */
int width = atoi (argv[1]);
char *input = read_file (stdin);
int length = strlen (input);
char *breaks = malloc (length);
int i;
u8_width_linebreaks ((uint8_t *) input, length, width, 0, 0, NULL, "UTF-8", breaks);
for (i = 0; i < length; i++)
{
switch (breaks[i])
{
case UC_BREAK_POSSIBLE:
putc ('\n', stdout);
break;
case UC_BREAK_MANDATORY:
break;
case UC_BREAK_PROHIBITED:
break;
default:
abort ();
}
putc (input[i], stdout);
}
free (breaks);
return 0;
}
else
return 1;
}
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;
}