#include <string.h>
#include "glib.h"
+#include "galias.h"
#ifdef G_PLATFORM_WIN32
#include <stdio.h>
* Converts from an integer character offset to a pointer to a position
* within the string.
*
+ * Since 2.10, this function allows to pass a negative @offset to
+ * step backwards. It is usually worth stepping backwards from the end
+ * instead of forwards if @offset is in the last fourth of the string,
+ * since moving forward is about 3 times faster than moving backward.
+ *
* Return value: the resulting pointer
**/
gchar *
glong offset)
{
const gchar *s = str;
- while (offset--)
- s = g_utf8_next_char (s);
-
+
+ if (offset > 0)
+ while (offset--)
+ s = g_utf8_next_char (s);
+ else
+ {
+ const char *s1;
+
+ /* This nice technique for fast backwards stepping
+ * through a UTF-8 string was dubbed "stutter stepping"
+ * by its inventor, Larry Ewing.
+ */
+ while (offset)
+ {
+ s1 = s;
+ s += offset;
+ while ((*s & 0xc0) == 0x80)
+ s--;
+
+ offset += g_utf8_pointer_to_offset (s, s1);
+ }
+ }
+
return (gchar *)s;
}
*
* Converts from a pointer to position within a string to a integer
* character offset.
+ *
+ * Since 2.10, this function allows @pos to be before @str, and returns
+ * a negative offset in this case.
*
* Return value: the resulting character offset
**/
{
const gchar *s = str;
glong offset = 0;
-
- while (s < pos)
- {
- s = g_utf8_next_char (s);
- offset++;
- }
+ if (pos < str)
+ offset = - g_utf8_pointer_to_offset (pos, str);
+ else
+ while (s < pos)
+ {
+ s = g_utf8_next_char (s);
+ offset++;
+ }
+
return offset;
}
* @src: UTF-8 encoded string
* @n: character count
*
- * Like the standard C <function>strncpy()</function> function, but
+ * Like the standard C strncpy() function, but
* copies a given number of characters instead of a given number of
* bytes. The @src string must be valid UTF-8 encoded text.
* (Use g_utf8_validate() on all text before trying to use UTF-8
/**
* g_unichar_to_utf8:
- * @c: a ISO10646 character code
+ * @c: a Unicode character code
* @outbuf: output buffer, must have at least 6 bytes of space.
* If %NULL, the length will be computed and returned
* and nothing will be written to @outbuf.
g_unichar_to_utf8 (gunichar c,
gchar *outbuf)
{
+ /* If this gets modified, also update the copy in g_string_insert_unichar() */
guint len = 0;
int first;
int i;
* g_utf8_strchr:
* @p: a nul-terminated UTF-8 encoded string
* @len: the maximum length of @p
- * @c: a ISO10646 character
+ * @c: a Unicode character
*
- * Finds the leftmost occurrence of the given ISO10646 character
+ * Finds the leftmost occurrence of the given Unicode character
* in a UTF-8 encoded string, while limiting the search to @len bytes.
* If @len is -1, allow unbounded search.
*
* g_utf8_strrchr:
* @p: a nul-terminated UTF-8 encoded string
* @len: the maximum length of @p
- * @c: a ISO10646 character
+ * @c: a Unicode character
*
- * Find the rightmost occurrence of the given ISO10646 character
+ * Find the rightmost occurrence of the given Unicode character
* in a UTF-8 encoded string, while limiting the search to @len bytes.
* If @len is -1, allow unbounded search.
*
* overlong encodings of valid characters.
*
* Return value: the resulting character. If @p points to a partial
- * sequence at the end of a string that could begin a valid character,
- * returns (gunichar)-2; otherwise, if @p does not point to a valid
- * UTF-8 encoded Unicode character, returns (gunichar)-1.
+ * sequence at the end of a string that could begin a valid
+ * character, returns (gunichar)-2; otherwise, if @p does not point
+ * to a valid UTF-8 encoded Unicode character, returns (gunichar)-1.
**/
gunichar
g_utf8_get_char_validated (const gchar *p,
n_chars = 0;
while ((len < 0 || str + len - in > 0) && *in)
{
- gunichar wc = g_utf8_get_char_extended (in, str + len - in);
+ gunichar wc = g_utf8_get_char_extended (in, len < 0 ? 6 : str + len - in);
if (wc & 0x80000000)
{
if (wc == (gunichar)-2)
/**
* g_ucs4_to_utf8:
* @str: a UCS-4 encoded string
- * @len: the maximum length of @str to use. If @len < 0, then
- * the string is terminated with a 0 character.
- * @items_read: location to store number of characters read read, or %NULL.
+ * @len: the maximum length (number of characters) of @str to use.
+ * If @len < 0, then the string is terminated with a 0 character.
+ * @items_read: location to store number of characters read, or %NULL.
* @items_written: location to store number of bytes written or %NULL.
* The value here stored does not include the trailing 0
* byte.
* Return value: a pointer to a newly allocated UTF-8 string.
* This value must be freed with g_free(). If an
* error occurs, %NULL will be returned and
- * @error set.
+ * @error set. In that case, @items_read will be
+ * set to the position of the first invalid input
+ * character.
**/
gchar *
g_ucs4_to_utf8 (const gunichar *str,
if (str[i] >= 0x80000000)
{
- if (items_read)
- *items_read = i;
-
g_set_error (error, G_CONVERT_ERROR, G_CONVERT_ERROR_ILLEGAL_SEQUENCE,
_("Character out of range for UTF-8"));
goto err_out;
/**
* g_utf16_to_utf8:
* @str: a UTF-16 encoded string
- * @len: the maximum length of @str to use. If @len < 0, then
- * the string is terminated with a 0 character.
+ * @len: the maximum length (number of <type>gunichar2</type>) of @str to use.
+ * If @len < 0, then the string is terminated with a 0 character.
* @items_read: location to store number of words read, or %NULL.
* If %NULL, then %G_CONVERT_ERROR_PARTIAL_INPUT will be
* returned in case @str contains a trailing partial
*
* Convert a string from UTF-16 to UTF-8. The result will be
* terminated with a 0 byte.
+ *
+ * Note that the input is expected to be already in native endianness,
+ * an initial byte-order-mark character is not handled specially.
+ * g_convert() can be used to convert a byte buffer of UTF-16 data of
+ * ambiguous endianess.
*
* Return value: a pointer to a newly allocated UTF-8 string.
* This value must be freed with g_free(). If an
/**
* g_utf16_to_ucs4:
* @str: a UTF-16 encoded string
- * @len: the maximum length of @str to use. If @len < 0, then
- * the string is terminated with a 0 character.
+ * @len: the maximum length (number of <type>gunichar2</type>) of @str to use.
+ * If @len < 0, then the string is terminated with a 0 character.
* @items_read: location to store number of words read, or %NULL.
* If %NULL, then %G_CONVERT_ERROR_PARTIAL_INPUT will be
* returned in case @str contains a trailing partial
/**
* g_utf8_to_utf16:
* @str: a UTF-8 encoded string
- * @len: the maximum length of @str to use. If @len < 0, then
- * the string is nul-terminated.
+ * @len: the maximum length (number of characters) of @str to use.
+ * If @len < 0, then the string is nul-terminated.
* @items_read: location to store number of bytes read, or %NULL.
* If %NULL, then %G_CONVERT_ERROR_PARTIAL_INPUT will be
* returned in case @str contains a trailing partial
* character. If an error occurs then the index of the
* invalid input is stored here.
- * @items_written: location to store number of words written, or %NULL.
- * The value stored here does not include the trailing
- * 0 word.
+ * @items_written: location to store number of <type>gunichar2</type> written,
+ * or %NULL.
+ * The value stored here does not include the trailing 0.
* @error: location to store the error occuring, or %NULL to ignore
* errors. Any of the errors in #GConvertError other than
* %G_CONVERT_ERROR_NO_CONVERSION may occur.
*
- * Convert a string from UTF-8 to UTF-16. A 0 word will be
+ * Convert a string from UTF-8 to UTF-16. A 0 character will be
* added to the result after the converted text.
*
* Return value: a pointer to a newly allocated UTF-16 string.
n16 = 0;
while ((len < 0 || str + len - in > 0) && *in)
{
- gunichar wc = g_utf8_get_char_extended (in, str + len - in);
+ gunichar wc = g_utf8_get_char_extended (in, len < 0 ? 6 : str + len - in);
if (wc & 0x80000000)
{
if (wc == (gunichar)-2)
/**
* g_ucs4_to_utf16:
* @str: a UCS-4 encoded string
- * @len: the maximum length of @str to use. If @len < 0, then
- * the string is terminated with a 0 character.
+ * @len: the maximum length (number of characters) of @str to use.
+ * If @len < 0, then the string is terminated with a 0 character.
* @items_read: location to store number of bytes read, or %NULL.
* If an error occurs then the index of the invalid input
* is stored here.
- * @items_written: location to store number of words written, or %NULL.
- * The value stored here does not include the trailing
- * 0 word.
+ * @items_written: location to store number of <type>gunichar2</type>
+ * written, or %NULL. The value stored here does not
+ * include the trailing 0.
* @error: location to store the error occuring, or %NULL to ignore
* errors. Any of the errors in #GConvertError other than
* %G_CONVERT_ERROR_NO_CONVERSION may occur.
*
- * Convert a string from UCS-4 to UTF-16. A 0 word will be
+ * Convert a string from UCS-4 to UTF-16. A 0 character will be
* added to the result after the converted text.
*
* Return value: a pointer to a newly allocated UTF-16 string.
return result;
}
+#define CONTINUATION_CHAR \
+ G_STMT_START { \
+ if ((*(guchar *)p & 0xc0) != 0x80) /* 10xxxxxx */ \
+ goto error; \
+ val <<= 6; \
+ val |= (*(guchar *)p) & 0x3f; \
+ } G_STMT_END
+
+static const gchar *
+fast_validate (const char *str)
+
+{
+ gunichar val = 0;
+ gunichar min = 0;
+ const gchar *p;
+
+ for (p = str; *p; p++)
+ {
+ if (*(guchar *)p < 128)
+ /* done */;
+ else
+ {
+ const gchar *last;
+
+ last = p;
+ if ((*(guchar *)p & 0xe0) == 0xc0) /* 110xxxxx */
+ {
+ if (G_UNLIKELY ((*(guchar *)p & 0x1e) == 0))
+ goto error;
+ p++;
+ if (G_UNLIKELY ((*(guchar *)p & 0xc0) != 0x80)) /* 10xxxxxx */
+ goto error;
+ }
+ else
+ {
+ if ((*(guchar *)p & 0xf0) == 0xe0) /* 1110xxxx */
+ {
+ min = (1 << 11);
+ val = *(guchar *)p & 0x0f;
+ goto TWO_REMAINING;
+ }
+ else if ((*(guchar *)p & 0xf8) == 0xf0) /* 11110xxx */
+ {
+ min = (1 << 16);
+ val = *(guchar *)p & 0x07;
+ }
+ else
+ goto error;
+
+ p++;
+ CONTINUATION_CHAR;
+ TWO_REMAINING:
+ p++;
+ CONTINUATION_CHAR;
+ p++;
+ CONTINUATION_CHAR;
+
+ if (G_UNLIKELY (val < min))
+ goto error;
+
+ if (G_UNLIKELY (!UNICODE_VALID(val)))
+ goto error;
+ }
+
+ continue;
+
+ error:
+ return last;
+ }
+ }
+
+ return p;
+}
+
+static const gchar *
+fast_validate_len (const char *str,
+ gssize max_len)
+
+{
+ gunichar val = 0;
+ gunichar min = 0;
+ const gchar *p;
+
+ for (p = str; (max_len < 0 || (p - str) < max_len) && *p; p++)
+ {
+ if (*(guchar *)p < 128)
+ /* done */;
+ else
+ {
+ const gchar *last;
+
+ last = p;
+ if ((*(guchar *)p & 0xe0) == 0xc0) /* 110xxxxx */
+ {
+ if (G_UNLIKELY (max_len >= 0 && max_len - (p - str) < 2))
+ goto error;
+
+ if (G_UNLIKELY ((*(guchar *)p & 0x1e) == 0))
+ goto error;
+ p++;
+ if (G_UNLIKELY ((*(guchar *)p & 0xc0) != 0x80)) /* 10xxxxxx */
+ goto error;
+ }
+ else
+ {
+ if ((*(guchar *)p & 0xf0) == 0xe0) /* 1110xxxx */
+ {
+ if (G_UNLIKELY (max_len >= 0 && max_len - (p - str) < 3))
+ goto error;
+
+ min = (1 << 11);
+ val = *(guchar *)p & 0x0f;
+ goto TWO_REMAINING;
+ }
+ else if ((*(guchar *)p & 0xf8) == 0xf0) /* 11110xxx */
+ {
+ if (G_UNLIKELY (max_len >= 0 && max_len - (p - str) < 4))
+ goto error;
+
+ min = (1 << 16);
+ val = *(guchar *)p & 0x07;
+ }
+ else
+ goto error;
+
+ p++;
+ CONTINUATION_CHAR;
+ TWO_REMAINING:
+ p++;
+ CONTINUATION_CHAR;
+ p++;
+ CONTINUATION_CHAR;
+
+ if (G_UNLIKELY (val < min))
+ goto error;
+ if (G_UNLIKELY (!UNICODE_VALID(val)))
+ goto error;
+ }
+
+ continue;
+
+ error:
+ return last;
+ }
+ }
+
+ return p;
+}
+
/**
* g_utf8_validate:
* @str: a pointer to character data
- * @max_len: max bytes to validate, or -1 to go until nul
+ * @max_len: max bytes to validate, or -1 to go until NUL
* @end: return location for end of valid data
*
* Validates UTF-8 encoded text. @str is the text to validate;
* if @str is nul-terminated, then @max_len can be -1, otherwise
* @max_len should be the number of bytes to validate.
* If @end is non-%NULL, then the end of the valid range
- * will be stored there (i.e. the address of the first invalid byte
- * if some bytes were invalid, or the end of the text being validated
- * otherwise).
+ * will be stored there (i.e. the start of the first invalid
+ * character if some bytes were invalid, or the end of the text
+ * being validated otherwise).
+ *
+ * Note that g_utf8_validate() returns %FALSE if @max_len is
+ * positive and NUL is met before @max_len bytes have been read.
*
* Returns %TRUE if all of @str was valid. Many GLib and GTK+
* routines <emphasis>require</emphasis> valid UTF-8 as input;
* Return value: %TRUE if the text was valid UTF-8
**/
gboolean
-g_utf8_validate (const gchar *str,
- gssize max_len,
- const gchar **end)
-{
+g_utf8_validate (const char *str,
+ gssize max_len,
+ const gchar **end)
+{
const gchar *p;
- g_return_val_if_fail (str != NULL, FALSE);
-
- if (end)
- *end = str;
-
- p = str;
-
- while ((max_len < 0 || (p - str) < max_len) && *p)
- {
- int i, mask = 0, len;
- gunichar result;
- unsigned char c = (unsigned char) *p;
-
- UTF8_COMPUTE (c, mask, len);
-
- if (len == -1)
- break;
-
- /* check that the expected number of bytes exists in str */
- if (max_len >= 0 &&
- ((max_len - (p - str)) < len))
- break;
-
- UTF8_GET (result, p, i, mask, len);
-
- if (UTF8_LENGTH (result) != len) /* Check for overlong UTF-8 */
- break;
-
- if (result == (gunichar)-1)
- break;
-
- if (!UNICODE_VALID (result))
- break;
-
- p += len;
- }
+ if (max_len < 0)
+ p = fast_validate (str);
+ else
+ p = fast_validate_len (str, max_len);
if (end)
*end = p;
- /* See that we covered the entire length if a length was
- * passed in, or that we ended on a nul if not
- */
- if (max_len >= 0 &&
- p != (str + max_len))
- return FALSE;
- else if (max_len < 0 &&
- *p != '\0')
+ if ((max_len >= 0 && p != str + max_len) ||
+ (max_len < 0 && *p != '\0'))
return FALSE;
else
return TRUE;
}
+
/**
* g_unichar_validate:
* @ch: a Unicode character
return result;
}
+
+#define __G_UTF8_C__
+#include "galiasdef.c"