dimmus
/
efl
forked from enlightenment/efl
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Evas encoding: Removed Evas_Encoding and moved to eina_unicode_utf8* instead. 

SVN revision: 57096
This commit is contained in:
Tom Hacohen 2011-02-16 16:00:17 +00:00
parent 0f686bb04b
commit bd0695d1b5
8 changed files with 9 additions and 302 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
legacy/evas/src/lib/canvas/evas_object_text.c
View File

@ -678,7 +678,7 @@ evas_object_text_text_set(Evas_Object *obj, const char *_text)
{
return;
}
text = evas_common_encoding_utf8_to_unicode(_text, &len);
text = eina_unicode_utf8_to_unicode(_text, &len);
if (!text) text = eina_unicode_strdup(EINA_UNICODE_EMPTY_STRING);
was = evas_object_is_in_output_rect(obj,
@ -1668,7 +1668,7 @@ evas_string_char_next_get(const char *str, int pos, int *decoded)
if (decoded) *decoded = 0;
if ((!str) || (pos < 0)) return 0;
p = pos;
d = evas_common_encoding_utf8_get_next(str, &p);
d = eina_unicode_utf8_get_next(str, &p);
if (decoded) *decoded = d;
return p;
}
@ -1700,7 +1700,7 @@ evas_string_char_prev_get(const char *str, int pos, int *decoded)
if (decoded) *decoded = 0;
if ((!str) || (pos < 1)) return 0;
p = pos;
d = evas_common_encoding_utf8_get_prev(str, &p);
d = eina_unicode_utf8_get_prev(str, &p);
if (decoded) *decoded = d;
return p;
}
@ -1715,7 +1715,7 @@ EAPI int
evas_string_char_len_get(const char *str)
{
if (!str) return 0;
return evas_common_encoding_utf8_get_len(str);
return eina_unicode_utf8_get_len(str);
}
/**
@ -2298,7 +2298,7 @@ _evas_object_text_recalc(Evas_Object *obj)
if (o->items) _evas_object_text_items_clear(o);
if (o->cur.utf8_text)
text = evas_common_encoding_utf8_to_unicode(o->cur.utf8_text,
text = eina_unicode_utf8_to_unicode(o->cur.utf8_text,
NULL);
if (!text) text = eina_unicode_strdup(EINA_UNICODE_EMPTY_STRING);

8
legacy/evas/src/lib/canvas/evas_object_textblock.c
View File

@ -2744,7 +2744,7 @@ _layout_text_append(Ctxt *c, Evas_Object_Textblock_Format *fmt, Evas_Object_Text
str = alloca((off + 1) * sizeof(Eina_Unicode));
tbase = str;
ind = 0;
urepch = evas_common_encoding_utf8_get_next(repch, &ind);
urepch = eina_unicode_utf8_get_next(repch, &ind);
for (i = 0, ptr = (Eina_Unicode *)tbase; i < off; ptr++, i++)
*ptr = urepch;
*ptr = 0;
@ -4563,7 +4563,7 @@ _markup_get_format_append(Evas_Object_Textblock *o, Eina_Strbuf *txt, Evas_Objec
static void
_markup_get_text_append(Eina_Strbuf *txt, const Eina_Unicode *text)
{
char *p = evas_common_encoding_unicode_to_utf8(text, NULL);
char *p = eina_unicode_unicode_to_utf8(text, NULL);
char *base = p;
while (*p)
{
@ -6391,7 +6391,7 @@ evas_textblock_cursor_text_append(Evas_Textblock_Cursor *cur, const char *_text)
int len = 0;
if (!cur) return 0;
text = evas_common_encoding_utf8_to_unicode(_text, &len);
text = eina_unicode_utf8_to_unicode(_text, &len);
o = (Evas_Object_Textblock *)(cur->obj->object_data);
/* Update all the cursors after our position. */
_evas_textblock_cursors_update_offset(cur, cur->node, cur->pos, len);
@ -6885,7 +6885,7 @@ evas_textblock_cursor_content_get(const Evas_Textblock_Cursor *cur)
ustr = eina_ustrbuf_string_get(cur->node->unicode);
buf[0] = ustr[cur->pos];
buf[1] = 0;
s = evas_common_encoding_unicode_to_utf8(buf, NULL);
s = eina_unicode_unicode_to_utf8(buf, NULL);
return s;
}

2
legacy/evas/src/lib/engines/common/Makefile.am
View File

@ -39,7 +39,6 @@ evas_convert_grypal_6.c \
evas_convert_yuv.c \
evas_cpu.c \
evas_draw_main.c \
evas_encoding.c \
evas_font_draw.c \
evas_font_load.c \
evas_font_main.c \
@ -82,7 +81,6 @@ evas_convert_rgb_32.h \
evas_convert_rgb_8.h \
evas_convert_yuv.h \
evas_draw.h \
evas_encoding.h \
evas_font.h \
evas_font_private.h \
evas_image.h \

259
legacy/evas/src/lib/engines/common/evas_encoding.c
View File

@ -1,259 +0,0 @@
#include "evas_common.h"
#include "evas_encoding.h"
/* The replacement range that will be used for bad utf8 chars. */
#define ERROR_REPLACEMENT_BASE 0xDC80
#define ERROR_REPLACEMENT_END 0xDCFF
#define IS_INVALID_BYTE(x) ((x == 192) || (x == 193) || (x >= 245))
#define IS_CONTINUATION_BYTE(x) ((x & 0xC0) == 0x80)
EAPI Eina_Unicode
evas_common_encoding_utf8_get_next(const char *buf, int *iindex)
{
/* Reads UTF8 bytes from @buf, starting at *@index and returns
* the decoded code point at iindex offset, and advances iindex
* to the next code point after this.
*
* Returns 0 to indicate there is no next char
*/
/* Note: we don't currently handle overlong forms and some other
* broken cases. */
int index = *iindex;
Eina_Unicode r;
unsigned char d;
/* if this char is the null terminator, exit */
if ((d = buf[index++]) == 0) return 0;
if ((d & 0x80) == 0)
{ // 1 byte (7bit) - 0xxxxxxx
*iindex = index;
return d;
}
if ((d & 0xe0) == 0xc0)
{ // 2 byte (11bit) - 110xxxxx 10xxxxxx
r = (d & 0x1f) << 6;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f);
if (!r) goto error;
*iindex = index;
return r;
}
if ((d & 0xf0) == 0xe0)
{ // 3 byte (16bit) - 1110xxxx 10xxxxxx 10xxxxxx
r = (d & 0x0f) << 12;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 6;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f);
if (!r) goto error;
*iindex = index;
return r;
}
if ((d & 0xf8) == 0xf0)
{ // 4 byte (21bit) - 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx
r = (d & 0x07) << 18;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 12;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 6;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f);
if (!r) goto error;
*iindex = index;
return r;
}
if ((d & 0xfc) == 0xf8)
{ // 5 byte (26bit) - 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
r = (d & 0x03) << 24;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 18;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 12;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 6;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f);
if (!r) goto error;
*iindex = index;
return r;
}
if ((d & 0xfe) == 0xfc)
{ // 6 byte (31bit) - 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
r = (d & 0x01) << 30;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 24;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 18;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 12;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f) << 6;
if (((d = buf[index++]) == 0) || IS_INVALID_BYTE(d) ||
!IS_CONTINUATION_BYTE(d)) goto error;
r |= (d & 0x3f);
if (!r) goto error;
*iindex = index;
return r;
}
/* Gets here where there was an error and we want to replace the char
* we just use the invalid unicode codepoints 8 lower bits represent
* the original char */
error:
d = buf[*iindex];
(*iindex)++;
return ERROR_REPLACEMENT_BASE | d;
}
EAPI Eina_Unicode
evas_common_encoding_utf8_get_prev(const char *buf, int *iindex)
{
/* Reads UTF8 bytes from @buf, starting at *@index and returns
* the decoded code point at iindex offset, and advances iindex
* to the prev code point after this.
*
* Returns 0 to indicate there is no prev char
*/
int r;
int index = *iindex;
/* First obtain the codepoint at iindex */
r = evas_common_encoding_utf8_get_next(buf, &index);
/* although when index == 0 there's no previous char, we still want to get
* the current char */
if (*iindex < 0)
return r;
/* Next advance iindex to previous codepoint */
index = *iindex;
index--;
while ((index > 0) && ((buf[index] & 0xc0) == 0x80))
index--;
*iindex = index;
return r;
}
EAPI Eina_Unicode
evas_common_encoding_utf8_get_last(const char *buf, int buflen)
{
/* jumps to the nul byte at the buffer end and decodes backwards and
* returns the offset index byte in the buffer where the last character
* in the buffer begins.
*
* Returns -1 to indicate an error
*/
/* Go one character backwards and then return the char at the new place */
evas_common_encoding_utf8_get_prev(buf, &buflen);
return evas_common_encoding_utf8_get_next(buf, &buflen);
}
EAPI int
evas_common_encoding_utf8_get_len(const char *buf)
{
/* returns the number of utf8 characters (not bytes) in the string */
int i = 0, len = 0;
while (evas_common_encoding_utf8_get_next(buf, &i))
len++;
return len;
}
/* FIXME: Should optimize! */
EAPI Eina_Unicode *
evas_common_encoding_utf8_to_unicode(const char *utf, int *_len)
{
int len, i;
int index;
Eina_Unicode *buf, *ind;
len = evas_common_encoding_utf8_get_len(utf);
if (_len)
*_len = len;
buf = (Eina_Unicode *) calloc(sizeof(Eina_Unicode), (len + 1));
if (!buf) return buf;
for (i = 0, index = 0, ind = buf ; i < len ; i++, ind++)
{
*ind = evas_common_encoding_utf8_get_next(utf, &index);
}
return buf;
}
EAPI char *
evas_common_encoding_unicode_to_utf8(const Eina_Unicode *uni, int *_len)
{
char *buf;
const Eina_Unicode *uind;
char *ind;
int ulen, len;
ulen = eina_unicode_strlen(uni);
buf = (char *) calloc(ulen + 1, EVAS_ENCODING_UTF8_BYTES_PER_CHAR);
len = 0;
for (uind = uni, ind = buf ; *uind ; uind++)
{
if (*uind <= 0x7F) /* 1 byte char */
{
*ind++ = *uind;
len += 1;
}
else if (*uind <= 0x7FF) /* 2 byte char */
{
*ind++ = 0xC0 | (unsigned char) (*uind >> 6);
*ind++ = 0x80 | (unsigned char) (*uind & 0x3F);
len += 2;
}
else if (*uind <= 0xFFFF) /* 3 byte char */
{
/* If it's a special replacement codepoint */
if (*uind >= ERROR_REPLACEMENT_BASE &&
*uind <= ERROR_REPLACEMENT_END)
{
*ind++ = *uind & 0xFF;
len += 1;
}
else
{
*ind++ = 0xE0 | (unsigned char) (*uind >> 12);
*ind++ = 0x80 | (unsigned char) ((*uind >> 6) & 0x3F);
*ind++ = 0x80 | (unsigned char) (*uind & 0x3F);
len += 3;
}
}
else /* 4 byte char */
{
*ind++ = 0xF0 | (unsigned char) ((*uind >> 18) & 0x07);
*ind++ = 0x80 | (unsigned char) ((*uind >> 12) & 0x3F);
*ind++ = 0x80 | (unsigned char) ((*uind >> 6) & 0x3F);
*ind++ = 0x80 | (unsigned char) (*uind & 0x3F);
len += 4;
}
}
buf = realloc(buf, len + 1);
buf[len] = '\0';
if (_len)
*_len = len;
return buf;
}

26
legacy/evas/src/lib/engines/common/evas_encoding.h
View File

@ -1,26 +0,0 @@
#ifndef EVAS_ENCODING_H
#define EVAS_ENCODING_H
#include <Eina.h>
/* FIXME: An assumption that will probably break in the future */
#define EVAS_ENCODING_UTF8_BYTES_PER_CHAR 4
EAPI Eina_Unicode
evas_common_encoding_utf8_get_next(const char *buf, int *iindex);
EAPI Eina_Unicode
evas_common_encoding_utf8_get_prev(const char *buf, int *iindex);
EAPI Eina_Unicode
evas_common_encoding_utf8_get_last(const char *buf, int buflen);
EAPI int
evas_common_encoding_utf8_get_len(const char *buf);
EAPI Eina_Unicode *
evas_common_encoding_utf8_to_unicode(const char *utf, int *_len) EINA_WARN_UNUSED_RESULT EINA_ARG_NONNULL(1) EINA_MALLOC;
EAPI char *
evas_common_encoding_unicode_to_utf8(const Eina_Unicode *uni, int *_len) EINA_WARN_UNUSED_RESULT EINA_ARG_NONNULL(1) EINA_MALLOC;
#endif

1
legacy/evas/src/lib/engines/common/evas_font_main.c
View File

@ -1,6 +1,5 @@
#include "evas_common.h"
#include "evas_private.h"
#include "evas_encoding.h"
#include "evas_font_private.h"

1
legacy/evas/src/lib/engines/common/language/evas_bidi_utils.c
View File

@ -3,7 +3,6 @@
#include "evas_common.h"
#include "evas_bidi_utils.h"
#include "evas_encoding.h"
#include "evas_font_private.h"

4
legacy/evas/src/lib/include/evas_common.h
View File

@ -49,10 +49,6 @@ extern EAPI int _evas_log_dom_global;
#endif
#define CRIT(...) EINA_LOG_DOM_CRIT(_EVAS_DEFAULT_LOG_DOM, __VA_ARGS__)
/************************ Unicode stuff **************************/
#include "../engines/common/evas_encoding.h"
/*****************************************************************/
#include "evas_options.h"
#if defined(__ARM_ARCH_3M__)