From ff076afaae6c9c736c54c34e10e931b939c7e403 Mon Sep 17 00:00:00 2001 From: Seven Du Date: Tue, 10 Feb 2015 12:21:04 +0800 Subject: [PATCH] FS-7506: utf8 support of draw text --- .../applications/mod_conference/Makefile.am | 6 +- .../mod_conference/mod_conference.c | 35 +- src/mod/applications/mod_conference/utf8.c | 477 ++++++++++++++++++ src/mod/applications/mod_conference/utf8.h | 71 +++ 4 files changed, 582 insertions(+), 7 deletions(-) create mode 100644 src/mod/applications/mod_conference/utf8.c create mode 100644 src/mod/applications/mod_conference/utf8.h diff --git a/src/mod/applications/mod_conference/Makefile.am b/src/mod/applications/mod_conference/Makefile.am index 17cc0f43aa..d0d9e072d6 100644 --- a/src/mod/applications/mod_conference/Makefile.am +++ b/src/mod/applications/mod_conference/Makefile.am @@ -2,10 +2,10 @@ include $(top_srcdir)/build/modmake.rulesam MODNAME=mod_conference mod_LTLIBRARIES = mod_conference.la -mod_conference_la_SOURCES = mod_conference.c -mod_conference_la_CFLAGS = $(AM_CFLAGS) +mod_conference_la_SOURCES = mod_conference.c utf8.c +mod_conference_la_CFLAGS = $(AM_CFLAGS) -I. -I/usr/include/freetype2 mod_conference_la_LIBADD = $(switch_builddir)/libfreeswitch.la -mod_conference_la_LDFLAGS = -avoid-version -module -no-undefined -shared -lyuv +mod_conference_la_LDFLAGS = -avoid-version -module -no-undefined -shared -lyuv -lfreetype if HAVE_OPENAL mod_conference_la_LDFLAGS += -lopenal -lm diff --git a/src/mod/applications/mod_conference/mod_conference.c b/src/mod/applications/mod_conference/mod_conference.c index 222b2268bf..27590df2ef 100644 --- a/src/mod/applications/mod_conference/mod_conference.c +++ b/src/mod/applications/mod_conference/mod_conference.c @@ -701,6 +701,31 @@ typedef struct layout_group_s { video_layout_node_t *layouts; } layout_group_t; + +#include "utf8.h" + +static const u_int32_t offsetsFromUTF8[6] = { + 0x00000000UL, 0x00003080UL, 0x000E2080UL, + 0x03C82080UL, 0xFA082080UL, 0x82082080UL +}; + +/* reads the next utf-8 sequence out of a string, updating an index */ +uint32_t get_utf8_char(char *s, int *i) +{ + u_int32_t ch = 0; + int sz = 0; + + do { + ch <<= 6; + ch += (unsigned char)s[(*i)++]; + sz++; + } while (s[*i] && !isutf(s[*i])); + + ch -= offsetsFromUTF8[sz-1]; + + return ch; +} + #include #include FT_FREETYPE_H #include FT_GLYPH_H @@ -735,13 +760,13 @@ SWITCH_DECLARE(void) switch_img_draw_text(switch_image_t *img, int x, int y, cha int font_size = 64; double angle; int target_height; - int n, num_chars; + int index = 0; + FT_ULong ch; switch_yuv_color_t color; if (zstr(text)) return; switch_color_set(&color, "#FFFFFF"); - num_chars = strlen(text); angle = 0; // (45.0 / 360 ) * 3.14159 * 2; target_height = img->d_h; @@ -768,12 +793,14 @@ SWITCH_DECLARE(void) switch_img_draw_text(switch_image_t *img, int x, int y, cha pen.x = x * 64; pen.y = (target_height - y) * 64; - for(n = 0; n < num_chars; n++) { + while(*(text + index)) { + ch = get_utf8_char(text, &index); + /* set transformation */ FT_Set_Transform(face, &matrix, &pen); /* load glyph image into the slot (erase previous one) */ - error = FT_Load_Char(face, text[n], FT_LOAD_RENDER); + error = FT_Load_Char(face, ch, FT_LOAD_RENDER); if (error) continue; /* now, draw to our target surface (convert position) */ diff --git a/src/mod/applications/mod_conference/utf8.c b/src/mod/applications/mod_conference/utf8.c new file mode 100644 index 0000000000..9dddb596c6 --- /dev/null +++ b/src/mod/applications/mod_conference/utf8.c @@ -0,0 +1,477 @@ +/* + Basic UTF-8 manipulation routines + by Jeff Bezanson + placed in the public domain Fall 2005 + + This code is designed to provide the utilities you need to manipulate + UTF-8 as an internal string encoding. These functions do not perform the + error checking normally needed when handling UTF-8 data, so if you happen + to be from the Unicode Consortium you will want to flay me alive. + I do this because error checking can be performed at the boundaries (I/O), + with these routines reserved for higher performance on data known to be + valid. +*/ +#include +#include +#include +#include +#ifdef WIN32 +#include +#else +#include +#include +#endif + +#include "utf8.h" + +static const uint32_t offsetsFromUTF8[6] = { + 0x00000000UL, 0x00003080UL, 0x000E2080UL, + 0x03C82080UL, 0xFA082080UL, 0x82082080UL +}; + +static const char trailingBytesForUTF8[256] = { + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5 +}; + +/* returns length of next utf-8 sequence */ +int u8_seqlen(char *s) +{ + return trailingBytesForUTF8[(unsigned int)(unsigned char)s[0]] + 1; +} + +/* conversions without error checking + only works for valid UTF-8, i.e. no 5- or 6-byte sequences + srcsz = source size in bytes, or -1 if 0-terminated + sz = dest size in # of wide characters + + returns # characters converted + dest will always be L'\0'-terminated, even if there isn't enough room + for all the characters. + if sz = srcsz+1 (i.e. 4*srcsz+4 bytes), there will always be enough space. +*/ +int u8_toucs(uint32_t *dest, int sz, char *src, int srcsz) +{ + uint32_t ch; + char *src_end = src + srcsz; + int nb; + int i=0; + + while (i < sz-1) { + nb = trailingBytesForUTF8[(unsigned char)*src]; + if (srcsz == -1) { + if (*src == 0) + goto done_toucs; + } + else { + if (src + nb >= src_end) + goto done_toucs; + } + ch = 0; + switch (nb) { + /* these fall through deliberately */ + case 3: ch += (unsigned char)*src++; ch <<= 6; + case 2: ch += (unsigned char)*src++; ch <<= 6; + case 1: ch += (unsigned char)*src++; ch <<= 6; + case 0: ch += (unsigned char)*src++; + } + ch -= offsetsFromUTF8[nb]; + dest[i++] = ch; + } + done_toucs: + dest[i] = 0; + return i; +} + +/* srcsz = number of source characters, or -1 if 0-terminated + sz = size of dest buffer in bytes + + returns # characters converted + dest will only be '\0'-terminated if there is enough space. this is + for consistency; imagine there are 2 bytes of space left, but the next + character requires 3 bytes. in this case we could NUL-terminate, but in + general we can't when there's insufficient space. therefore this function + only NUL-terminates if all the characters fit, and there's space for + the NUL as well. + the destination string will never be bigger than the source string. +*/ +int u8_toutf8(char *dest, int sz, uint32_t *src, int srcsz) +{ + uint32_t ch; + int i = 0; + char *dest_end = dest + sz; + + while (srcsz<0 ? src[i]!=0 : i < srcsz) { + ch = src[i]; + if (ch < 0x80) { + if (dest >= dest_end) + return i; + *dest++ = (char)ch; + } + else if (ch < 0x800) { + if (dest >= dest_end-1) + return i; + *dest++ = (ch>>6) | 0xC0; + *dest++ = (ch & 0x3F) | 0x80; + } + else if (ch < 0x10000) { + if (dest >= dest_end-2) + return i; + *dest++ = (ch>>12) | 0xE0; + *dest++ = ((ch>>6) & 0x3F) | 0x80; + *dest++ = (ch & 0x3F) | 0x80; + } + else if (ch < 0x110000) { + if (dest >= dest_end-3) + return i; + *dest++ = (ch>>18) | 0xF0; + *dest++ = ((ch>>12) & 0x3F) | 0x80; + *dest++ = ((ch>>6) & 0x3F) | 0x80; + *dest++ = (ch & 0x3F) | 0x80; + } + i++; + } + if (dest < dest_end) + *dest = '\0'; + return i; +} + +int u8_wc_toutf8(char *dest, uint32_t ch) +{ + if (ch < 0x80) { + dest[0] = (char)ch; + return 1; + } + if (ch < 0x800) { + dest[0] = (ch>>6) | 0xC0; + dest[1] = (ch & 0x3F) | 0x80; + return 2; + } + if (ch < 0x10000) { + dest[0] = (ch>>12) | 0xE0; + dest[1] = ((ch>>6) & 0x3F) | 0x80; + dest[2] = (ch & 0x3F) | 0x80; + return 3; + } + if (ch < 0x110000) { + dest[0] = (ch>>18) | 0xF0; + dest[1] = ((ch>>12) & 0x3F) | 0x80; + dest[2] = ((ch>>6) & 0x3F) | 0x80; + dest[3] = (ch & 0x3F) | 0x80; + return 4; + } + return 0; +} + +/* charnum => byte offset */ +int u8_offset(char *str, int charnum) +{ + int offs=0; + + while (charnum > 0 && str[offs]) { + (void)(isutf(str[++offs]) || isutf(str[++offs]) || + isutf(str[++offs]) || ++offs); + charnum--; + } + return offs; +} + +/* byte offset => charnum */ +int u8_charnum(char *s, int offset) +{ + int charnum = 0, offs=0; + + while (offs < offset && s[offs]) { + (void)(isutf(s[++offs]) || isutf(s[++offs]) || + isutf(s[++offs]) || ++offs); + charnum++; + } + return charnum; +} + +/* number of characters */ +int u8_strlen(char *s) +{ + int count = 0; + int i = 0; + + while (u8_nextchar(s, &i) != 0) + count++; + + return count; +} + +/* reads the next utf-8 sequence out of a string, updating an index */ +uint32_t u8_nextchar(char *s, int *i) +{ + uint32_t ch = 0; + int sz = 0; + + do { + ch <<= 6; + ch += (unsigned char)s[(*i)++]; + sz++; + } while (s[*i] && !isutf(s[*i])); + ch -= offsetsFromUTF8[sz-1]; + + return ch; +} + +void u8_inc(char *s, int *i) +{ + (void)(isutf(s[++(*i)]) || isutf(s[++(*i)]) || + isutf(s[++(*i)]) || ++(*i)); +} + +void u8_dec(char *s, int *i) +{ + (void)(isutf(s[--(*i)]) || isutf(s[--(*i)]) || + isutf(s[--(*i)]) || --(*i)); +} + +int octal_digit(char c) +{ + return (c >= '0' && c <= '7'); +} + +int hex_digit(char c) +{ + return ((c >= '0' && c <= '9') || + (c >= 'A' && c <= 'F') || + (c >= 'a' && c <= 'f')); +} + +/* assumes that src points to the character after a backslash + returns number of input characters processed */ +int u8_read_escape_sequence(char *str, uint32_t *dest) +{ + uint32_t ch; + char digs[9]="\0\0\0\0\0\0\0\0"; + int dno=0, i=1; + + ch = (uint32_t)str[0]; /* take literal character */ + if (str[0] == 'n') + ch = L'\n'; + else if (str[0] == 't') + ch = L'\t'; + else if (str[0] == 'r') + ch = L'\r'; + else if (str[0] == 'b') + ch = L'\b'; + else if (str[0] == 'f') + ch = L'\f'; + else if (str[0] == 'v') + ch = L'\v'; + else if (str[0] == 'a') + ch = L'\a'; + else if (octal_digit(str[0])) { + i = 0; + do { + digs[dno++] = str[i++]; + } while (octal_digit(str[i]) && dno < 3); + ch = strtol(digs, NULL, 8); + } + else if (str[0] == 'x') { + while (hex_digit(str[i]) && dno < 2) { + digs[dno++] = str[i++]; + } + if (dno > 0) + ch = strtol(digs, NULL, 16); + } + else if (str[0] == 'u') { + while (hex_digit(str[i]) && dno < 4) { + digs[dno++] = str[i++]; + } + if (dno > 0) + ch = strtol(digs, NULL, 16); + } + else if (str[0] == 'U') { + while (hex_digit(str[i]) && dno < 8) { + digs[dno++] = str[i++]; + } + if (dno > 0) + ch = strtol(digs, NULL, 16); + } + *dest = ch; + + return i; +} + +/* convert a string with literal \uxxxx or \Uxxxxxxxx characters to UTF-8 + example: u8_unescape(mybuf, 256, "hello\\u220e") + note the double backslash is needed if called on a C string literal */ +int u8_unescape(char *buf, int sz, char *src) +{ + int c=0, amt; + uint32_t ch; + char temp[4]; + + while (*src && c < sz) { + if (*src == '\\') { + src++; + amt = u8_read_escape_sequence(src, &ch); + } + else { + ch = (uint32_t)*src; + amt = 1; + } + src += amt; + amt = u8_wc_toutf8(temp, ch); + if (amt > sz-c) + break; + memcpy(&buf[c], temp, amt); + c += amt; + } + if (c < sz) + buf[c] = '\0'; + return c; +} + +int u8_escape_wchar(char *buf, int sz, uint32_t ch) +{ + if (ch == L'\n') + return snprintf(buf, sz, "\\n"); + else if (ch == L'\t') + return snprintf(buf, sz, "\\t"); + else if (ch == L'\r') + return snprintf(buf, sz, "\\r"); + else if (ch == L'\b') + return snprintf(buf, sz, "\\b"); + else if (ch == L'\f') + return snprintf(buf, sz, "\\f"); + else if (ch == L'\v') + return snprintf(buf, sz, "\\v"); + else if (ch == L'\a') + return snprintf(buf, sz, "\\a"); + else if (ch == L'\\') + return snprintf(buf, sz, "\\\\"); + else if (ch < 32 || ch == 0x7f) + return snprintf(buf, sz, "\\x%hhX", (unsigned char)ch); + else if (ch > 0xFFFF) + return snprintf(buf, sz, "\\U%.8X", (uint32_t)ch); + else if (ch >= 0x80 && ch <= 0xFFFF) + return snprintf(buf, sz, "\\u%.4hX", (unsigned short)ch); + + return snprintf(buf, sz, "%c", (char)ch); +} + +int u8_escape(char *buf, int sz, char *src, int escape_quotes) +{ + int c=0, i=0, amt; + + while (src[i] && c < sz) { + if (escape_quotes && src[i] == '"') { + amt = snprintf(buf, sz - c, "\\\""); + i++; + } + else { + amt = u8_escape_wchar(buf, sz - c, u8_nextchar(src, &i)); + } + c += amt; + buf += amt; + } + if (c < sz) + *buf = '\0'; + return c; +} + +char *u8_strchr(char *s, uint32_t ch, int *charn) +{ + int i = 0, lasti=0; + uint32_t c; + + *charn = 0; + while (s[i]) { + c = u8_nextchar(s, &i); + if (c == ch) { + return &s[lasti]; + } + lasti = i; + (*charn)++; + } + return NULL; +} + +char *u8_memchr(char *s, uint32_t ch, size_t sz, int *charn) +{ + int i = 0, lasti=0; + uint32_t c; + int csz; + + *charn = 0; + while (i < sz) { + c = csz = 0; + do { + c <<= 6; + c += (unsigned char)s[i++]; + csz++; + } while (i < sz && !isutf(s[i])); + c -= offsetsFromUTF8[csz-1]; + + if (c == ch) { + return &s[lasti]; + } + lasti = i; + (*charn)++; + } + return NULL; +} + +int u8_is_locale_utf8(char *locale) +{ + /* this code based on libutf8 */ + const char* cp = locale; + + for (; *cp != '\0' && *cp != '@' && *cp != '+' && *cp != ','; cp++) { + if (*cp == '.') { + const char* encoding = ++cp; + for (; *cp != '\0' && *cp != '@' && *cp != '+' && *cp != ','; cp++) + ; + if ((cp-encoding == 5 && !strncmp(encoding, "UTF-8", 5)) + || (cp-encoding == 4 && !strncmp(encoding, "utf8", 4))) + return 1; /* it's UTF-8 */ + break; + } + } + return 0; +} + +int u8_vprintf(char *fmt, va_list ap) +{ + int cnt, sz=0; + char *buf; + uint32_t *wcs; + + sz = 512; + buf = (char*)alloca(sz); + try_print: + cnt = vsnprintf(buf, sz, fmt, ap); + if (cnt >= sz) { + buf = (char*)alloca(cnt - sz + 1); + sz = cnt + 1; + goto try_print; + } + wcs = (uint32_t*)alloca((cnt+1) * sizeof(uint32_t)); + cnt = u8_toucs(wcs, cnt+1, buf, cnt); + printf("%ls", (wchar_t*)wcs); + return cnt; +} + +int u8_printf(char *fmt, ...) +{ + int cnt; + va_list args; + + va_start(args, fmt); + + cnt = u8_vprintf(fmt, args); + + va_end(args); + return cnt; +} diff --git a/src/mod/applications/mod_conference/utf8.h b/src/mod/applications/mod_conference/utf8.h new file mode 100644 index 0000000000..c62b1bf48f --- /dev/null +++ b/src/mod/applications/mod_conference/utf8.h @@ -0,0 +1,71 @@ +#include + +/* is c the start of a utf8 sequence? */ +#define isutf(c) (((c)&0xC0)!=0x80) + +/* convert UTF-8 data to wide character */ +int u8_toucs(uint32_t *dest, int sz, char *src, int srcsz); + +/* the opposite conversion */ +int u8_toutf8(char *dest, int sz, uint32_t *src, int srcsz); + +/* single character to UTF-8 */ +int u8_wc_toutf8(char *dest, uint32_t ch); + +/* character number to byte offset */ +int u8_offset(char *str, int charnum); + +/* byte offset to character number */ +int u8_charnum(char *s, int offset); + +/* return next character, updating an index variable */ +uint32_t u8_nextchar(char *s, int *i); + +/* move to next character */ +void u8_inc(char *s, int *i); + +/* move to previous character */ +void u8_dec(char *s, int *i); + +/* returns length of next utf-8 sequence */ +int u8_seqlen(char *s); + +/* assuming src points to the character after a backslash, read an + escape sequence, storing the result in dest and returning the number of + input characters processed */ +int u8_read_escape_sequence(char *src, uint32_t *dest); + +/* given a wide character, convert it to an ASCII escape sequence stored in + buf, where buf is "sz" bytes. returns the number of characters output.*/ +int u8_escape_wchar(char *buf, int sz, uint32_t ch); + +/* convert a string "src" containing escape sequences to UTF-8 */ +int u8_unescape(char *buf, int sz, char *src); + +/* convert UTF-8 "src" to ASCII with escape sequences. + if escape_quotes is nonzero, quote characters will be preceded by + backslashes as well. */ +int u8_escape(char *buf, int sz, char *src, int escape_quotes); + +/* utility predicates used by the above */ +int octal_digit(char c); +int hex_digit(char c); + +/* return a pointer to the first occurrence of ch in s, or NULL if not + found. character index of found character returned in *charn. */ +char *u8_strchr(char *s, uint32_t ch, int *charn); + +/* same as the above, but searches a buffer of a given size instead of + a NUL-terminated string. */ +char *u8_memchr(char *s, uint32_t ch, size_t sz, int *charn); + +/* count the number of characters in a UTF-8 string */ +int u8_strlen(char *s); + +int u8_is_locale_utf8(char *locale); + +/* printf where the format string and arguments may be in UTF-8. + you can avoid this function and just use ordinary printf() if the current + locale is UTF-8. */ +int u8_vprintf(char *fmt, va_list ap); +int u8_printf(char *fmt, ...);