+++ /dev/null
-/*\r
- * This is an implementation of wcwidth() and wcswidth() (defined in\r
- * IEEE Std 1002.1-2001) for Unicode.\r
- *\r
- * http://www.opengroup.org/onlinepubs/007904975/functions/wcwidth.html\r
- * http://www.opengroup.org/onlinepubs/007904975/functions/wcswidth.html\r
- *\r
- * In fixed-width output devices, Latin characters all occupy a single\r
- * "cell" position of equal width, whereas ideographic CJK characters\r
- * occupy two such cells. Interoperability between terminal-line\r
- * applications and (teletype-style) character terminals using the\r
- * UTF-8 encoding requires agreement on which character should advance\r
- * the cursor by how many cell positions. No established formal\r
- * standards exist at present on which Unicode character shall occupy\r
- * how many cell positions on character terminals. These routines are\r
- * a first attempt of defining such behavior based on simple rules\r
- * applied to data provided by the Unicode Consortium.\r
- *\r
- * For some graphical characters, the Unicode standard explicitly\r
- * defines a character-cell width via the definition of the East Asian\r
- * FullWidth (F), Wide (W), Half-width (H), and Narrow (Na) classes.\r
- * In all these cases, there is no ambiguity about which width a\r
- * terminal shall use. For characters in the East Asian Ambiguous (A)\r
- * class, the width choice depends purely on a preference of backward\r
- * compatibility with either historic CJK or Western practice.\r
- * Choosing single-width for these characters is easy to justify as\r
- * the appropriate long-term solution, as the CJK practice of\r
- * displaying these characters as double-width comes from historic\r
- * implementation simplicity (8-bit encoded characters were displayed\r
- * single-width and 16-bit ones double-width, even for Greek,\r
- * Cyrillic, etc.) and not any typographic considerations.\r
- *\r
- * Much less clear is the choice of width for the Not East Asian\r
- * (Neutral) class. Existing practice does not dictate a width for any\r
- * of these characters. It would nevertheless make sense\r
- * typographically to allocate two character cells to characters such\r
- * as for instance EM SPACE or VOLUME INTEGRAL, which cannot be\r
- * represented adequately with a single-width glyph. The following\r
- * routines at present merely assign a single-cell width to all\r
- * neutral characters, in the interest of simplicity. This is not\r
- * entirely satisfactory and should be reconsidered before\r
- * establishing a formal standard in this area. At the moment, the\r
- * decision which Not East Asian (Neutral) characters should be\r
- * represented by double-width glyphs cannot yet be answered by\r
- * applying a simple rule from the Unicode database content. Setting\r
- * up a proper standard for the behavior of UTF-8 character terminals\r
- * will require a careful analysis not only of each Unicode character,\r
- * but also of each presentation form, something the author of these\r
- * routines has avoided to do so far.\r
- *\r
- * http://www.unicode.org/unicode/reports/tr11/\r
- *\r
- * Markus Kuhn -- 2003-05-20 (Unicode 4.0)\r
- *\r
- * Permission to use, copy, modify, and distribute this software\r
- * for any purpose and without fee is hereby granted. The author\r
- * disclaims all warranties with regard to this software.\r
- *\r
- * Latest version: http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c\r
- */\r
-\r
-#include <wchar.h>\r
-\r
-#include "putty.h" /* for prototypes */\r
-\r
-struct interval {\r
- int first;\r
- int last;\r
-};\r
-\r
-/* auxiliary function for binary search in interval table */\r
-static int bisearch(wchar_t ucs, const struct interval *table, int max) {\r
- int min = 0;\r
- int mid;\r
-\r
- if (ucs < table[0].first || ucs > table[max].last)\r
- return 0;\r
- while (max >= min) {\r
- mid = (min + max) / 2;\r
- if (ucs > table[mid].last)\r
- min = mid + 1;\r
- else if (ucs < table[mid].first)\r
- max = mid - 1;\r
- else\r
- return 1;\r
- }\r
-\r
- return 0;\r
-}\r
-\r
-\r
-/* The following two functions define the column width of an ISO 10646\r
- * character as follows:\r
- *\r
- * - The null character (U+0000) has a column width of 0.\r
- *\r
- * - Other C0/C1 control characters and DEL will lead to a return\r
- * value of -1.\r
- *\r
- * - Non-spacing and enclosing combining characters (general\r
- * category code Mn or Me in the Unicode database) have a\r
- * column width of 0.\r
- *\r
- * - SOFT HYPHEN (U+00AD) has a column width of 1.\r
- *\r
- * - Other format characters (general category code Cf in the Unicode\r
- * database) and ZERO WIDTH SPACE (U+200B) have a column width of 0.\r
- *\r
- * - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF)\r
- * have a column width of 0.\r
- *\r
- * - Spacing characters in the East Asian Wide (W) or East Asian\r
- * Full-width (F) category as defined in Unicode Technical\r
- * Report #11 have a column width of 2.\r
- *\r
- * - All remaining characters (including all printable\r
- * ISO 8859-1 and WGL4 characters, Unicode control characters,\r
- * etc.) have a column width of 1.\r
- *\r
- * This implementation assumes that wchar_t characters are encoded\r
- * in ISO 10646.\r
- */\r
-\r
-int mk_wcwidth(wchar_t ucs)\r
-{\r
- /* sorted list of non-overlapping intervals of non-spacing characters */\r
- /* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c" */\r
- static const struct interval combining[] = {\r
- { 0x0300, 0x0357 }, { 0x035D, 0x036F }, { 0x0483, 0x0486 },\r
- { 0x0488, 0x0489 }, { 0x0591, 0x05A1 }, { 0x05A3, 0x05B9 },\r
- { 0x05BB, 0x05BD }, { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 },\r
- { 0x05C4, 0x05C4 }, { 0x0600, 0x0603 }, { 0x0610, 0x0615 },\r
- { 0x064B, 0x0658 }, { 0x0670, 0x0670 }, { 0x06D6, 0x06E4 },\r
- { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED }, { 0x070F, 0x070F },\r
- { 0x0711, 0x0711 }, { 0x0730, 0x074A }, { 0x07A6, 0x07B0 },\r
- { 0x0901, 0x0902 }, { 0x093C, 0x093C }, { 0x0941, 0x0948 },\r
- { 0x094D, 0x094D }, { 0x0951, 0x0954 }, { 0x0962, 0x0963 },\r
- { 0x0981, 0x0981 }, { 0x09BC, 0x09BC }, { 0x09C1, 0x09C4 },\r
- { 0x09CD, 0x09CD }, { 0x09E2, 0x09E3 }, { 0x0A01, 0x0A02 },\r
- { 0x0A3C, 0x0A3C }, { 0x0A41, 0x0A42 }, { 0x0A47, 0x0A48 },\r
- { 0x0A4B, 0x0A4D }, { 0x0A70, 0x0A71 }, { 0x0A81, 0x0A82 },\r
- { 0x0ABC, 0x0ABC }, { 0x0AC1, 0x0AC5 }, { 0x0AC7, 0x0AC8 },\r
- { 0x0ACD, 0x0ACD }, { 0x0AE2, 0x0AE3 }, { 0x0B01, 0x0B01 },\r
- { 0x0B3C, 0x0B3C }, { 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 },\r
- { 0x0B4D, 0x0B4D }, { 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 },\r
- { 0x0BC0, 0x0BC0 }, { 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 },\r
- { 0x0C46, 0x0C48 }, { 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 },\r
- { 0x0CBC, 0x0CBC }, { 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 },\r
- { 0x0CCC, 0x0CCD }, { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D },\r
- { 0x0DCA, 0x0DCA }, { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 },\r
- { 0x0E31, 0x0E31 }, { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E },\r
- { 0x0EB1, 0x0EB1 }, { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC },\r
- { 0x0EC8, 0x0ECD }, { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 },\r
- { 0x0F37, 0x0F37 }, { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E },\r
- { 0x0F80, 0x0F84 }, { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 },\r
- { 0x0F99, 0x0FBC }, { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 },\r
- { 0x1032, 0x1032 }, { 0x1036, 0x1037 }, { 0x1039, 0x1039 },\r
- { 0x1058, 0x1059 }, { 0x1160, 0x11FF }, { 0x1712, 0x1714 },\r
- { 0x1732, 0x1734 }, { 0x1752, 0x1753 }, { 0x1772, 0x1773 },\r
- { 0x17B4, 0x17B5 }, { 0x17B7, 0x17BD }, { 0x17C6, 0x17C6 },\r
- { 0x17C9, 0x17D3 }, { 0x17DD, 0x17DD }, { 0x180B, 0x180D },\r
- { 0x18A9, 0x18A9 }, { 0x1920, 0x1922 }, { 0x1927, 0x1928 },\r
- { 0x1932, 0x1932 }, { 0x1939, 0x193B }, { 0x200B, 0x200F },\r
- { 0x202A, 0x202E }, { 0x2060, 0x2063 }, { 0x206A, 0x206F },\r
- { 0x20D0, 0x20EA }, { 0x302A, 0x302F }, { 0x3099, 0x309A },\r
- { 0xFB1E, 0xFB1E }, { 0xFE00, 0xFE0F }, { 0xFE20, 0xFE23 },\r
- { 0xFEFF, 0xFEFF }, { 0xFFF9, 0xFFFB }, { 0x1D167, 0x1D169 },\r
- { 0x1D173, 0x1D182 }, { 0x1D185, 0x1D18B }, { 0x1D1AA, 0x1D1AD },\r
- { 0xE0001, 0xE0001 }, { 0xE0020, 0xE007F }, { 0xE0100, 0xE01EF }\r
- };\r
-\r
- /* test for 8-bit control characters */\r
- if (ucs == 0)\r
- return 0;\r
- if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0))\r
- return -1;\r
-\r
- /* binary search in table of non-spacing characters */\r
- if (bisearch(ucs, combining,\r
- sizeof(combining) / sizeof(struct interval) - 1))\r
- return 0;\r
-\r
- /* if we arrive here, ucs is not a combining or C0/C1 control character */\r
-\r
- return 1 + \r
- (ucs >= 0x1100 &&\r
- (ucs <= 0x115f || /* Hangul Jamo init. consonants */\r
- ucs == 0x2329 || ucs == 0x232a ||\r
- (ucs >= 0x2e80 && ucs <= 0xa4cf &&\r
- ucs != 0x303f) || /* CJK ... Yi */\r
- (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */\r
- (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */\r
- (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */\r
- (ucs >= 0xff00 && ucs <= 0xff60) || /* Fullwidth Forms */\r
- (ucs >= 0xffe0 && ucs <= 0xffe6) ||\r
- (ucs >= 0x20000 && ucs <= 0x2fffd) ||\r
- (ucs >= 0x30000 && ucs <= 0x3fffd)));\r
-}\r
-\r
-\r
-int mk_wcswidth(const wchar_t *pwcs, size_t n)\r
-{\r
- int w, width = 0;\r
-\r
- for (;*pwcs && n-- > 0; pwcs++)\r
- if ((w = mk_wcwidth(*pwcs)) < 0)\r
- return -1;\r
- else\r
- width += w;\r
-\r
- return width;\r
-}\r
-\r
-\r
-/*\r
- * The following functions are the same as mk_wcwidth() and\r
- * mk_wcwidth_cjk(), except that spacing characters in the East Asian\r
- * Ambiguous (A) category as defined in Unicode Technical Report #11\r
- * have a column width of 2. This variant might be useful for users of\r
- * CJK legacy encodings who want to migrate to UCS without changing\r
- * the traditional terminal character-width behaviour. It is not\r
- * otherwise recommended for general use.\r
- */\r
-int mk_wcwidth_cjk(wchar_t ucs)\r
-{\r
- /* sorted list of non-overlapping intervals of East Asian Ambiguous\r
- * characters, generated by "uniset +WIDTH-A -cat=Me -cat=Mn -cat=Cf c" */\r
- static const struct interval ambiguous[] = {\r
- { 0x00A1, 0x00A1 }, { 0x00A4, 0x00A4 }, { 0x00A7, 0x00A8 },\r
- { 0x00AA, 0x00AA }, { 0x00AE, 0x00AE }, { 0x00B0, 0x00B4 },\r
- { 0x00B6, 0x00BA }, { 0x00BC, 0x00BF }, { 0x00C6, 0x00C6 },\r
- { 0x00D0, 0x00D0 }, { 0x00D7, 0x00D8 }, { 0x00DE, 0x00E1 },\r
- { 0x00E6, 0x00E6 }, { 0x00E8, 0x00EA }, { 0x00EC, 0x00ED },\r
- { 0x00F0, 0x00F0 }, { 0x00F2, 0x00F3 }, { 0x00F7, 0x00FA },\r
- { 0x00FC, 0x00FC }, { 0x00FE, 0x00FE }, { 0x0101, 0x0101 },\r
- { 0x0111, 0x0111 }, { 0x0113, 0x0113 }, { 0x011B, 0x011B },\r
- { 0x0126, 0x0127 }, { 0x012B, 0x012B }, { 0x0131, 0x0133 },\r
- { 0x0138, 0x0138 }, { 0x013F, 0x0142 }, { 0x0144, 0x0144 },\r
- { 0x0148, 0x014B }, { 0x014D, 0x014D }, { 0x0152, 0x0153 },\r
- { 0x0166, 0x0167 }, { 0x016B, 0x016B }, { 0x01CE, 0x01CE },\r
- { 0x01D0, 0x01D0 }, { 0x01D2, 0x01D2 }, { 0x01D4, 0x01D4 },\r
- { 0x01D6, 0x01D6 }, { 0x01D8, 0x01D8 }, { 0x01DA, 0x01DA },\r
- { 0x01DC, 0x01DC }, { 0x0251, 0x0251 }, { 0x0261, 0x0261 },\r
- { 0x02C4, 0x02C4 }, { 0x02C7, 0x02C7 }, { 0x02C9, 0x02CB },\r
- { 0x02CD, 0x02CD }, { 0x02D0, 0x02D0 }, { 0x02D8, 0x02DB },\r
- { 0x02DD, 0x02DD }, { 0x02DF, 0x02DF }, { 0x0391, 0x03A1 },\r
- { 0x03A3, 0x03A9 }, { 0x03B1, 0x03C1 }, { 0x03C3, 0x03C9 },\r
- { 0x0401, 0x0401 }, { 0x0410, 0x044F }, { 0x0451, 0x0451 },\r
- { 0x2010, 0x2010 }, { 0x2013, 0x2016 }, { 0x2018, 0x2019 },\r
- { 0x201C, 0x201D }, { 0x2020, 0x2022 }, { 0x2024, 0x2027 },\r
- { 0x2030, 0x2030 }, { 0x2032, 0x2033 }, { 0x2035, 0x2035 },\r
- { 0x203B, 0x203B }, { 0x203E, 0x203E }, { 0x2074, 0x2074 },\r
- { 0x207F, 0x207F }, { 0x2081, 0x2084 }, { 0x20AC, 0x20AC },\r
- { 0x2103, 0x2103 }, { 0x2105, 0x2105 }, { 0x2109, 0x2109 },\r
- { 0x2113, 0x2113 }, { 0x2116, 0x2116 }, { 0x2121, 0x2122 },\r
- { 0x2126, 0x2126 }, { 0x212B, 0x212B }, { 0x2153, 0x2154 },\r
- { 0x215B, 0x215E }, { 0x2160, 0x216B }, { 0x2170, 0x2179 },\r
- { 0x2190, 0x2199 }, { 0x21B8, 0x21B9 }, { 0x21D2, 0x21D2 },\r
- { 0x21D4, 0x21D4 }, { 0x21E7, 0x21E7 }, { 0x2200, 0x2200 },\r
- { 0x2202, 0x2203 }, { 0x2207, 0x2208 }, { 0x220B, 0x220B },\r
- { 0x220F, 0x220F }, { 0x2211, 0x2211 }, { 0x2215, 0x2215 },\r
- { 0x221A, 0x221A }, { 0x221D, 0x2220 }, { 0x2223, 0x2223 },\r
- { 0x2225, 0x2225 }, { 0x2227, 0x222C }, { 0x222E, 0x222E },\r
- { 0x2234, 0x2237 }, { 0x223C, 0x223D }, { 0x2248, 0x2248 },\r
- { 0x224C, 0x224C }, { 0x2252, 0x2252 }, { 0x2260, 0x2261 },\r
- { 0x2264, 0x2267 }, { 0x226A, 0x226B }, { 0x226E, 0x226F },\r
- { 0x2282, 0x2283 }, { 0x2286, 0x2287 }, { 0x2295, 0x2295 },\r
- { 0x2299, 0x2299 }, { 0x22A5, 0x22A5 }, { 0x22BF, 0x22BF },\r
- { 0x2312, 0x2312 }, { 0x2460, 0x24E9 }, { 0x24EB, 0x254B },\r
- { 0x2550, 0x2573 }, { 0x2580, 0x258F }, { 0x2592, 0x2595 },\r
- { 0x25A0, 0x25A1 }, { 0x25A3, 0x25A9 }, { 0x25B2, 0x25B3 },\r
- { 0x25B6, 0x25B7 }, { 0x25BC, 0x25BD }, { 0x25C0, 0x25C1 },\r
- { 0x25C6, 0x25C8 }, { 0x25CB, 0x25CB }, { 0x25CE, 0x25D1 },\r
- { 0x25E2, 0x25E5 }, { 0x25EF, 0x25EF }, { 0x2605, 0x2606 },\r
- { 0x2609, 0x2609 }, { 0x260E, 0x260F }, { 0x2614, 0x2615 },\r
- { 0x261C, 0x261C }, { 0x261E, 0x261E }, { 0x2640, 0x2640 },\r
- { 0x2642, 0x2642 }, { 0x2660, 0x2661 }, { 0x2663, 0x2665 },\r
- { 0x2667, 0x266A }, { 0x266C, 0x266D }, { 0x266F, 0x266F },\r
- { 0x273D, 0x273D }, { 0x2776, 0x277F }, { 0xE000, 0xF8FF },\r
- { 0xFFFD, 0xFFFD }, { 0xF0000, 0xFFFFD }, { 0x100000, 0x10FFFD }\r
- };\r
-\r
- /* binary search in table of non-spacing characters */\r
- if (bisearch(ucs, ambiguous,\r
- sizeof(ambiguous) / sizeof(struct interval) - 1))\r
- return 2;\r
-\r
- return mk_wcwidth(ucs);\r
-}\r
-\r
-\r
-int mk_wcswidth_cjk(const wchar_t *pwcs, size_t n)\r
-{\r
- int w, width = 0;\r
-\r
- for (;*pwcs && n-- > 0; pwcs++)\r
- if ((w = mk_wcwidth_cjk(*pwcs)) < 0)\r
- return -1;\r
- else\r
- width += w;\r
-\r
- return width;\r
-}\r
OSDN Git Service
