| 1 | #include "mupdf/fitz.h" |
| 2 | #include "mupdf/pdf.h" |
| 3 | |
| 4 | #include <string.h> |
| 5 | #include <stdlib.h> |
| 6 | |
| 7 | #include "encodings.h" |
| 8 | #include "glyphlist.h" |
| 9 | #include "smallcaps.h" |
| 10 | |
| 11 | #define FROM_UNICODE(ENC) \ |
| 12 | int l = 0; \ |
| 13 | int r = nelem(ENC##_from_unicode) - 1; \ |
| 14 | if (u < 128) \ |
| 15 | return u; \ |
| 16 | while (l <= r) \ |
| 17 | { \ |
| 18 | int m = (l + r) >> 1; \ |
| 19 | if (u < ENC##_from_unicode[m].u) \ |
| 20 | r = m - 1; \ |
| 21 | else if (u > ENC##_from_unicode[m].u) \ |
| 22 | l = m + 1; \ |
| 23 | else \ |
| 24 | return ENC##_from_unicode[m].c; \ |
| 25 | } \ |
| 26 | return -1; \ |
| 27 | |
| 28 | int fz_iso8859_1_from_unicode(int u) { FROM_UNICODE(iso8859_1) } |
| 29 | int fz_iso8859_7_from_unicode(int u) { FROM_UNICODE(iso8859_7) } |
| 30 | int fz_koi8u_from_unicode(int u) { FROM_UNICODE(koi8u) } |
| 31 | int fz_windows_1250_from_unicode(int u) { FROM_UNICODE(windows_1250) } |
| 32 | int fz_windows_1251_from_unicode(int u) { FROM_UNICODE(windows_1251) } |
| 33 | int fz_windows_1252_from_unicode(int u) { FROM_UNICODE(windows_1252) } |
| 34 | |
| 35 | int |
| 36 | fz_unicode_from_glyph_name_strict(const char *name) |
| 37 | { |
| 38 | int l = 0; |
| 39 | int r = nelem(single_name_list) - 1; |
| 40 | |
| 41 | while (l <= r) |
| 42 | { |
| 43 | int m = (l + r) >> 1; |
| 44 | int c = strcmp(name, single_name_list[m]); |
| 45 | if (c < 0) |
| 46 | r = m - 1; |
| 47 | else if (c > 0) |
| 48 | l = m + 1; |
| 49 | else |
| 50 | return single_code_list[m]; |
| 51 | } |
| 52 | return 0; |
| 53 | } |
| 54 | |
| 55 | int |
| 56 | fz_unicode_from_glyph_name(const char *name) |
| 57 | { |
| 58 | char buf[64]; |
| 59 | char *p; |
| 60 | int l = 0; |
| 61 | int r = nelem(single_name_list) - 1; |
| 62 | int code = 0; |
| 63 | |
| 64 | fz_strlcpy(buf, name, sizeof buf); |
| 65 | |
| 66 | /* kill anything after first period and underscore */ |
| 67 | p = strchr(buf, '.'); |
| 68 | if (p) p[0] = 0; |
| 69 | p = strchr(buf, '_'); |
| 70 | if (p) p[0] = 0; |
| 71 | |
| 72 | while (l <= r) |
| 73 | { |
| 74 | int m = (l + r) >> 1; |
| 75 | int c = strcmp(buf, single_name_list[m]); |
| 76 | if (c < 0) |
| 77 | r = m - 1; |
| 78 | else if (c > 0) |
| 79 | l = m + 1; |
| 80 | else |
| 81 | return single_code_list[m]; |
| 82 | } |
| 83 | |
| 84 | if (buf[0] == 'u' && buf[1] == 'n' && buf[2] == 'i') |
| 85 | code = strtol(buf + 3, NULL, 16); |
| 86 | else if (buf[0] == 'u') |
| 87 | code = strtol(buf + 1, NULL, 16); |
| 88 | else if (buf[0] == 'a' && buf[1] != 0 && buf[2] != 0) |
| 89 | code = strtol(buf + 1, NULL, 10); |
| 90 | |
| 91 | return (code > 0 && code <= 0x10ffff) ? code : FZ_REPLACEMENT_CHARACTER; |
| 92 | } |
| 93 | |
| 94 | static const char *empty_dup_list[] = { 0 }; |
| 95 | |
| 96 | const char ** |
| 97 | fz_duplicate_glyph_names_from_unicode(int ucs) |
| 98 | { |
| 99 | int l = 0; |
| 100 | int r = nelem(agl_dup_offsets) / 2 - 1; |
| 101 | while (l <= r) |
| 102 | { |
| 103 | int m = (l + r) >> 1; |
| 104 | if (ucs < agl_dup_offsets[m << 1]) |
| 105 | r = m - 1; |
| 106 | else if (ucs > agl_dup_offsets[m << 1]) |
| 107 | l = m + 1; |
| 108 | else |
| 109 | return agl_dup_names + agl_dup_offsets[(m << 1) + 1]; |
| 110 | } |
| 111 | return empty_dup_list; |
| 112 | } |
| 113 | |
| 114 | const char * |
| 115 | fz_glyph_name_from_unicode_sc(int u) |
| 116 | { |
| 117 | int l = 0; |
| 118 | int r = nelem(glyph_name_from_unicode_sc) / 2 - 1; |
| 119 | while (l <= r) |
| 120 | { |
| 121 | int m = (l + r) >> 1; |
| 122 | if (u < glyph_name_from_unicode_sc[m].u) |
| 123 | r = m - 1; |
| 124 | else if (u > glyph_name_from_unicode_sc[m].u) |
| 125 | l = m + 1; |
| 126 | else |
| 127 | return glyph_name_from_unicode_sc[m].n; |
| 128 | } |
| 129 | return NULL; |
| 130 | } |
| 131 | |