1 /*
2 * Copyright (C) 2023 The Android Open Source Project
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 * http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License.
15 */
16
17 #include "ScriptUtils.h"
18
19 #include <unicode/ubidi.h>
20 #include <unicode/uscript.h>
21 #include <unicode/utf16.h>
22 #include <unicode/utypes.h>
23
24 #include <algorithm>
25
26 #include "MinikinInternal.h"
27 #include "minikin/Emoji.h"
28
29 namespace minikin {
30
decodeUtf16(U16StringPiece text,Range range,uint32_t pos)31 static hb_codepoint_t decodeUtf16(U16StringPiece text, Range range, uint32_t pos) {
32 uint32_t result;
33 U16_NEXT(text.data(), pos, range.getEnd(), result);
34 if (U_IS_SURROGATE(result)) { // isolated surrogate
35 result = CHAR_REPLACEMENT_CHARACTER;
36 }
37 return static_cast<hb_codepoint_t>(result);
38 }
39
getICUScript(uint32_t cp)40 static UScriptCode getICUScript(uint32_t cp) {
41 UErrorCode status = U_ZERO_ERROR;
42 UScriptCode scriptCode = uscript_getScript(cp, &status);
43 if (U_FAILURE(status)) [[unlikely]] {
44 return USCRIPT_INVALID_CODE;
45 }
46 return scriptCode;
47 }
48
getHbScript(uint32_t cp)49 static hb_script_t getHbScript(uint32_t cp) {
50 hb_unicode_funcs_t* unicode_func = hb_unicode_funcs_get_default();
51 return hb_unicode_script(unicode_func, cp);
52 }
53
54 // static
getScriptRun(U16StringPiece text,Range range,uint32_t pos)55 std::pair<uint32_t, hb_script_t> ScriptText::getScriptRun(U16StringPiece text, Range range,
56 uint32_t pos) {
57 if (!range.contains(pos)) {
58 return std::make_pair(range.getEnd(), HB_SCRIPT_UNKNOWN);
59 }
60
61 uint32_t cp = decodeUtf16(text, range, pos);
62 UScriptCode current_script = getICUScript(cp);
63 hb_script_t current_hb_script = getHbScript(cp);
64 uint32_t i;
65 for (i = pos + U16_LENGTH(cp); i < range.getEnd(); i += U16_LENGTH(cp)) {
66 cp = decodeUtf16(text, range, i);
67 UScriptCode next_script = getICUScript(cp);
68 if (current_script != next_script) {
69 if (current_script == USCRIPT_INHERITED || current_script == USCRIPT_COMMON) {
70 current_script = next_script;
71 current_hb_script = getHbScript(cp);
72 } else if (next_script == USCRIPT_INHERITED || next_script == USCRIPT_COMMON) {
73 continue;
74 } else {
75 break;
76 }
77 }
78 }
79 if (current_script == USCRIPT_INHERITED) {
80 return std::make_pair(i, HB_SCRIPT_COMMON);
81 } else {
82 return std::make_pair(i, current_hb_script);
83 }
84 }
85
86 } // namespace minikin
87