1 /*
2 * Copyright (C) 2015 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 "LayoutUtils.h"
18
19 #include "StringPiece.h"
20
21 namespace minikin {
22
23 /*
24 * Determine whether the code unit is a word space for the purposes of justification.
25 * TODO: Support NBSP and other stretchable whitespace (b/34013491 and b/68204709).
26 */
isWordSpace(uint16_t code_unit)27 bool isWordSpace(uint16_t code_unit) {
28 return code_unit == ' ';
29 }
30
31 /**
32 * For the purpose of layout, a word break is a boundary with no
33 * kerning or complex script processing. This is necessarily a
34 * heuristic, but should be accurate most of the time.
35 */
isWordBreakAfter(uint16_t c)36 static bool isWordBreakAfter(uint16_t c) {
37 if (c == ' ' || (0x2000 <= c && c <= 0x200A) || c == 0x3000) {
38 // spaces
39 return true;
40 }
41 // Break layout context before and after BiDi control character.
42 if ((0x2066 <= c && c <= 0x2069) || (0x202A <= c && c <= 0x202E) || c == 0x200E ||
43 c == 0x200F) {
44 return true;
45 }
46 // Note: kana is not included, as sophisticated fonts may kern kana
47 return false;
48 }
49
isWordBreakBefore(uint16_t c)50 static bool isWordBreakBefore(uint16_t c) {
51 // CJK ideographs (and yijing hexagram symbols)
52 return isWordBreakAfter(c) || (0x3400 <= c && c <= 0x9FFF);
53 }
54
55 /**
56 * Return offset of previous word break. It is either < offset or == 0.
57 */
getPrevWordBreakForCache(const U16StringPiece & textBuf,uint32_t offset)58 uint32_t getPrevWordBreakForCache(const U16StringPiece& textBuf, uint32_t offset) {
59 if (offset == 0) return 0;
60 if (offset > textBuf.size()) offset = textBuf.size();
61 if (isWordBreakBefore(textBuf[offset - 1])) {
62 return offset - 1;
63 }
64 for (uint32_t i = offset - 1; i > 0; i--) {
65 if (isWordBreakBefore(textBuf[i]) || isWordBreakAfter(textBuf[i - 1])) {
66 return i;
67 }
68 }
69 return 0;
70 }
71
72 /**
73 * Return offset of next word break. It is either > offset or == len.
74 */
getNextWordBreakForCache(const U16StringPiece & textBuf,uint32_t offset)75 uint32_t getNextWordBreakForCache(const U16StringPiece& textBuf, uint32_t offset) {
76 if (offset >= textBuf.size()) return textBuf.size();
77 if (isWordBreakAfter(textBuf[offset])) {
78 return offset + 1;
79 }
80 for (uint32_t i = offset + 1; i < textBuf.size(); i++) {
81 // No need to check isWordBreakAfter(chars[i - 1]) since it is checked
82 // in previous iteration. Note that isWordBreakBefore returns true
83 // whenever isWordBreakAfter returns true.
84 if (isWordBreakBefore(textBuf[i])) {
85 return i;
86 }
87 }
88 return textBuf.size();
89 }
90
91 } // namespace minikin
92