GraphemeBreakTests.cpp revision d8dd94b81ea7efd776859fbbdf4a76458e270eab
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 <gtest/gtest.h> 18#include <UnicodeUtils.h> 19#include <minikin/GraphemeBreak.h> 20 21using namespace android; 22 23bool IsBreak(const char* src) { 24 const size_t BUF_SIZE = 256; 25 uint16_t buf[BUF_SIZE]; 26 size_t offset; 27 size_t size; 28 ParseUnicode(buf, BUF_SIZE, src, &size, &offset); 29 return GraphemeBreak::isGraphemeBreak(buf, 0, size, offset); 30} 31 32TEST(GraphemeBreak, utf16) { 33 EXPECT_FALSE(IsBreak("U+D83C | U+DC31")); // emoji, U+1F431 34 35 // tests for invalid UTF-16 36 EXPECT_TRUE(IsBreak("U+D800 | U+D800")); // two leading surrogates 37 EXPECT_TRUE(IsBreak("U+DC00 | U+DC00")); // two trailing surrogates 38 EXPECT_TRUE(IsBreak("'a' | U+D800")); // lonely leading surrogate 39 EXPECT_TRUE(IsBreak("U+DC00 | 'a'")); // lonely trailing surrogate 40 EXPECT_TRUE(IsBreak("U+D800 | 'a'")); // leading surrogate followed by non-surrogate 41 EXPECT_TRUE(IsBreak("'a' | U+DC00")); // non-surrogate followed by trailing surrogate 42} 43 44TEST(GraphemeBreak, rules) { 45 // Rule GB1, sot ÷; Rule GB2, ÷ eot 46 EXPECT_TRUE(IsBreak("| 'a'")); 47 EXPECT_TRUE(IsBreak("'a' |")); 48 49 // Rule GB3, CR x LF 50 EXPECT_FALSE(IsBreak("U+000D | U+000A")); // CR x LF 51 52 // Rule GB4, (Control | CR | LF) ÷ 53 EXPECT_TRUE(IsBreak("'a' | U+2028")); // Line separator 54 EXPECT_TRUE(IsBreak("'a' | U+000D")); // LF 55 EXPECT_TRUE(IsBreak("'a' | U+000A")); // CR 56 57 // Rule GB5, ÷ (Control | CR | LF) 58 EXPECT_TRUE(IsBreak("U+2028 | 'a'")); // Line separator 59 EXPECT_TRUE(IsBreak("U+000D | 'a'")); // LF 60 EXPECT_TRUE(IsBreak("U+000A | 'a'")); // CR 61 62 // Rule GB6, L x ( L | V | LV | LVT ) 63 EXPECT_FALSE(IsBreak("U+1100 | U+1100")); // L x L 64 EXPECT_FALSE(IsBreak("U+1100 | U+1161")); // L x V 65 EXPECT_FALSE(IsBreak("U+1100 | U+AC00")); // L x LV 66 EXPECT_FALSE(IsBreak("U+1100 | U+AC01")); // L x LVT 67 68 // Rule GB7, ( LV | V ) x ( V | T ) 69 EXPECT_FALSE(IsBreak("U+AC00 | U+1161")); // LV x V 70 EXPECT_FALSE(IsBreak("U+1161 | U+1161")); // V x V 71 EXPECT_FALSE(IsBreak("U+AC00 | U+11A8")); // LV x T 72 EXPECT_FALSE(IsBreak("U+1161 | U+11A8")); // V x T 73 74 // Rule GB8, ( LVT | T ) x T 75 EXPECT_FALSE(IsBreak("U+AC01 | U+11A8")); // LVT x T 76 EXPECT_FALSE(IsBreak("U+11A8 | U+11A8")); // T x T 77 78 // Other hangul pairs not counted above _are_ breaks (GB10) 79 EXPECT_TRUE(IsBreak("U+AC00 | U+1100")); // LV x L 80 EXPECT_TRUE(IsBreak("U+AC01 | U+1100")); // LVT x L 81 EXPECT_TRUE(IsBreak("U+11A8 | U+1100")); // T x L 82 EXPECT_TRUE(IsBreak("U+11A8 | U+AC00")); // T x LV 83 EXPECT_TRUE(IsBreak("U+11A8 | U+AC01")); // T x LVT 84 85 // Rule GB8a, Regional_Indicator x Regional_Indicator 86 EXPECT_FALSE(IsBreak("U+1F1FA | U+1F1F8")); 87 88 // Rule GB9, x Extend 89 EXPECT_FALSE(IsBreak("'a' | U+0301")); // combining accent 90 // Rule GB9a, x SpacingMark 91 EXPECT_FALSE(IsBreak("U+0915 | U+093E")); // KA, AA (spacing mark) 92 // Rule GB9b, Prepend x 93 // see tailoring test for prepend, as current ICU doesn't have any characters in the class 94 95 // Rule GB10, Any ÷ Any 96 EXPECT_TRUE(IsBreak("'a' | 'b'")); 97 EXPECT_TRUE(IsBreak("'f' | 'i'")); // probable ligature 98 EXPECT_TRUE(IsBreak("U+0644 | U+0627")); // probable ligature, lam + alef 99 EXPECT_TRUE(IsBreak("U+4E00 | U+4E00")); // CJK ideographs 100 EXPECT_TRUE(IsBreak("'a' | U+1F1FA U+1F1F8")); // Regional indicator pair (flag) 101 EXPECT_TRUE(IsBreak("U+1F1FA U+1F1F8 | 'a'")); // Regional indicator pair (flag) 102} 103 104TEST(GraphemeBreak, tailoring) { 105 // control characters that we interpret as "extend" 106 EXPECT_FALSE(IsBreak("'a' | U+00AD")); // soft hyphen 107 EXPECT_FALSE(IsBreak("'a' | U+200B")); // zwsp 108 EXPECT_FALSE(IsBreak("'a' | U+200E")); // lrm 109 EXPECT_FALSE(IsBreak("'a' | U+202A")); // lre 110 EXPECT_FALSE(IsBreak("'a' | U+E0041")); // tag character 111 112 // UTC-approved characters for the Prepend class 113 EXPECT_FALSE(IsBreak("U+06DD | U+0661")); // arabic subtending mark + digit one 114 115 EXPECT_TRUE(IsBreak("U+0E01 | U+0E33")); // Thai sara am 116 117 // virama is not a grapheme break, but "pure killer" is 118 EXPECT_FALSE(IsBreak("U+0915 | U+094D U+0915")); // Devanagari ka+virama+ka 119 EXPECT_FALSE(IsBreak("U+0915 U+094D | U+0915")); // Devanagari ka+virama+ka 120 EXPECT_FALSE(IsBreak("U+0E01 | U+0E3A U+0E01")); // thai phinthu = pure killer 121 EXPECT_TRUE(IsBreak("U+0E01 U+0E3A | U+0E01")); // thai phinthu = pure killer 122} 123 124TEST(GraphemeBreak, offsets) { 125 uint16_t string[] = { 0x0041, 0x06DD, 0x0045, 0x0301, 0x0049, 0x0301 }; 126 EXPECT_TRUE(GraphemeBreak::isGraphemeBreak(string, 2, 3, 2)); 127 EXPECT_FALSE(GraphemeBreak::isGraphemeBreak(string, 2, 3, 3)); 128 EXPECT_TRUE(GraphemeBreak::isGraphemeBreak(string, 2, 3, 4)); 129 EXPECT_TRUE(GraphemeBreak::isGraphemeBreak(string, 2, 3, 5)); 130} 131