1 //===- unittests/Support/UnicodeTest.cpp - Unicode.h tests ----------------===//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
8 
9 #include "llvm/Support/Unicode.h"
10 #include "llvm/Support/ConvertUTF.h"
11 #include "gtest/gtest.h"
12 
13 namespace llvm {
14 namespace sys {
15 namespace unicode {
16 namespace {
17 
TEST(Unicode,columnWidthUTF8)18 TEST(Unicode, columnWidthUTF8) {
19   EXPECT_EQ(0, columnWidthUTF8(""));
20   EXPECT_EQ(1, columnWidthUTF8(" "));
21   EXPECT_EQ(1, columnWidthUTF8("a"));
22   EXPECT_EQ(1, columnWidthUTF8("~"));
23 
24   EXPECT_EQ(6, columnWidthUTF8("abcdef"));
25 
26   EXPECT_EQ(-1, columnWidthUTF8("\x01"));
27   EXPECT_EQ(-1, columnWidthUTF8("\t"));
28   EXPECT_EQ(-1, columnWidthUTF8("aaaaaaaaaa\x01"));
29   EXPECT_EQ(-1, columnWidthUTF8("\342\200\213")); // 200B ZERO WIDTH SPACE
30 
31   // 00AD SOFT HYPHEN is displayed on most terminals as a space or a dash. Some
32   // text editors display it only when a line is broken at it, some use it as a
33   // line-break hint, but don't display. We choose terminal-oriented
34   // interpretation.
35   EXPECT_EQ(1, columnWidthUTF8("\302\255"));
36 
37   EXPECT_EQ(0, columnWidthUTF8("\314\200"));     // 0300 COMBINING GRAVE ACCENT
38   EXPECT_EQ(1, columnWidthUTF8("\340\270\201")); // 0E01 THAI CHARACTER KO KAI
39   EXPECT_EQ(2, columnWidthUTF8("\344\270\200")); // CJK UNIFIED IDEOGRAPH-4E00
40 
41   EXPECT_EQ(4, columnWidthUTF8("\344\270\200\344\270\200"));
42   EXPECT_EQ(3, columnWidthUTF8("q\344\270\200"));
43   EXPECT_EQ(3, columnWidthUTF8("\314\200\340\270\201\344\270\200"));
44 
45   // Invalid UTF-8 strings, columnWidthUTF8 should error out.
46   EXPECT_EQ(-2, columnWidthUTF8("\344"));
47   EXPECT_EQ(-2, columnWidthUTF8("\344\270"));
48   EXPECT_EQ(-2, columnWidthUTF8("\344\270\033"));
49   EXPECT_EQ(-2, columnWidthUTF8("\344\270\300"));
50   EXPECT_EQ(-2, columnWidthUTF8("\377\366\355"));
51 
52   EXPECT_EQ(-2, columnWidthUTF8("qwer\344"));
53   EXPECT_EQ(-2, columnWidthUTF8("qwer\344\270"));
54   EXPECT_EQ(-2, columnWidthUTF8("qwer\344\270\033"));
55   EXPECT_EQ(-2, columnWidthUTF8("qwer\344\270\300"));
56   EXPECT_EQ(-2, columnWidthUTF8("qwer\377\366\355"));
57 
58   // UTF-8 sequences longer than 4 bytes correspond to unallocated Unicode
59   // characters.
60   EXPECT_EQ(-2, columnWidthUTF8("\370\200\200\200\200"));     // U+200000
61   EXPECT_EQ(-2, columnWidthUTF8("\374\200\200\200\200\200")); // U+4000000
62 }
63 
TEST(Unicode,isPrintable)64 TEST(Unicode, isPrintable) {
65   EXPECT_FALSE(isPrintable(0)); // <control-0000>-<control-001F>
66   EXPECT_FALSE(isPrintable(0x01));
67   EXPECT_FALSE(isPrintable(0x1F));
68   EXPECT_TRUE(isPrintable(' '));
69   EXPECT_TRUE(isPrintable('A'));
70   EXPECT_TRUE(isPrintable('~'));
71   EXPECT_FALSE(isPrintable(0x7F)); // <control-007F>..<control-009F>
72   EXPECT_FALSE(isPrintable(0x90));
73   EXPECT_FALSE(isPrintable(0x9F));
74 
75   EXPECT_TRUE(isPrintable(0xAC));
76   EXPECT_TRUE(isPrintable(0xAD)); // SOFT HYPHEN is displayed on most terminals
77                                   // as either a space or a dash.
78   EXPECT_TRUE(isPrintable(0xAE));
79 
80   EXPECT_TRUE(isPrintable(0x0377));  // GREEK SMALL LETTER PAMPHYLIAN DIGAMMA
81   EXPECT_FALSE(isPrintable(0x0378)); // <reserved-0378>..<reserved-0379>
82 
83   EXPECT_FALSE(isPrintable(0x0600)); // ARABIC NUMBER SIGN
84 
85   EXPECT_FALSE(isPrintable(0x1FFFF)); // <reserved-1F774>..<noncharacter-1FFFF>
86   EXPECT_TRUE(isPrintable(0x20000));  // CJK UNIFIED IDEOGRAPH-20000
87 
88   EXPECT_FALSE(isPrintable(0x10FFFF)); // noncharacter
89 
90   // test the validity of a fast path in columnWidthUTF8
91   for (unsigned char c = 0; c < 128; ++c) {
92     const UTF8 buf8[2] = {c, 0};
93     const UTF8 *Target8 = &buf8[0];
94     UTF32 buf32[1];
95     UTF32 *Target32 = &buf32[0];
96     auto status = ConvertUTF8toUTF32(&Target8, Target8 + 1, &Target32,
97                                      Target32 + 1, strictConversion);
98     EXPECT_TRUE(status == conversionOK);
99     EXPECT_TRUE((columnWidthUTF8(reinterpret_cast<const char *>(buf8)) == 1) ==
100                 (bool)isPrintable(buf32[0]));
101   }
102 }
103 
104 } // namespace
105 } // namespace unicode
106 } // namespace sys
107 } // namespace llvm
108