Blender  V3.3
BLI_string_utf8_test.cc
Go to the documentation of this file.
1 /* SPDX-License-Identifier: Apache-2.0 */
2 
3 #include "testing/testing.h"
4 
5 #include "BLI_rand.h"
6 #include "BLI_string.h"
7 #include "BLI_string_utf8.h"
8 #include "BLI_utildefines.h"
9 
10 /* Note that 'common' utf-8 variants of string functions (like copy, etc.) are tested in
11  * BLI_string_test.cc However, tests below are specific utf-8 conformance ones, and since they eat
12  * quite their share of lines, they deserved their own file. */
13 
14 /* -------------------------------------------------------------------- */
18 /* Breaking strings is confusing here, prefer over-long lines. */
19 /* clang-format off */
20 
21 /* Each test is made of a 79 bytes (80 with NULL char) string to test, expected string result after
22  * stripping invalid utf8 bytes, and a single-byte string encoded with expected number of errors.
23  *
24  * Based on utf-8 decoder stress-test (https://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-test.txt)
25  * by Markus Kuhn <http://www.cl.cam.ac.uk/~mgk25/> - 2015-08-28 - CC BY 4.0
26  */
27 static const char *utf8_invalid_tests[][3] = {
28 /* 1 Some correct UTF-8 text. */
29  {"You should see the Greek word 'kosme': \"\xce\xba\xe1\xbd\xb9\xcf\x83\xce\xbc\xce\xb5\" |",
30  "You should see the Greek word 'kosme': \"\xce\xba\xe1\xbd\xb9\xcf\x83\xce\xbc\xce\xb5\" |", "\x00"},
31 
32 /* 2 Boundary condition test cases
33  * Note that those will pass for us, those are not erronĂ©ous unicode code points
34  * (aside from \x00, which is only valid as string terminator).
35  * 2.1 First possible sequence of a certain length */
36  {"2.1.1 1 byte (U-00000000): \"\x00\" |",
37  "2.1.1 1 byte (U-00000000): \"\" |", "\x01"},
38  {"2.1.2 2 bytes (U-00000080): \"\xc2\x80\" |",
39  "2.1.2 2 bytes (U-00000080): \"\xc2\x80\" |", "\x00"},
40  {"2.1.3 3 bytes (U-00000800): \"\xe0\xa0\x80\" |",
41  "2.1.3 3 bytes (U-00000800): \"\xe0\xa0\x80\" |", "\x00"},
42  {"2.1.4 4 bytes (U-00010000): \"\xf0\x90\x80\x80\" |",
43  "2.1.4 4 bytes (U-00010000): \"\xf0\x90\x80\x80\" |", "\x00"},
44  {"2.1.5 5 bytes (U-00200000): \"\xf8\x88\x80\x80\x80\" |",
45  "2.1.5 5 bytes (U-00200000): \"\xf8\x88\x80\x80\x80\" |", "\x00"},
46  {"2.1.6 6 bytes (U-04000000): \"\xfc\x84\x80\x80\x80\x80\" |",
47  "2.1.6 6 bytes (U-04000000): \"\xfc\x84\x80\x80\x80\x80\" |", "\x00"},
48 /* 2.2 Last possible sequence of a certain length */
49  {"2.2.1 1 byte (U-0000007F): \"\x7f\" |",
50  "2.2.1 1 byte (U-0000007F): \"\x7f\" |", "\x00"},
51  {"2.2.2 2 bytes (U-000007FF): \"\xdf\xbf\" |",
52  "2.2.2 2 bytes (U-000007FF): \"\xdf\xbf\" |", "\x00"},
53  {"2.2.3 3 bytes (U-0000FFFF): \"\xef\xbf\xbf\" |",
54  "2.2.3 3 bytes (U-0000FFFF): \"\" |", "\x03"}, /* matches one of 5.3 sequences... */
55  {"2.2.4 4 bytes (U-001FFFFF): \"\xf7\xbf\xbf\xbf\" |",
56  "2.2.4 4 bytes (U-001FFFFF): \"\xf7\xbf\xbf\xbf\" |", "\x00"},
57  {"2.2.5 5 bytes (U-03FFFFFF): \"\xfb\xbf\xbf\xbf\xbf\" |",
58  "2.2.5 5 bytes (U-03FFFFFF): \"\xfb\xbf\xbf\xbf\xbf\" |", "\x00"},
59  {"2.2.6 6 bytes (U-7FFFFFFF): \"\xfd\xbf\xbf\xbf\xbf\xbf\" |",
60  "2.2.6 6 bytes (U-7FFFFFFF): \"\xfd\xbf\xbf\xbf\xbf\xbf\" |", "\x00"},
61 /* 2.3 Other boundary conditions */
62  {"2.3.1 U-0000D7FF = ed 9f bf = \"\xed\x9f\xbf\" |",
63  "2.3.1 U-0000D7FF = ed 9f bf = \"\xed\x9f\xbf\" |", "\x00"},
64  {"2.3.2 U-0000E000 = ee 80 80 = \"\xee\x80\x80\" |",
65  "2.3.2 U-0000E000 = ee 80 80 = \"\xee\x80\x80\" |", "\x00"},
66  {"2.3.3 U-0000FFFD = ef bf bd = \"\xef\xbf\xbd\" |",
67  "2.3.3 U-0000FFFD = ef bf bd = \"\xef\xbf\xbd\" |", "\x00"},
68  {"2.3.4 U-0010FFFF = f4 8f bf bf = \"\xf4\x8f\xbf\xbf\" |",
69  "2.3.4 U-0010FFFF = f4 8f bf bf = \"\xf4\x8f\xbf\xbf\" |", "\x00"},
70  {"2.3.5 U-00110000 = f4 90 80 80 = \"\xf4\x90\x80\x80\" |",
71  "2.3.5 U-00110000 = f4 90 80 80 = \"\xf4\x90\x80\x80\" |", "\x00"},
72 
73 /* 3 Malformed sequences
74  * 3.1 Unexpected continuation bytes
75  * Each unexpected continuation byte should be separately signaled as a malformed sequence of its own. */
76  {"3.1.1 First continuation byte 0x80: \"\x80\" |",
77  "3.1.1 First continuation byte 0x80: \"\" |", "\x01"},
78  {"3.1.2 Last continuation byte 0xbf: \"\xbf\" |",
79  "3.1.2 Last continuation byte 0xbf: \"\" |", "\x01"},
80  {"3.1.3 2 continuation bytes: \"\x80\xbf\" |",
81  "3.1.3 2 continuation bytes: \"\" |", "\x02"},
82  {"3.1.4 3 continuation bytes: \"\x80\xbf\x80\" |",
83  "3.1.4 3 continuation bytes: \"\" |", "\x03"},
84  {"3.1.5 4 continuation bytes: \"\x80\xbf\x80\xbf\" |",
85  "3.1.5 4 continuation bytes: \"\" |", "\x04"},
86  {"3.1.6 5 continuation bytes: \"\x80\xbf\x80\xbf\x80\" |",
87  "3.1.6 5 continuation bytes: \"\" |", "\x05"},
88  {"3.1.7 6 continuation bytes: \"\x80\xbf\x80\xbf\x80\xbf\" |",
89  "3.1.7 6 continuation bytes: \"\" |", "\x06"},
90  {"3.1.8 7 continuation bytes: \"\x80\xbf\x80\xbf\x80\xbf\x80\" |",
91  "3.1.8 7 continuation bytes: \"\" |", "\x07"},
92 /* 3.1.9 Sequence of all 64 possible continuation bytes (0x80-0xbf): | */
93  {"3.1.9 \"\x80\x81\x82\x83\x84\x85\x86\x87\x88\x89\x8a\x8b\x8c\x8d\x8e\x8f"
94  "\x90\x91\x92\x93\x94\x95\x96\x97\x98\x99\x9a\x9b\x9c\x9d\x9e\x9f"
95  "\xa0\xa1\xa2\xa3\xa4\xa5\xa6\xa7\xa8\xa9\xaa\xab\xac\xad\xae\xaf"
96  "\xb0\xb1\xb2\xb3\xb4\xb5\xb6\xb7\xb8\xb9\xba\xbb\xbc\xbd\xbe\xbf\" |",
97  "3.1.9 \"\" |", "\x40"}, /* NOLINT: modernize-raw-string-literal. */
98 /* 3.2 Lonely start characters
99  * 3.2.1 All 32 first bytes of 2-byte sequences (0xc0-0xdf), each followed by a space character: */
100  {"3.2.1 \"\xc0 \xc1 \xc2 \xc3 \xc4 \xc5 \xc6 \xc7 \xc8 \xc9 \xca \xcb \xcc \xcd \xce \xcf "
101  "\xd0 \xd1 \xd2 \xd3 \xd4 \xd5 \xd6 \xd7 \xd8 \xd9 \xda \xdb \xdc \xdd \xde \xdf \" |",
102  "3.2.1 \" \" |", "\x20"}, /* NOLINT: modernize-raw-string-literal. */
103 /* 3.2.2 All 16 first bytes of 3-byte sequences (0xe0-0xef), each followed by a space character: */
104  {"3.2.2 \"\xe0 \xe1 \xe2 \xe3 \xe4 \xe5 \xe6 \xe7 \xe8 \xe9 \xea \xeb \xec \xed \xee \xef \" |",
105  "3.2.2 \" \" |", "\x10"},
106 /* 3.2.3 All 8 first bytes of 4-byte sequences (0xf0-0xf7), each followed by a space character: */
107  {"3.2.3 \"\xf0 \xf1 \xf2 \xf3 \xf4 \xf5 \xf6 \xf7 \" |",
108  "3.2.3 \" \" |", "\x08"},
109 /* 3.2.4 All 4 first bytes of 5-byte sequences (0xf8-0xfb), each followed by a space character: */
110  {"3.2.4 \"\xf8 \xf9 \xfa \xfb \" |",
111  "3.2.4 \" \" |", "\x04"},
112 /* 3.2.5 All 2 first bytes of 6-byte sequences (0xfc-0xfd), each followed by a space character: */
113  {"3.2.4 \"\xfc \xfd \" |",
114  "3.2.4 \" \" |", "\x02"},
115 /* 3.3 Sequences with last continuation byte missing
116  * All bytes of an incomplete sequence should be signaled as a single malformed sequence,
117  * i.e., you should see only a single replacement character in each of the next 10 tests.
118  * (Characters as in section 2) */
119  {"3.3.1 2-byte sequence with last byte missing (U+0000): \"\xc0\" |",
120  "3.3.1 2-byte sequence with last byte missing (U+0000): \"\" |", "\x01"},
121  {"3.3.2 3-byte sequence with last byte missing (U+0000): \"\xe0\x80\" |",
122  "3.3.2 3-byte sequence with last byte missing (U+0000): \"\" |", "\x02"},
123  {"3.3.3 4-byte sequence with last byte missing (U+0000): \"\xf0\x80\x80\" |",
124  "3.3.3 4-byte sequence with last byte missing (U+0000): \"\" |", "\x03"},
125  {"3.3.4 5-byte sequence with last byte missing (U+0000): \"\xf8\x80\x80\x80\" |",
126  "3.3.4 5-byte sequence with last byte missing (U+0000): \"\" |", "\x04"},
127  {"3.3.5 6-byte sequence with last byte missing (U+0000): \"\xfc\x80\x80\x80\x80\" |",
128  "3.3.5 6-byte sequence with last byte missing (U+0000): \"\" |", "\x05"},
129  {"3.3.6 2-byte sequence with last byte missing (U-000007FF): \"\xdf\" |",
130  "3.3.6 2-byte sequence with last byte missing (U-000007FF): \"\" |", "\x01"},
131  {"3.3.7 3-byte sequence with last byte missing (U-0000FFFF): \"\xef\xbf\" |",
132  "3.3.7 3-byte sequence with last byte missing (U-0000FFFF): \"\" |", "\x02"},
133  {"3.3.8 4-byte sequence with last byte missing (U-001FFFFF): \"\xf7\xbf\xbf\" |",
134  "3.3.8 4-byte sequence with last byte missing (U-001FFFFF): \"\" |", "\x03"},
135  {"3.3.9 5-byte sequence with last byte missing (U-03FFFFFF): \"\xfb\xbf\xbf\xbf\" |",
136  "3.3.9 5-byte sequence with last byte missing (U-03FFFFFF): \"\" |", "\x04"},
137  {"3.3.10 6-byte sequence with last byte missing (U-7FFFFFFF): \"\xfd\xbf\xbf\xbf\xbf\" |",
138  "3.3.10 6-byte sequence with last byte missing (U-7FFFFFFF): \"\" |", "\x05"},
139 /* 3.4 Concatenation of incomplete sequences
140  * All the 10 sequences of 3.3 concatenated, you should see 10 malformed sequences being signaled: */
141  {"3.4 \"\xc0\xe0\x80\xf0\x80\x80\xf8\x80\x80\x80\xfc\x80\x80\x80\x80"
142  "\xdf\xef\xbf\xf7\xbf\xbf\xfb\xbf\xbf\xbf\xfd\xbf\xbf\xbf\xbf\""
143  " |",
144  "3.4 \"\" |", "\x1e"},
145 /* 3.5 Impossible bytes
146  * The following two bytes cannot appear in a correct UTF-8 string */
147  {"3.5.1 fe = \"\xfe\" |",
148  "3.5.1 fe = \"\" |", "\x01"},
149  {"3.5.2 ff = \"\xff\" |",
150  "3.5.2 ff = \"\" |", "\x01"},
151  {"3.5.3 fe fe ff ff = \"\xfe\xfe\xff\xff\" |",
152  "3.5.3 fe fe ff ff = \"\" |", "\x04"},
153 
154 /* 4 Overlong sequences
155  * The following sequences are not malformed according to the letter of the Unicode 2.0 standard.
156  * However, they are longer then necessary and a correct UTF-8 encoder is not allowed to produce them.
157  * A "safe UTF-8 decoder" should reject them just like malformed sequences for two reasons:
158  * (1) It helps to debug applications if overlong sequences are not treated as valid representations
159  * of characters, because this helps to spot problems more quickly. (2) Overlong sequences provide
160  * alternative representations of characters, that could maliciously be used to bypass filters that check
161  * only for ASCII characters. For instance, a 2-byte encoded line feed (LF) would not be caught by a
162  * line counter that counts only 0x0a bytes, but it would still be processed as a line feed by an unsafe
163  * UTF-8 decoder later in the pipeline. From a security point of view, ASCII compatibility of UTF-8
164  * sequences means also, that ASCII characters are *only* allowed to be represented by ASCII bytes
165  * in the range 0x00-0x7f. To ensure this aspect of ASCII compatibility, use only "safe UTF-8 decoders"
166  * that reject overlong UTF-8 sequences for which a shorter encoding exists.
167  *
168  * 4.1 Examples of an overlong ASCII character
169  * With a safe UTF-8 decoder, all of the following five overlong representations of the ASCII character
170  * slash ("/") should be rejected like a malformed UTF-8 sequence, for instance by substituting it with
171  * a replacement character. If you see a slash below, you do not have a safe UTF-8 decoder! */
172  {"4.1.1 U+002F = c0 af = \"\xc0\xaf\" |",
173  "4.1.1 U+002F = c0 af = \"\" |", "\x02"},
174  {"4.1.2 U+002F = e0 80 af = \"\xe0\x80\xaf\" |",
175  "4.1.2 U+002F = e0 80 af = \"\" |", "\x03"},
176  {"4.1.3 U+002F = f0 80 80 af = \"\xf0\x80\x80\xaf\" |",
177  "4.1.3 U+002F = f0 80 80 af = \"\" |", "\x04"},
178  {"4.1.4 U+002F = f8 80 80 80 af = \"\xf8\x80\x80\x80\xaf\" |",
179  "4.1.4 U+002F = f8 80 80 80 af = \"\" |", "\x05"},
180  {"4.1.5 U+002F = fc 80 80 80 80 af = \"\xfc\x80\x80\x80\x80\xaf\" |",
181  "4.1.5 U+002F = fc 80 80 80 80 af = \"\" |", "\x06"},
182 /* 4.2 Maximum overlong sequences
183  * Below you see the highest Unicode value that is still resulting in an overlong sequence if represented
184  * with the given number of bytes. This is a boundary test for safe UTF-8 decoders. All five characters
185  * should be rejected like malformed UTF-8 sequences. */
186  {"4.2.1 U-0000007F = c1 bf = \"\xc1\xbf\" |",
187  "4.2.1 U-0000007F = c1 bf = \"\" |", "\x02"},
188  {"4.2.2 U-000007FF = e0 9f bf = \"\xe0\x9f\xbf\" |",
189  "4.2.2 U-000007FF = e0 9f bf = \"\" |", "\x03"},
190  {"4.2.3 U-0000FFFF = f0 8f bf bf = \"\xf0\x8f\xbf\xbf\" |",
191  "4.2.3 U-0000FFFF = f0 8f bf bf = \"\" |", "\x04"},
192  {"4.2.4 U-001FFFFF = f8 87 bf bf bf = \"\xf8\x87\xbf\xbf\xbf\" |",
193  "4.2.4 U-001FFFFF = f8 87 bf bf bf = \"\" |", "\x05"},
194  {"4.2.5 U+0000 = fc 83 bf bf bf bf = \"\xfc\x83\xbf\xbf\xbf\xbf\" |",
195  "4.2.5 U+0000 = fc 83 bf bf bf bf = \"\" |", "\x06"},
196 /* 4.3 Overlong representation of the NUL character
197  * The following five sequences should also be rejected like malformed UTF-8 sequences and should not be
198  * treated like the ASCII NUL character. */
199  {"4.3.1 U+0000 = c0 80 = \"\xc0\x80\" |",
200  "4.3.1 U+0000 = c0 80 = \"\" |", "\x02"},
201  {"4.3.2 U+0000 = e0 80 80 = \"\xe0\x80\x80\" |",
202  "4.3.2 U+0000 = e0 80 80 = \"\" |", "\x03"},
203  {"4.3.3 U+0000 = f0 80 80 80 = \"\xf0\x80\x80\x80\" |",
204  "4.3.3 U+0000 = f0 80 80 80 = \"\" |", "\x04"},
205  {"4.3.4 U+0000 = f8 80 80 80 80 = \"\xf8\x80\x80\x80\x80\" |",
206  "4.3.4 U+0000 = f8 80 80 80 80 = \"\" |", "\x05"},
207  {"4.3.5 U+0000 = fc 80 80 80 80 80 = \"\xfc\x80\x80\x80\x80\x80\" |",
208  "4.3.5 U+0000 = fc 80 80 80 80 80 = \"\" |", "\x06"},
209 
210 /* 5 Illegal code positions
211  * The following UTF-8 sequences should be rejected like malformed sequences, because they never represent
212  * valid ISO 10646 characters and a UTF-8 decoder that accepts them might introduce security problems
213  * comparable to overlong UTF-8 sequences.
214  * 5.1 Single UTF-16 surrogates */
215  {"5.1.1 U+D800 = ed a0 80 = \"\xed\xa0\x80\" |",
216  "5.1.1 U+D800 = ed a0 80 = \"\" |", "\x03"},
217  {"5.1.2 U+DB7F = ed ad bf = \"\xed\xad\xbf\" |",
218  "5.1.2 U+DB7F = ed ad bf = \"\" |", "\x03"},
219  {"5.1.3 U+DB80 = ed ae 80 = \"\xed\xae\x80\" |",
220  "5.1.3 U+DB80 = ed ae 80 = \"\" |", "\x03"},
221  {"5.1.4 U+DBFF = ed af bf = \"\xed\xaf\xbf\" |",
222  "5.1.4 U+DBFF = ed af bf = \"\" |", "\x03"},
223  {"5.1.5 U+DC00 = ed b0 80 = \"\xed\xb0\x80\" |",
224  "5.1.5 U+DC00 = ed b0 80 = \"\" |", "\x03"},
225  {"5.1.6 U+DF80 = ed be 80 = \"\xed\xbe\x80\" |",
226  "5.1.6 U+DF80 = ed be 80 = \"\" |", "\x03"},
227  {"5.1.7 U+DFFF = ed bf bf = \"\xed\xbf\xbf\" |",
228  "5.1.7 U+DFFF = ed bf bf = \"\" |", "\x03"},
229 /* 5.2 Paired UTF-16 surrogates */
230  {"5.2.1 U+D800 U+DC00 = ed a0 80 ed b0 80 = \"\xed\xa0\x80\xed\xb0\x80\" |",
231  "5.2.1 U+D800 U+DC00 = ed a0 80 ed b0 80 = \"\" |", "\x06"},
232  {"5.2.2 U+D800 U+DFFF = ed a0 80 ed bf bf = \"\xed\xa0\x80\xed\xbf\xbf\" |",
233  "5.2.2 U+D800 U+DFFF = ed a0 80 ed bf bf = \"\" |", "\x06"},
234  {"5.2.3 U+DB7F U+DC00 = ed ad bf ed b0 80 = \"\xed\xad\xbf\xed\xb0\x80\" |",
235  "5.2.3 U+DB7F U+DC00 = ed ad bf ed b0 80 = \"\" |", "\x06"},
236  {"5.2.4 U+DB7F U+DFFF = ed ad bf ed bf bf = \"\xed\xad\xbf\xed\xbf\xbf\" |",
237  "5.2.4 U+DB7F U+DFFF = ed ad bf ed bf bf = \"\" |", "\x06"},
238  {"5.2.5 U+DB80 U+DC00 = ed ae 80 ed b0 80 = \"\xed\xae\x80\xed\xb0\x80\" |",
239  "5.2.5 U+DB80 U+DC00 = ed ae 80 ed b0 80 = \"\" |", "\x06"},
240  {"5.2.6 U+DB80 U+DFFF = ed ae 80 ed bf bf = \"\xed\xae\x80\xed\xbf\xbf\" |",
241  "5.2.6 U+DB80 U+DFFF = ed ae 80 ed bf bf = \"\" |", "\x06"},
242  {"5.2.7 U+DBFF U+DC00 = ed af bf ed b0 80 = \"\xed\xaf\xbf\xed\xb0\x80\" |",
243  "5.2.7 U+DBFF U+DC00 = ed af bf ed b0 80 = \"\" |", "\x06"},
244  {"5.2.8 U+DBFF U+DFFF = ed af bf ed bf bf = \"\xed\xaf\xbf\xed\xbf\xbf\" |",
245  "5.2.8 U+DBFF U+DFFF = ed af bf ed bf bf = \"\" |", "\x06"},
246 /* 5.3 Non-character code positions
247  * The following "non-characters" are "reserved for internal use" by applications, and according to older versions
248  * of the Unicode Standard "should never be interchanged". Unicode Corrigendum #9 dropped the latter restriction.
249  * Nevertheless, their presence in incoming UTF-8 data can remain a potential security risk, depending
250  * on what use is made of these codes subsequently. Examples of such internal use:
251  * - Some file APIs with 16-bit characters may use the integer value -1 = U+FFFF to signal
252  * an end-of-file (EOF) or error condition.
253  * - In some UTF-16 receivers, code point U+FFFE might trigger a byte-swap operation
254  * (to convert between UTF-16LE and UTF-16BE).
255  * With such internal use of non-characters, it may be desirable and safer to block those code points in
256  * UTF-8 decoders, as they should never occur legitimately in incoming UTF-8 data, and could trigger
257  * unsafe behavior in subsequent processing.
258  *
259  * Particularly problematic non-characters in 16-bit applications: */
260  {"5.3.1 U+FFFE = ef bf be = \"\xef\xbf\xbe\" |",
261  "5.3.1 U+FFFE = ef bf be = \"\" |", "\x03"},
262  {"5.3.2 U+FFFF = ef bf bf = \"\xef\xbf\xbf\" |",
263  "5.3.2 U+FFFF = ef bf bf = \"\" |", "\x03"},
264  /* For now, we ignore those, they do not seem to be crucial anyway... */
265 /* 5.3.3 U+FDD0 .. U+FDEF
266  * 5.3.4 U+nFFFE U+nFFFF (for n = 1..10) */
267  {nullptr, nullptr, nullptr},
268 };
269 /* clang-format on */
270 
271 /* BLI_str_utf8_invalid_strip (and indirectly, BLI_str_utf8_invalid_byte). */
272 TEST(string, Utf8InvalidBytes)
273 {
274  for (int i = 0; utf8_invalid_tests[i][0] != nullptr; i++) {
275  const char *tst = utf8_invalid_tests[i][0];
276  const char *tst_stripped = utf8_invalid_tests[i][1];
277  const int errors_num = (int)utf8_invalid_tests[i][2][0];
278 
279  char buff[80];
280  memcpy(buff, tst, sizeof(buff));
281 
282  const int errors_found_num = BLI_str_utf8_invalid_strip(buff, sizeof(buff) - 1);
283 
284  printf("[%02d] -> [%02d] \"%s\" -> \"%s\"\n", errors_num, errors_found_num, tst, buff);
285  EXPECT_EQ(errors_found_num, errors_num);
286  EXPECT_STREQ(buff, tst_stripped);
287  }
288 }
289 
292 /* -------------------------------------------------------------------- */
296 static size_t utf8_as_char32(const char *str, const char str_len, char32_t *r_result)
297 {
298  size_t i = 0, result_len = 0;
299  while ((i < str_len) && (str[i] != '\0')) {
300  char32_t c = BLI_str_utf8_as_unicode_step(str, str_len, &i);
301  if (c != BLI_UTF8_ERR) {
302  r_result[result_len++] = c;
303  }
304  }
305  return i;
306 }
307 
308 template<size_t Size, size_t SizeWithPadding>
309 void utf8_as_char32_test_compare_with_pad_bytes(const char utf8_src[Size])
310 {
311  char utf8_src_with_pad[SizeWithPadding] = {0};
312 
313  memcpy(utf8_src_with_pad, utf8_src, Size);
314 
315  char32_t unicode_dst_a[Size], unicode_dst_b[Size];
316 
317  memset(unicode_dst_a, 0xff, sizeof(unicode_dst_a));
318  const size_t index_a = utf8_as_char32(utf8_src, Size, unicode_dst_a);
319 
320  /* Test with padded and un-padded size,
321  * to ensure that extra available space doesn't yield a different result. */
322  for (int pass = 0; pass < 2; pass++) {
323  memset(unicode_dst_b, 0xff, sizeof(unicode_dst_b));
324  const size_t index_b = utf8_as_char32(
325  utf8_src_with_pad, pass ? Size : SizeWithPadding, unicode_dst_b);
326 
327  /* Check the resulting content matches. */
328  EXPECT_EQ_ARRAY(unicode_dst_a, unicode_dst_b, Size);
329  /* Check the index of the source strings match. */
330  EXPECT_EQ(index_a, index_b);
331  }
332 }
333 
334 template<size_t Size> void utf8_as_char32_test_compare(const char utf8_src[Size])
335 {
336  /* Note that 7 is a little arbitrary,
337  * chosen since it's the maximum length of multi-byte character + 1
338  * to account for any errors that read past null bytes. */
339  utf8_as_char32_test_compare_with_pad_bytes<Size, Size + 1>(utf8_src);
340  utf8_as_char32_test_compare_with_pad_bytes<Size, Size + 7>(utf8_src);
341 }
342 
343 template<size_t Size> void utf8_as_char32_test_at_buffer_size()
344 {
345  char utf8_src[Size];
346 
347  /* Test uniform bytes, also with offsets ascending & descending. */
348  for (int i = 0; i <= 0xff; i++) {
349  memset(utf8_src, i, sizeof(utf8_src));
350  utf8_as_char32_test_compare<Size>(utf8_src);
351 
352  /* Offset trailing bytes up and down in steps of 1, 2, 4 .. etc. */
353  if (Size > 1) {
354  for (int mul = 1; mul < 256; mul *= 2) {
355  for (int ofs = 1; ofs < (int)Size; ofs++) {
356  utf8_src[ofs] = (char)(i + (ofs * mul));
357  }
358  utf8_as_char32_test_compare<Size>(utf8_src);
359 
360  for (int ofs = 1; ofs < (int)Size; ofs++) {
361  utf8_src[ofs] = (char)(i - (ofs * mul));
362  }
363  utf8_as_char32_test_compare<Size>(utf8_src);
364  }
365  }
366  }
367 
368  /* Random bytes. */
369  RNG *rng = BLI_rng_new(1);
370  for (int i = 0; i < 256; i++) {
371  BLI_rng_get_char_n(rng, utf8_src, sizeof(utf8_src));
372  utf8_as_char32_test_compare<Size>(utf8_src);
373  }
374  BLI_rng_free(rng);
375 }
376 
377 TEST(string, Utf8AsUnicodeStep)
378 {
379 
380  /* Run tests at different buffer sizes. */
381  utf8_as_char32_test_at_buffer_size<1>();
382  utf8_as_char32_test_at_buffer_size<2>();
383  utf8_as_char32_test_at_buffer_size<3>();
384  utf8_as_char32_test_at_buffer_size<4>();
385  utf8_as_char32_test_at_buffer_size<5>();
386  utf8_as_char32_test_at_buffer_size<6>();
387  utf8_as_char32_test_at_buffer_size<7>();
388  utf8_as_char32_test_at_buffer_size<8>();
389  utf8_as_char32_test_at_buffer_size<9>();
390  utf8_as_char32_test_at_buffer_size<10>();
391  utf8_as_char32_test_at_buffer_size<11>();
392  utf8_as_char32_test_at_buffer_size<12>();
393 }
394 
EXPECT_EQ(BLI_expr_pylike_eval(expr, nullptr, 0, &result), EXPR_PYLIKE_INVALID)
Random number functions.
void BLI_rng_free(struct RNG *rng) ATTR_NONNULL(1)
Definition: rand.cc:58
struct RNG * BLI_rng_new(unsigned int seed)
Definition: rand.cc:39
void BLI_rng_get_char_n(RNG *rng, char *bytes, size_t bytes_len) ATTR_NONNULL(1
unsigned int BLI_str_utf8_as_unicode_step(const char *__restrict p, size_t p_len, size_t *__restrict index) ATTR_WARN_UNUSED_RESULT ATTR_NONNULL(1
#define BLI_UTF8_ERR
int BLI_str_utf8_invalid_strip(char *str, size_t length) ATTR_NONNULL(1)
Definition: string_utf8.c:181
void utf8_as_char32_test_at_buffer_size()
static size_t utf8_as_char32(const char *str, const char str_len, char32_t *r_result)
void utf8_as_char32_test_compare_with_pad_bytes(const char utf8_src[Size])
void utf8_as_char32_test_compare(const char utf8_src[Size])
TEST(string, Utf8InvalidBytes)
static const char * utf8_invalid_tests[][3]
static void mul(btAlignedObjectArray< T > &items, const Q &value)
#define str(s)
static unsigned c
Definition: RandGen.cpp:83
Definition: rand.cc:33