]>
Commit | Line | Data |
---|---|---|
f3c0d7a5 A |
1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html | |
73c04bcf A |
3 | /* |
4 | ******************************************************************************* | |
5 | * | |
57a6839d | 6 | * Copyright (C) 2005-2014, International Business Machines |
73c04bcf A |
7 | * Corporation and others. All Rights Reserved. |
8 | * | |
9 | ******************************************************************************* | |
10 | * | |
11 | * created on: 2005jun15 | |
12 | * created by: Raymond Yang | |
13 | */ | |
14 | ||
3d1f044b A |
15 | #include "unicode/utypes.h" |
16 | ||
17 | #if !UCONFIG_NO_IDNA | |
73c04bcf A |
18 | |
19 | #include <stdio.h> | |
20 | #include <stdlib.h> | |
21 | #include <string.h> | |
73c04bcf A |
22 | #include "unicode/ucnv.h" |
23 | #include "unicode/ustring.h" | |
24 | #include "unicode/uidna.h" | |
0f5d89e8 | 25 | #include "unicode/utf16.h" |
73c04bcf A |
26 | #include "idnaconf.h" |
27 | ||
28 | static const UChar C_TAG[] = {0x3D, 0x3D, 0x3D, 0x3D, 0x3D, 0}; // ===== | |
29 | static const UChar C_NAMEZONE[] = {0x6E, 0x61, 0x6D, 0x65, 0x7A, 0x6F, 0x6E, 0x65, 0}; // namezone | |
30 | static const UChar C_NAMEBASE[] = {0x6E, 0x61, 0x6D, 0x65, 0x62, 0x61, 0x73, 0x65, 0}; // namebase | |
73c04bcf A |
31 | |
32 | static const UChar C_TYPE[] = {0x74, 0x79, 0x70, 0x65, 0}; // type | |
33 | static const UChar C_TOASCII[] = {0x74, 0x6F, 0x61, 0x73, 0x63, 0x69, 0x69, 0}; // toascii | |
34 | static const UChar C_TOUNICODE[] = {0x74, 0x6F, 0x75, 0x6E, 0x69, 0x63, 0x6F, 0x64, 0x65, 0}; // tounicode | |
35 | ||
36 | static const UChar C_PASSFAIL[] = {0x70, 0x61, 0x73, 0x73, 0x66, 0x61, 0x69, 0x6C, 0}; // passfail | |
37 | static const UChar C_PASS[] = {0x70, 0x61, 0x73, 0x73, 0}; // pass | |
38 | static const UChar C_FAIL[] = {0x66, 0x61, 0x69, 0x6C, 0}; // fail | |
39 | ||
40 | static const UChar C_DESC[] = {0x64, 0x65, 0x73, 0x63, 0}; // desc | |
41 | static const UChar C_USESTD3ASCIIRULES[] = {0x55, 0x73, 0x65, 0x53, 0x54, 0x44, | |
42 | 0x33, 0x41, 0x53, 0x43, 0x49, 0x49, 0x52, 0x75, 0x6C, 0x65, 0x73, 0}; // UseSTD3ASCIIRules | |
43 | ||
44 | IdnaConfTest::IdnaConfTest(){ | |
45 | base = NULL; | |
46 | len = 0; | |
47 | curOffset = 0; | |
48 | ||
49 | type = option = passfail = -1; | |
50 | namebase.setToBogus(); | |
51 | namezone.setToBogus(); | |
52 | } | |
53 | IdnaConfTest::~IdnaConfTest(){ | |
54 | delete [] base; | |
55 | } | |
56 | ||
57 | #if !UCONFIG_NO_IDNA | |
58 | /* this function is modified from RBBITest::ReadAndConvertFile() | |
59 | * | |
60 | */ | |
61 | UBool IdnaConfTest::ReadAndConvertFile(){ | |
62 | ||
63 | char * source = NULL; | |
64 | size_t source_len; | |
65 | ||
66 | // read the test data file to memory | |
67 | FILE* f = NULL; | |
68 | UErrorCode status = U_ZERO_ERROR; | |
69 | ||
70 | const char *path = IntlTest::getSourceTestData(status); | |
71 | if (U_FAILURE(status)) { | |
72 | errln("%s", u_errorName(status)); | |
73 | return FALSE; | |
74 | } | |
75 | ||
76 | const char* name = "idna_conf.txt"; // test data file | |
3d1f044b | 77 | int t = static_cast<int>(strlen(path) + strlen(name) + 1); |
73c04bcf A |
78 | char* absolute_name = new char[t]; |
79 | strcpy(absolute_name, path); | |
80 | strcat(absolute_name, name); | |
81 | f = fopen(absolute_name, "rb"); | |
82 | delete [] absolute_name; | |
83 | ||
84 | if (f == NULL){ | |
729e4ab9 | 85 | dataerrln("fopen error on %s", name); |
73c04bcf A |
86 | return FALSE; |
87 | } | |
88 | ||
89 | fseek( f, 0, SEEK_END); | |
90 | if ((source_len = ftell(f)) <= 0){ | |
91 | errln("Error reading test data file."); | |
92 | fclose(f); | |
93 | return FALSE; | |
94 | } | |
95 | ||
96 | source = new char[source_len]; | |
97 | fseek(f, 0, SEEK_SET); | |
98 | if (fread(source, 1, source_len, f) != source_len) { | |
99 | errln("Error reading test data file."); | |
100 | delete [] source; | |
101 | fclose(f); | |
102 | return FALSE; | |
103 | } | |
104 | fclose(f); | |
105 | ||
106 | // convert the UTF-8 encoded stream to UTF-16 stream | |
107 | UConverter* conv = ucnv_open("utf-8", &status); | |
108 | int dest_len = ucnv_toUChars(conv, | |
109 | NULL, // dest, | |
110 | 0, // destCapacity, | |
111 | source, | |
3d1f044b | 112 | static_cast<int32_t>(source_len), |
73c04bcf A |
113 | &status); |
114 | if (status == U_BUFFER_OVERFLOW_ERROR) { | |
115 | // Buffer Overflow is expected from the preflight operation. | |
116 | status = U_ZERO_ERROR; | |
117 | UChar * dest = NULL; | |
118 | dest = new UChar[ dest_len + 1]; | |
3d1f044b | 119 | ucnv_toUChars(conv, dest, dest_len + 1, source, static_cast<int32_t>(source_len), &status); |
73c04bcf A |
120 | // Do not know the "if possible" behavior of ucnv_toUChars() |
121 | // Do it by ourself. | |
122 | dest[dest_len] = 0; | |
123 | len = dest_len; | |
124 | base = dest; | |
125 | delete [] source; | |
126 | ucnv_close(conv); | |
127 | return TRUE; // The buffer will owned by caller. | |
128 | } | |
129 | errln("UConverter error: %s", u_errorName(status)); | |
130 | delete [] source; | |
131 | ucnv_close(conv); | |
132 | return FALSE; | |
133 | } | |
134 | ||
135 | int IdnaConfTest::isNewlineMark(){ | |
136 | static const UChar LF = 0x0a; | |
137 | static const UChar CR = 0x0d; | |
138 | UChar c = base[curOffset]; | |
139 | // CR LF | |
140 | if ( c == CR && curOffset + 1 < len && base[curOffset + 1] == LF){ | |
141 | return 2; | |
142 | } | |
143 | ||
144 | // CR or LF | |
145 | if ( c == CR || c == LF) { | |
146 | return 1; | |
147 | } | |
148 | ||
149 | return 0; | |
150 | } | |
151 | ||
152 | /* Read a logical line. | |
153 | * | |
154 | * All lines ending in a backslash (\) and immediately followed by a newline | |
155 | * character are joined with the next line in the source file forming logical | |
156 | * lines from the physical lines. | |
157 | * | |
158 | */ | |
159 | UBool IdnaConfTest::ReadOneLine(UnicodeString& buf){ | |
160 | if ( !(curOffset < len) ) return FALSE; // stream end | |
161 | ||
162 | static const UChar BACKSLASH = 0x5c; | |
163 | buf.remove(); | |
164 | int t = 0; | |
165 | while (curOffset < len){ | |
166 | if ((t = isNewlineMark())) { // end of line | |
167 | curOffset += t; | |
168 | break; | |
169 | } | |
170 | UChar c = base[curOffset]; | |
171 | if (c == BACKSLASH && curOffset < len -1){ // escaped new line mark | |
172 | if ((t = isNewlineMark())){ | |
173 | curOffset += 1 + t; // BACKSLAH and NewlineMark | |
174 | continue; | |
175 | } | |
176 | }; | |
177 | buf.append(c); | |
178 | curOffset++; | |
179 | } | |
180 | return TRUE; | |
181 | } | |
182 | ||
183 | // | |
184 | //=============================================================== | |
185 | // | |
186 | ||
187 | /* Explain <xxxxx> tag to a native value | |
188 | * | |
189 | * Since <xxxxx> is always larger than the native value, | |
190 | * the operation will replace the tag directly in the buffer, | |
191 | * and, of course, will shift tail elements. | |
192 | */ | |
193 | void IdnaConfTest::ExplainCodePointTag(UnicodeString& buf){ | |
194 | buf.append((UChar)0); // add a terminal NULL | |
195 | UChar* bufBase = buf.getBuffer(buf.length()); | |
196 | UChar* p = bufBase; | |
197 | while (*p != 0){ | |
198 | if ( *p != 0x3C){ // < | |
199 | *bufBase++ = *p++; | |
200 | } else { | |
201 | p++; // skip < | |
202 | UChar32 cp = 0; | |
203 | for ( ;*p != 0x3E; p++){ // > | |
204 | if (0x30 <= *p && *p <= 0x39){ // 0-9 | |
205 | cp = (cp * 16) + (*p - 0x30); | |
206 | } else if (0x61 <= *p && *p <= 0x66){ // a-f | |
207 | cp = (cp * 16) + (*p - 0x61) + 10; | |
208 | } else if (0x41 <= *p && *p <= 0x46) {// A-F | |
209 | cp = (cp * 16) + (*p - 0x41) + 10; | |
210 | } | |
211 | // no else. hope everything is good. | |
212 | } | |
213 | p++; // skip > | |
214 | if (U_IS_BMP(cp)){ | |
215 | *bufBase++ = cp; | |
216 | } else { | |
217 | *bufBase++ = U16_LEAD(cp); | |
218 | *bufBase++ = U16_TRAIL(cp); | |
219 | } | |
220 | } | |
221 | } | |
222 | *bufBase = 0; // close our buffer | |
223 | buf.releaseBuffer(); | |
224 | } | |
225 | ||
226 | void IdnaConfTest::Call(){ | |
227 | if (type == -1 || option == -1 || passfail == -1 || namebase.isBogus() || namezone.isBogus()){ | |
228 | errln("Incomplete record"); | |
229 | } else { | |
230 | UErrorCode status = U_ZERO_ERROR; | |
231 | UChar result[200] = {0,}; // simple life | |
232 | const UChar *p = namebase.getTerminatedBuffer(); | |
233 | const int p_len = namebase.length(); | |
234 | ||
235 | if (type == 0 && option == 0){ | |
236 | uidna_IDNToASCII(p, p_len, result, 200, UIDNA_USE_STD3_RULES, NULL, &status); | |
237 | } else if (type == 0 && option == 1){ | |
238 | uidna_IDNToASCII(p, p_len, result, 200, UIDNA_ALLOW_UNASSIGNED, NULL, &status); | |
239 | } else if (type == 1 && option == 0){ | |
240 | uidna_IDNToUnicode(p, p_len, result, 200, UIDNA_USE_STD3_RULES, NULL, &status); | |
241 | } else if (type == 1 && option == 1){ | |
242 | uidna_IDNToUnicode(p, p_len, result, 200, UIDNA_ALLOW_UNASSIGNED, NULL, &status); | |
243 | } | |
244 | if (passfail == 0){ | |
245 | if (U_FAILURE(status)){ | |
729e4ab9 A |
246 | id.append(" should pass, but failed. - "); |
247 | id.append(u_errorName(status)); | |
248 | errcheckln(status, id); | |
73c04bcf A |
249 | } else{ |
250 | if (namezone.compare(result, -1) == 0){ | |
251 | // expected | |
252 | logln(UnicodeString("namebase: ") + prettify(namebase) + UnicodeString(" result: ") + prettify(result)); | |
253 | } else { | |
254 | id.append(" no error, but result is not as expected."); | |
255 | errln(id); | |
256 | } | |
257 | } | |
258 | } else if (passfail == 1){ | |
259 | if (U_FAILURE(status)){ | |
260 | // expected | |
261 | // TODO: Uncomment this when U_IDNA_ZERO_LENGTH_LABEL_ERROR is added to u_errorName | |
262 | //logln("Got the expected error: " + UnicodeString(u_errorName(status))); | |
263 | } else{ | |
264 | if (namebase.compare(result, -1) == 0){ | |
265 | // garbage in -> garbage out | |
266 | logln(UnicodeString("ICU will not recognize malformed ACE-Prefixes or incorrect ACE-Prefixes. ") + UnicodeString("namebase: ") + prettify(namebase) + UnicodeString(" result: ") + prettify(result)); | |
267 | } else { | |
268 | id.append(" should fail, but not failed. "); | |
269 | id.append(u_errorName(status)); | |
270 | errln(id); | |
271 | } | |
272 | } | |
273 | } | |
274 | } | |
275 | type = option = passfail = -1; | |
276 | namebase.setToBogus(); | |
277 | namezone.setToBogus(); | |
278 | id.remove(); | |
279 | return; | |
280 | } | |
281 | ||
282 | void IdnaConfTest::Test(void){ | |
283 | if (!ReadAndConvertFile())return; | |
284 | ||
285 | UnicodeString s; | |
286 | UnicodeString key; | |
287 | UnicodeString value; | |
288 | ||
289 | // skip everything before the first "=====" and "=====" itself | |
290 | do { | |
291 | if (!ReadOneLine(s)) { | |
292 | errln("End of file prematurely found"); | |
293 | break; | |
294 | } | |
295 | } | |
296 | while (s.compare(C_TAG, -1) != 0); //"=====" | |
297 | ||
298 | while(ReadOneLine(s)){ | |
299 | s.trim(); | |
300 | key.remove(); | |
301 | value.remove(); | |
302 | if (s.compare(C_TAG, -1) == 0){ //"=====" | |
303 | Call(); | |
304 | } else { | |
305 | // explain key:value | |
306 | int p = s.indexOf((UChar)0x3A); // : | |
307 | key.setTo(s,0,p).trim(); | |
308 | value.setTo(s,p+1).trim(); | |
309 | if (key.compare(C_TYPE, -1) == 0){ | |
310 | if (value.compare(C_TOASCII, -1) == 0) { | |
311 | type = 0; | |
312 | } else if (value.compare(C_TOUNICODE, -1) == 0){ | |
313 | type = 1; | |
314 | } | |
315 | } else if (key.compare(C_PASSFAIL, -1) == 0){ | |
316 | if (value.compare(C_PASS, -1) == 0){ | |
317 | passfail = 0; | |
318 | } else if (value.compare(C_FAIL, -1) == 0){ | |
319 | passfail = 1; | |
320 | } | |
321 | } else if (key.compare(C_DESC, -1) == 0){ | |
322 | if (value.indexOf(C_USESTD3ASCIIRULES, u_strlen(C_USESTD3ASCIIRULES), 0) == -1){ | |
323 | option = 1; // not found | |
324 | } else { | |
325 | option = 0; | |
326 | } | |
327 | id.setTo(value, 0, value.indexOf((UChar)0x20)); // space | |
328 | } else if (key.compare(C_NAMEZONE, -1) == 0){ | |
329 | ExplainCodePointTag(value); | |
330 | namezone.setTo(value); | |
331 | } else if (key.compare(C_NAMEBASE, -1) == 0){ | |
332 | ExplainCodePointTag(value); | |
333 | namebase.setTo(value); | |
334 | } | |
335 | // just skip other lines | |
336 | } | |
337 | } | |
338 | ||
339 | Call(); // for last record | |
340 | } | |
341 | #else | |
342 | void IdnaConfTest::Test(void) | |
343 | { | |
344 | // test nothing... | |
345 | } | |
346 | #endif | |
347 | ||
348 | void IdnaConfTest::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/){ | |
349 | switch (index) { | |
350 | TESTCASE(0,Test); | |
351 | default: name = ""; break; | |
352 | } | |
353 | } | |
354 | ||
355 | #endif |