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