1 // Copyright 2017 PDFium Authors. All rights reserved.
2 // Use of this source code is governed by a BSD-style license that can be
3 // found in the LICENSE file.
4
5 // Original code copyright 2014 Foxit Software Inc. http://www.foxitsoftware.com
6
7 #include "xfa/fgas/crt/cfgas_stringformatter.h"
8
9 #include <time.h>
10
11 #include <memory>
12
13 #include "build/build_config.h"
14 #include "core/fpdfapi/page/cpdf_pagemodule.h"
15 #include "testing/fx_string_testhelpers.h"
16 #include "testing/gtest/include/gtest/gtest.h"
17 #include "third_party/base/ptr_util.h"
18 #include "xfa/fxfa/parser/cxfa_localemgr.h"
19
20 class CFGAS_StringFormatterTest : public testing::Test {
21 public:
CFGAS_StringFormatterTest()22 CFGAS_StringFormatterTest() {
23 SetTZ("UTC");
24 CPDF_PageModule::Create();
25 }
26
~CFGAS_StringFormatterTest()27 ~CFGAS_StringFormatterTest() override { CPDF_PageModule::Destroy(); }
28
TearDown()29 void TearDown() override {
30 fmt_.reset();
31 mgr_.reset();
32 }
33
SetTZ(const char * tz)34 void SetTZ(const char* tz) {
35 #if defined(OS_WIN)
36 _putenv_s("TZ", tz);
37 _tzset();
38 #else
39 setenv("TZ", tz, 1);
40 tzset();
41 #endif
42 }
43
44 // Note, this re-creates the fmt on each call. If you need to multiple
45 // times store it locally.
fmt(const WideString & locale,const WideString & pattern)46 CFGAS_StringFormatter* fmt(const WideString& locale,
47 const WideString& pattern) {
48 fmt_.reset(); // Can't outlive |mgr_|.
49 mgr_ = pdfium::MakeUnique<CXFA_LocaleMgr>(nullptr, locale);
50 fmt_ = pdfium::MakeUnique<CFGAS_StringFormatter>(mgr_.get(), pattern);
51 return fmt_.get();
52 }
53
54 protected:
55 std::unique_ptr<CXFA_LocaleMgr> mgr_;
56 std::unique_ptr<CFGAS_StringFormatter> fmt_;
57 };
58
59 // TODO(dsinclair): Looks like the formatter/parser does not handle the various
60 // 'g' flags.
TEST_F(CFGAS_StringFormatterTest,DateFormat)61 TEST_F(CFGAS_StringFormatterTest, DateFormat) {
62 struct {
63 const wchar_t* locale;
64 const wchar_t* input;
65 const wchar_t* pattern;
66 const wchar_t* output;
67 } tests[] = {
68 {L"en", L"2002-10-25", L"MMMM DD, YYYY", L"October 25, 2002"},
69 // Note, this is in the doc as 5 but it's wrong and should be 3 by the
70 // example in the Picture Clause Reference section.
71 {L"en", L"20040722", L"'Week of the month is' w",
72 L"Week of the month is 3"},
73 {L"en", L"20040722", L"e 'days after Sunday'", L"4 days after Sunday"},
74 {L"en", L"20040722", L"YYYY-'W'WW-e", L"2004-W30-4"},
75 {L"en", L"20040722", L"E 'days after Saturday'",
76 L"5 days after Saturday"},
77 {L"en", L"2000-01-01", L"EEE, 'the' D 'of' MMMM, YYYY",
78 L"Sat, the 1 of January, 2000"},
79 {L"en", L"2000-01-01", L"EEEE, 'the' D 'of' MMMM, YYYY",
80 L"Saturday, the 1 of January, 2000"},
81 {L"en", L"19991202", L"MM/D/YY", L"12/2/99"},
82 {L"en", L"19990110", L"MMM D, YYYY", L"Jan 10, 1999"},
83 {L"en", L"19990202", L"J", L"33"},
84 {L"en", L"19990202", L"JJJ", L"033"},
85 {L"en", L"19991231", L"J", L"365"},
86 {L"en", L"20001231", L"J", L"366"},
87 {L"en", L"19990501", L"J", L"121"},
88 {L"en", L"19990901", L"J", L"244"},
89 {L"en", L"19990228", L"J", L"59"},
90 {L"en", L"20000229", L"J", L"60"},
91 {L"en", L"21000501", L"J", L"121"},
92 {L"en", L"19990102", L"M", L"1"},
93 {L"en", L"19990102", L"MMM", L"Jan"},
94 {L"en", L"19990102", L"YYYY G", L"1999 AD"},
95 // Week 01 of the year is the week containing Jan 04.
96 // {L"en", L"19990102", L"WW", L"00"}, -- Returns 01 incorrectly
97 // {L"en", L"19990104", L"WW", L"01"}, -- Returns 02 incorrectly
98 // The ?*+ should format as whitespace.
99 // {L"en", L"19990104", L"YYYY?*+MM", L"1999 01"},
100 // {L"en", L"1999-07-16", L"date{DD/MM/YY} '('date{MMM DD, YYYY}')'",
101 // L"16/07/99 (Jul 16, 1999)"},
102 {L"de_CH", L"20041030", L"D. MMMM YYYY", L"30. Oktober 2004"},
103 {L"fr_CA", L"20041030", L"D MMMM YYYY", L"30 octobre 2004"},
104 {L"en", L"2002-10-25", L"date(fr){DD MMMM, YYYY}", L"25 octobre, 2002"},
105 {L"en", L"2002-10-25", L"date(es){EEEE, D 'de' MMMM 'de' YYYY}",
106 L"viernes, 25 de octubre de 2002"},
107 // {L"en", L"2002-20-25", L"date.long(fr)()", L"25 octobre, 2002"},
108 // {L"ja", L"2003-11-03", L"gY/M/D", L"H15/11/3"},
109 // {L"ja", L"1989-01-08", L"ggY-M-D", L"\u5e731-1-8"},
110 // {L"ja", L"1989-11-03", L"gggYY/MM/DD", L"\u5e73\u621089/11/03"},
111 };
112 // Note, none of the full width date symbols are listed here
113 // as they are not supported. In theory there are the full width versions
114 // of DDD, DDDD, MMM, MMMM, E, e, gg, YYY, YYYYY.
115
116 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
117 WideString result;
118 EXPECT_TRUE(
119 fmt(tests[i].locale, tests[i].pattern)
120 ->FormatDateTime(tests[i].input, FX_DATETIMETYPE_Date, &result));
121 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
122 }
123 }
124
TEST_F(CFGAS_StringFormatterTest,TimeFormat)125 TEST_F(CFGAS_StringFormatterTest, TimeFormat) {
126 struct {
127 const wchar_t* locale;
128 const wchar_t* input;
129 const wchar_t* pattern;
130 const wchar_t* output;
131 } tests[] = {{L"en", L"01:01:11", L"h:M A", L"1:1 AM"},
132 {L"en", L"13:01:11", L"h:M A", L"1:1 PM"},
133 {L"en", L"01:01:11", L"hh:MM:SS A", L"01:01:11 AM"},
134 {L"en", L"13:01:11", L"hh:MM:SS A", L"01:01:11 PM"},
135 {L"en", L"01:01:11", L"hh:MM:SS A Z", L"01:01:11 AM GMT-02:00"},
136 {L"en", L"01:01:11", L"hh:MM:SS A z", L"01:01:11 AM -02:00"},
137 // {L"en", L"01:01:11", L"hh:MM:SS A zz", L"01:01:11 AM GMT"},
138 // Should change ?*+ into ' ' when formatting.
139 // {L"en", L"01:01:11", L"hh:MM:SS?*+A", L"01:01:11 AM"},
140 {L"en", L"12:01:01", L"k:MM:SS", L"12:01:01"},
141 {L"en", L"14:01:01", L"k:MM:SS", L"2:01:01"},
142 {L"en", L"12:01:11", L"kk:MM", L"12:01"},
143 {L"en", L"14:01:11", L"kk:MM", L"02:01"},
144 {L"en", L"12:01:11 +04:30", L"kk:MM", L"05:31"},
145 {L"en", L"12:01:11", L"kk:MM A", L"12:01 PM"},
146 {L"en", L"00:01:01", L"H:M:S", L"0:1:1"},
147 {L"en", L"13:02:11", L"H:M:S", L"13:2:11"},
148 {L"en", L"00:01:11.001", L"HH:M:S.FFF", L"00:1:11.001"},
149 {L"en", L"13:02:11", L"HH:M", L"13:2"},
150 {L"en", L"00:01:11", L"K:M", L"24:1"},
151 {L"en", L"00:02:11", L"KK:M", L"24:2"},
152 {L"en", L"11:11:11", L"HH:MM:SS 'o''clock' A Z",
153 L"11:11:11 o'clock AM GMT-02:00"},
154 {L"en", L"14:30:59", L"h:MM A", L"2:30 PM"},
155 {L"en", L"14:30:59", L"HH:MM:SS A Z", L"14:30:59 PM GMT-02:00"}};
156 // Note, none of the full width time symbols are listed here
157 // as they are not supported. In theory there are the full
158 // width versions of kkk, kkkk, HHH, HHHH, KKK, KKKK, MMM, MMMM,
159 // SSS, SSSS plus 2 more that the spec apparently forgot to
160 // list the symbol.
161
162 // The z modifier only appends if the TZ is outside of +0
163 SetTZ("UTC+2");
164
165 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
166 WideString result;
167 EXPECT_TRUE(
168 fmt(tests[i].locale, tests[i].pattern)
169 ->FormatDateTime(tests[i].input, FX_DATETIMETYPE_Time, &result));
170 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
171 }
172
173 SetTZ("UTC");
174 }
175
TEST_F(CFGAS_StringFormatterTest,DateTimeFormat)176 TEST_F(CFGAS_StringFormatterTest, DateTimeFormat) {
177 struct {
178 const wchar_t* locale;
179 const wchar_t* input;
180 const wchar_t* pattern;
181 const wchar_t* output;
182 } tests[] = {
183 {L"en", L"1999-07-16T10:30Z",
184 L"'At' time{HH:MM Z} 'on' date{MMM DD, YYYY}",
185 L"At 10:30 GMT on Jul 16, 1999"},
186 {L"en", L"1999-07-16T10:30", L"'At' time{HH:MM} 'on' date{MMM DD, YYYY}",
187 L"At 10:30 on Jul 16, 1999"},
188 {L"en", L"1999-07-16T10:30Z",
189 L"time{'At' HH:MM Z} date{'on' MMM DD, YYYY}",
190 L"At 10:30 GMT on Jul 16, 1999"},
191 {L"en", L"1999-07-16T10:30Z",
192 L"time{'At 'HH:MM Z}date{' on 'MMM DD, YYYY}",
193 L"At 10:30 GMT on Jul 16, 1999"},
194 {L"en", L"9111T1111:", L"MMM D, YYYYTh:MM:SS A",
195 L"Jan 1, 9111 11:11:00 AM"}};
196
197 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
198 WideString result;
199 EXPECT_TRUE(fmt(tests[i].locale, tests[i].pattern)
200 ->FormatDateTime(tests[i].input, FX_DATETIMETYPE_DateTime,
201 &result));
202 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
203 }
204 }
205
TEST_F(CFGAS_StringFormatterTest,TimeDateFormat)206 TEST_F(CFGAS_StringFormatterTest, TimeDateFormat) {
207 struct {
208 const wchar_t* locale;
209 const wchar_t* input;
210 const wchar_t* pattern;
211 const wchar_t* output;
212 } tests[] = {
213 {L"en", L"1999-07-16T10:30Z",
214 L"'At' time{HH:MM Z} 'on' date{MMM DD, YYYY}",
215 L"At 10:30 GMT on Jul 16, 1999"},
216 {L"en", L"1999-07-16T10:30", L"'At' time{HH:MM} 'on' date{MMM DD, YYYY}",
217 L"At 10:30 on Jul 16, 1999"},
218 {L"en", L"1999-07-16T10:30Z",
219 L"time{'At' HH:MM Z} date{'on' MMM DD, YYYY}",
220 L"At 10:30 GMT on Jul 16, 1999"},
221 {L"en", L"1999-07-16T10:30Z",
222 L"time{'At 'HH:MM Z}date{' on 'MMM DD, YYYY}",
223 L"At 10:30 GMT on Jul 16, 1999"}};
224
225 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
226 WideString result;
227 EXPECT_TRUE(fmt(tests[i].locale, tests[i].pattern)
228 ->FormatDateTime(tests[i].input, FX_DATETIMETYPE_TimeDate,
229 &result));
230 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
231 }
232 }
233
TEST_F(CFGAS_StringFormatterTest,DateParse)234 TEST_F(CFGAS_StringFormatterTest, DateParse) {
235 struct {
236 const wchar_t* locale;
237 const wchar_t* input;
238 const wchar_t* pattern;
239 CFX_DateTime output;
240 } tests[] = {
241 {L"en", L"12/2/99", L"MM/D/YY", CFX_DateTime(1999, 12, 2, 0, 0, 0, 0)},
242 {L"en", L"2/2/99", L"M/D/YY", CFX_DateTime(1999, 2, 2, 0, 0, 0, 0)},
243 {L"en", L"2/2/10", L"M/D/YY", CFX_DateTime(2010, 2, 2, 0, 0, 0, 0)},
244 {L"en", L"Jan 10, 1999", L"MMM D, YYYY",
245 CFX_DateTime(1999, 1, 10, 0, 0, 0, 0)},
246 {L"en", L"Jan 10, 1999 AD", L"MMM D, YYYY G",
247 CFX_DateTime(1999, 1, 10, 0, 0, 0, 0)},
248 // TODO(dsinclair): Should this be -2 instead of 2?
249 {L"en", L"Jan 10, 0002 BC", L"MMM D, YYYY G",
250 CFX_DateTime(2, 1, 10, 0, 0, 0, 0)},
251 {L"en", L"October 25, 2002", L"MMMM DD, YYYY",
252 CFX_DateTime(2002, 10, 25, 0, 0, 0, 0)},
253 // TODO(dsinclair): The J and JJJ are ignored during parsing when they
254 // could be turned back into a date.
255 {L"en", L"1999-33", L"YYYY-J", CFX_DateTime(1999, 1, 1, 0, 0, 0, 0)},
256 {L"en", L"1999-033", L"YYYY-JJJ", CFX_DateTime(1999, 1, 1, 0, 0, 0, 0)},
257 {L"de_CH", L"30. Oktober 2004", L"D. MMMM YYYY",
258 CFX_DateTime(2004, 10, 30, 0, 0, 0, 0)},
259 {L"fr_CA", L"30 octobre 2004", L"D MMMM YYYY",
260 CFX_DateTime(2004, 10, 30, 0, 0, 0, 0)},
261 {L"en", L"Saturday, the 1 of January, 2000",
262 L"EEEE, 'the' D 'of' MMMM, YYYY", CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
263 {L"en", L"Sat, the 1 of January, 2000", L"EEE, 'the' D 'of' MMMM, YYYY",
264 CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
265 {L"en", L"7, the 1 of January, 2000", // 7 == Saturday as 1 == Sunday
266 L"E, 'the' D 'of' MMMM, YYYY", CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
267 {L"en", L"6, the 1 of January, 2000", // 6 == Saturday as 1 == Monday
268 L"e, 'the' D 'of' MMMM, YYYY", CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
269 {L"en", L"2004-07-22 Week of the month is 3",
270 L"YYYY-MM-DD 'Week of the month is' w",
271 CFX_DateTime(2004, 7, 22, 0, 0, 0, 0)},
272 {L"en", L"2004-07-22 Week of the year is 03",
273 L"YYYY-MM-DD 'Week of the year is' WW",
274 CFX_DateTime(2004, 7, 22, 0, 0, 0, 0)}
275 // {L"ja", L"H15/11/3", L"gY/M/D", CFX_DateTime(2003, 11, 3, 0, 0, 0, 0)},
276 // {L"ja", L"\u5e731-1-8", L"ggY-M-D", CFX_DateTime(1989, 1, 8, 0, 0, 0,
277 // 0)}, {L"ja", L"\u5e73\u621089/11/03", L"gggYY/MM/DD",
278 // CFX_DateTime(1989, 11, 3, 0, 0, 0, 0)},
279 // {L"ja", L"u337b99/01/08", L"\u0067\u0067YY/MM/DD",
280 // CFX_DateTime(1999, 1, 8, 0, 0, 0, 0)}
281 };
282 // Note, none of the full width date symbols are listed here as they are
283 // not supported. In theory there are the full width versions of DDD,
284 // DDDD, MMM, MMMM, E, e, gg, YYY, YYYYY.
285
286 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
287 CFX_DateTime result;
288 EXPECT_TRUE(
289 fmt(tests[i].locale, tests[i].pattern)
290 ->ParseDateTime(tests[i].input, FX_DATETIMETYPE_Date, &result));
291 EXPECT_EQ(tests[i].output, result) << " TEST: " << i;
292 }
293 }
294
295 // TODO(dsinclair): GetDateTimeFormat is broken and doesn't allow just returning
296 // a parsed Time. It will assume it's a Date. The method needs to be re-written.
297 // TEST_F(CFGAS_StringFormatterTest, TimeParse) {
298 // struct {
299 // const wchar_t* locale;
300 // const wchar_t* input;
301 // const wchar_t* pattern;
302 // CFX_DateTime output;
303 // } tests[] = {
304 // {L"en", L"18:00", L"HH:MM", CFX_DateTime(0, 0, 0, 18, 0, 0, 0)},
305 // {L"en", L"12.59 Uhr", L"H.MM 'Uhr'", CFX_DateTime(0, 0, 0, 12, 59, 0,
306 // 0)}, {L"en", L"1:05:10 PM PST", L"h:MM:SS A Z",
307 // CFX_DateTime(0, 0, 0, 17, 05, 10, 0)}};
308 // // Note, none of the full width date symbols are listed here as they are
309 // // not supported. In theory there are the full width versions of kkk,
310 // // kkkk, HHH, HHHH, KKK, KKKK, MMM, MMMM, SSS, SSSS plus 2 more that the
311 // // spec apparently forgot to list the symbol.
312
313 // for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
314 // CFX_DateTime result;
315 // EXPECT_TRUE(fmt(tests[i].locale)
316 // ->ParseDateTime(tests[i].input, tests[i].pattern,
317 // FX_DATETIMETYPE_Time, &result));
318 // EXPECT_EQ(tests[i].output, result) << " TEST: " << i;
319 // }
320 // }
321
TEST_F(CFGAS_StringFormatterTest,SplitFormatString)322 TEST_F(CFGAS_StringFormatterTest, SplitFormatString) {
323 std::vector<WideString> results = CFGAS_StringFormatter::SplitOnBars(L"");
324 EXPECT_EQ(1UL, results.size());
325 EXPECT_TRUE(results[0].IsEmpty());
326
327 results = CFGAS_StringFormatter::SplitOnBars(L"|");
328 EXPECT_EQ(2UL, results.size());
329 EXPECT_TRUE(results[0].IsEmpty());
330 EXPECT_TRUE(results[1].IsEmpty());
331
332 results = CFGAS_StringFormatter::SplitOnBars(
333 L"null{'No|data'} | null{} | text{999*9999} | text{999*999*9999}");
334 EXPECT_EQ(4UL, results.size());
335
336 const wchar_t* patterns[] = {L"null{'No|data'} ", L" null{} ",
337 L" text{999*9999} ", L" text{999*999*9999}"};
338 for (size_t i = 0; i < results.size(); ++i) {
339 EXPECT_STREQ(patterns[i], results[i].c_str());
340 }
341 }
342
TEST_F(CFGAS_StringFormatterTest,NumParse)343 TEST_F(CFGAS_StringFormatterTest, NumParse) {
344 struct TestCase {
345 const wchar_t* locale;
346 const wchar_t* input;
347 const wchar_t* pattern;
348 const wchar_t* output;
349 };
350
351 static const TestCase tests[] = {
352 // {L"en", L"€100.00", L"num(en_GB){$z,zz9.99}", L"100"},
353 // {L"en", L"1050", L"99V99", L"10.50"},
354 // {L"en", L"3125", L"99V99", L"31.25"},
355 {L"en", L"12.345e3", L"99.999E", L"12345.000000"},
356 {L"en", L"12.345e+3", L"99.999E", L"12345.000000"},
357 {L"en", L"12.345E-2", L"99.999E", L"0.123450"},
358 // TODO(dsinclair): Returns 0.000?
359 // {L"en", L"12e-2", L"99E", L"0.12"},
360 {L"en", L"150", L"z999", L"150"},
361 {L"en", L"150.50$", L"zzz.zz$", L"150.50"},
362 {L"en", L"0150", L"z999", L"0150"},
363 {L"en", L"123CR", L"999cr", L"-123"},
364 {L"en", L"123", L"999cr", L"123"},
365 {L"en", L"123CR", L"999CR", L"-123"},
366 {L"en", L"123 ", L"999CR", L"123"},
367 {L"en", L"123DB", L"999db", L"-123"},
368 {L"en", L"123", L"999db", L"123"},
369 {L"en", L"123DB", L"999DB", L"-123"},
370 {L"en", L"123 ", L"999DB", L"123"},
371 {L"en", L"123.5CR", L"999.9cr", L"-123.5"},
372 {L"en", L"123.5", L"999.9cr", L"123.5"},
373 {L"en", L"123.5CR", L"999.9CR", L"-123.5"},
374 // {L"en", L"123.5 ", L"999.9CR", L"123.5"},
375 {L"en", L"123.5DB", L"999.9db", L"-123.5"},
376 {L"en", L"123.5", L"999.9db", L"123.5"},
377 {L"en", L"123.5DB", L"999.9DB", L"-123.5"},
378 // {L"en", L"123.5 ", L"999.9DB", L"123.5"},
379 {L"en", L"10.50", L"z,zz9.99", L"10.50"},
380 {L"en", L"3,125.00", L"z,zz9.99", L"3125.00"},
381 {L"en", L"$1,234.00", L"$z,zz9.99DB", L"1234.00"},
382 // TODO(dsinclair): Comes out as 1234 instead of -1234.
383 // {L"en", L"$,1234.00DB", L"$z,zz9.99DB", L"-1234.00"},
384 {L"en", L"1.234", L"zz9.zzz", L"1.234"},
385 {L"en", L"1 text", L"num{z 'text'}", L"1"},
386 {L"en", L"1.234 text", L"z.zzz 'text'", L"1.234"},
387 {L"en", L" 1.234", L"ZZ9.ZZZ", L"1.234"},
388 {L"en", L"12.345", L"zz9.zzz", L"12.345"},
389 {L"en", L" 12.345", L"ZZ9.ZZZ", L"12.345"},
390 {L"en", L"123.456", L"zz9.zzz", L"123.456"},
391 {L"en", L"123.456", L"ZZ9.ZZZ", L"123.456"},
392 {L"en", L"123.456-", L"ZZ9.ZZZS", L"-123.456"},
393 {L"en", L"123.456+", L"ZZ9.ZZZS", L"123.456"},
394 {L"en", L"123.456 ", L"ZZ9.ZZZS", L"123.456"},
395 {L"en", L"123.456-", L"ZZ9.ZZZS", L"-123.456"},
396 {L"en", L"123.456+", L"ZZ9.ZZZS", L"123.456"},
397 {L"en", L"123", L"zz9.zzz", L"123"},
398 {L"en", L"123.", L"ZZ9.ZZZ", L"123."},
399 {L"en", L"123.", L"zz9.zzz", L"123."},
400 {L"en", L"123.", L"ZZ9.ZZZ", L"123."},
401 {L"en", L"123.0", L"zz9.zzz", L"123.0"},
402 {L"en", L"123.0", L"ZZ9.ZZZ", L"123.0"},
403 {L"en", L"123.000", L"zz9.zzz", L"123.000"},
404 {L"en", L"123.000", L"ZZ9.ZZZ", L"123.000"},
405 {L"en", L"12,345.67", L"zzz,zz9.88888888", L"12345.67"},
406 {L"en", L"12,345.0000", L"zzz,zz9.88888888", L"12345.0000"},
407 {L"en", L"12,345.6789", L"zzz,zz9.8", L"12345.6789"},
408 {L"en", L"12,345.", L"zzz,zz9.8", L"12345."},
409 {L"en", L"123,456.000", L"zzz,zz9.8888", L"123456.000"},
410 {L"en", L"123,456.0", L"zzz,zz9.8888", L"123456.0"},
411 {L"en", L"123,456", L"zzz,zz9.8888", L"123456"},
412 {L"en", L"123,456", L"ZZZ,ZZ9.88", L"123456"},
413 {L"en", L"12,345.67", L"zzz,zz9.88888888", L"12345.67"},
414 {L"en", L"12,345.0000", L"zzz,zz9.88888888", L"12345.0000"},
415 {L"en", L"12,345.6789", L"zzz,zz9.8", L"12345.6789"},
416 {L"en", L"12,345.", L"zzz,zz9.8", L"12345."},
417 // TODO(dsinclair): Parses to 0
418 // {L"en", L"12%", L"zz9.%%", L".12"},
419 {L"en", L"1,234.50%", L"zzz,zz9.99%%", L"12.345"},
420 // {L"en", L"-00123", L"S999v99", L"-1.23"},
421 {L"en", L" 001.23", L"S999V99", L"001.23"},
422 // {L"en", L" 123.00", L"S999V99", L"123"},
423 {L"en", L" 12.30", L"SZZ9.99", L"12.30"},
424 {L"en", L"- 12.30", L"SZ99.99", L"-12.30"},
425 {L"en", L"123.00", L"szz9.99", L"123.00"},
426 {L"en", L"-123.00", L"szz9.99", L"-123.00"},
427 // {L"en", L"$ 1,234.00 ", L"$ZZ,ZZ9.99CR", L"1234"},
428 // {L"en", L"$ 1,234.00CR", L"$ZZ,ZZ9.99CR", L"-1234"},
429 // {L"en", L"$1,23400", L"$z,zz9.99DB", L"1234"},
430 {L"en", L"$1,234.00DB", L"$z,zz9.99DB", L"-1234.00"},
431 {L"en",
432 L"1\xA0"
433 L"234",
434 L"num(fr){z,zzz}", L"1234"},
435 // TODO(dsinclair): Parses to blank
436 // {L"en", L"1,234%", L"num.percent{}", L"12.34"},
437 // {L"en", L"1\xA0" L"234%%", L"num(fr).percent{}", L"12.34"},
438 // TODO(dsinclair): Parses to blank
439 // {L"en", L"1,234%", L"num{9,999%%}", L"12.34"},
440 {L"fr",
441 L"123\xA0"
442 L"456",
443 L"zzz,zzz", L"123456"},
444 {L"en", L"12%", L"zz%", L"0.12"},
445 {L"en", L"(123", L"(zzz", L"-123"},
446 {L"en", L"123)", L"zzz)", L"-123"},
447 {L"en", L"(123)", L"(zzz)", L"-123"},
448 {L"en", L"123 ", L"zzz)", L"123"},
449 {L"en", L" 123", L"(zzz", L"123"},
450 {L"en", L" 123 ", L"(zzz)", L"123"},
451 {L"en", L"123.5(", L"zzz.z(", L"-123.5"},
452 {L"en", L"123.5)", L"zzz.z)", L"-123.5"},
453 {L"en", L"123.5 ", L"zzz.z)", L"123.5"},
454 {L"en", L"123.5 ", L"zzz.z(", L"123.5"},
455 {L"en", L"123.545,4", L"zzz.zzz,z", L"123.5454"},
456 // https://crbug.com/938724
457 {L"en", L"1", L" num.().().}", L"1"},
458 };
459
460 static const TestCase failures[] = {
461 // https://crbug.com/pdfium/1260
462 {L"en", L"..", L"VC", L"."},
463
464 // https://crbug.com/938626
465 {L"en", L"PDF", L"num( ", L"."},
466
467 // https://crbug.com/945836
468 {L"en", L"9.E99999999999", L"EdEE.E999", L""},
469
470 // https://crbug.com/947188
471 {L"en", L"-3.E98998998 ", L" 35EEEE.EE98", L""},
472 };
473
474 for (const auto& test : tests) {
475 WideString result;
476 EXPECT_TRUE(fmt(test.locale, test.pattern)->ParseNum(test.input, &result))
477 << " TEST: " << test.input << ", " << test.pattern;
478 EXPECT_STREQ(test.output, result.c_str())
479 << " TEST: " << test.input << ", " << test.pattern;
480 }
481
482 for (const auto& test : failures) {
483 WideString result;
484 EXPECT_FALSE(fmt(test.locale, test.pattern)->ParseNum(test.input, &result))
485 << " TEST: " << test.input << ", " << test.pattern;
486 }
487 }
488
TEST_F(CFGAS_StringFormatterTest,NumFormat)489 TEST_F(CFGAS_StringFormatterTest, NumFormat) {
490 struct TestCase {
491 const wchar_t* locale;
492 const wchar_t* input;
493 const wchar_t* pattern;
494 const wchar_t* output;
495 };
496
497 static const TestCase tests[] = {
498 {L"en", L"1.234", L"zz9.zzz", L"1.234"},
499 {L"en", L"1", L"num{z 'text'}", L"1 text"},
500 {L"en", L"1", L"num{'text' z}", L"text 1"},
501 {L"en", L"1.234", L"ZZ9.ZZZ", L" 1.234"},
502 {L"en", L"12.345", L"zz9.zzz", L"12.345"},
503 {L"en", L"12.345", L"ZZ9.ZZZ", L" 12.345"},
504 {L"en", L"123.456", L"zz9.zzz", L"123.456"},
505 {L"en", L"123.456", L"ZZ9.ZZZ", L"123.456"},
506 {L"en", L"123", L"zz9.zzz", L"123"},
507 {L"en", L"123", L"ZZ9.ZZZ", L"123.000"},
508 {L"en", L"123.", L"zz9.zzz", L"123."},
509 {L"en", L"123.", L"ZZ9.ZZZ", L"123.000"},
510 {L"en", L"123.0", L"zz9.zzz", L"123"},
511 {L"en", L"123.0", L"ZZ9.ZZZ", L"123.000"},
512 {L"en", L"123.000", L"zz9.zzz", L"123"},
513 {L"en", L"123.000", L"ZZ9.ZZZ", L"123.000"},
514 // {L"en", L"12345.67", L"zzz,zz9.88888888", L"12,345.67"},
515 // {L"en", L"12345.0000", L"zzz,zz9.88888888", L"12,345.0000"},
516 // {L"en", L"12345.6789", L"zzz,zz9.8", L"12,345.6789"},
517 // {L"en", L"12345.", L"zzz,zz9.8", L"12,345"},
518 // {L"en", L"123456.000", L"zzz,zz9.8888", L"123,456.000"},
519 // {L"en", L"123456.0", L"zzz,zz9.8888", L"123,456.0"},
520 {L"en", L"123456", L"zzz,zz9.8888", L"123,456"},
521 {L"en", L"123456", L"ZZZ,ZZ9.88", L"123,456"},
522 // {L"en", L"12345.67", L"zzz,zz9.88888888", L"12,345.67"},
523 // {L"en", L"12345.0000", L"zzz,zz9.88888888", L"12,345.0000"},
524 // {L"en", L"12345.6789", L"zzz,zz9.8", L"12,345.6789"},
525 // {L"en", L"12345.", L"zzz,zz9.8", L"12,345"},
526 // {L"en", L"12%%", L"zz9.%%", L"12%%"},
527 // {L"en", L"1,234.5%%", L"zzz,zz9.99%%", L"1,234.50%%"},
528 {L"en", L"-1.23", L"S999v99", L"-00123"},
529 {L"en", L"1.23", L"S999V99", L" 001.23"},
530 {L"en", L"123", L"S999V99", L" 123.00"},
531 {L"en", L"12.3", L"SZZ9.99", L" 12.30"},
532 {L"en", L"-12.3", L"SZ99.99", L"- 12.30"},
533 {L"en", L"123", L"szz9.99", L"123.00"},
534 {L"en", L"-123", L"szz9.99", L"-123.00"},
535 // {L"en", L"1234", L"$ZZ,ZZ9.99CR", L"$ 1,234.00 "},
536 // {L"en", L"-1234", L"$ZZ,ZZ9.99CR", L"$ 1,234.00CR"},
537 // {L"en", L"1234", L"$z,zz9.99DB", L"$1,234.00"},
538 {L"en", L"-1234", L"$z,zz9.99DB", L"$1,234.00DB"},
539 {L"en", L"12345", L"99.999E", L"12.345E+3"},
540 {L"en", L"12345", L"99999E", L"12345E+0"},
541 {L"en", L".12345", L"99.999E", L"12.345E-2"},
542 {L"en", L"12345", L"99,999", L"12,345"},
543 {L"en", L"1234", L"num(fr){z,zzz}",
544 L"1\xA0"
545 L"234"},
546 {L"en", L"12.34", L"num.percent{}", L"1,234%"},
547 {L"en", L"12.34", L"num(fr).percent{}",
548 L"1\xA0"
549 L"234%"},
550 // {L"en", L"12.34", L"num{9,999%%}", L"1,234%"},
551 {L"en", L"-123", L"zzzCR", L"123CR"},
552 {L"en", L"123", L"zzzCR", L"123 "},
553 {L"en", L"-123", L"zzzcr", L"123CR"},
554 {L"en", L"123", L"zzzcr", L"123"},
555 {L"en", L"123", L"zzz$", L"123$"},
556 {L"en", L"-123.5", L"zzz.zCR", L"123.5CR"},
557 {L"en", L"123.5", L"zzz.zCR", L"123.5 "},
558 {L"en", L"-123.5", L"zzz.zcr", L"123.5CR"},
559 {L"en", L"123.5", L"zzz.zcr", L"123.5"},
560
561 {L"en", L"-123.5", L"999.9db", L"123.5db"},
562 {L"en", L"123.5", L"999.9db", L"123.5"},
563 {L"en", L"-123.5", L"999.9DB", L"123.5DB"},
564 {L"en", L"123.5", L"999.9DB", L"123.5 "},
565
566 {L"en", L"-123", L"(zzz", L"(123"},
567 // {L"en", L"-123", L"zzz)", L"123)"},
568 {L"en", L"-123", L"(zzz)", L"(123)"},
569 {L"en", L"123", L"zzz)", L"123 "},
570 {L"en", L"123", L"(zzz", L" 123"},
571 {L"en", L"123", L"(zzz)", L" 123 "},
572 {L"en", L"-123.5", L"zzz.z(", L"123.5("},
573 // {L"en", L"-123.5", L"zzz.z)", L"123.5)"},
574 {L"en", L"123.5", L"zzz.z)", L"123.5 "},
575 {L"en", L"123.5", L"zzz.z(", L"123.5 "},
576 // https://crbug.com/pdfium/1233
577 {L"en", L"1", L"r9", L"r1"},
578 {L"en", L"1", L"R9", L"R1"},
579 {L"en", L"1", L"b9", L"b1"},
580 {L"en", L"1", L"B9", L"B1"},
581 {L"en", L"1", L"9.c", L"1"},
582 {L"en", L"1", L"9.C", L"1"},
583 {L"en", L"1", L"9.d", L"1"},
584 {L"en", L"1", L"9.D", L"1"},
585 // https://crbug.com/pdfium/1244
586 {L"en", L"1", L"E", L"1"},
587 {L"en", L"0", L"E", L"0"},
588 {L"en", L"-1", L"E", L"-1"},
589 {L"en", L"900000000000000000000", L"E", L"900,000,000,000,000,000,000"},
590 // TODO(tsepez): next one seems wrong
591 // {L"en", L".000000000000000000009", L"E", L"9"},
592 // https://crbug.com/938724
593 {L"en", L"1", L"| num.().().", L"1"},
594 // https://crbug.com/942449
595 {L"en", L"1", L"9.", L"1"},
596 };
597
598 static const TestCase failures[] = {
599 // https://crbug.com/pdfium/1271
600 {L"en", L"1", L"num.{E", L""},
601 };
602
603 for (const auto& test : tests) {
604 WideString result;
605 EXPECT_TRUE(fmt(test.locale, test.pattern)->FormatNum(test.input, &result))
606 << " TEST: " << test.input << ", " << test.pattern;
607 EXPECT_STREQ(test.output, result.c_str())
608 << " TEST: " << test.input << ", " << test.pattern;
609 }
610
611 for (const auto& test : failures) {
612 WideString result;
613 EXPECT_FALSE(fmt(test.locale, test.pattern)->FormatNum(test.input, &result))
614 << " TEST: " << test.input << ", " << test.pattern;
615 }
616 }
617
TEST_F(CFGAS_StringFormatterTest,TextParse)618 TEST_F(CFGAS_StringFormatterTest, TextParse) {
619 struct {
620 const wchar_t* locale;
621 const wchar_t* input;
622 const wchar_t* pattern;
623 const wchar_t* output;
624 } tests[] = {// TODO(dsinclair) Missing support for the global modifiers:
625 // ? - wildcard
626 // * - zero or more whitespace
627 // + - one or more whitespace
628 // {L"en", L"555-1212", L"text(th_TH){999*9999}", L"5551212"},
629 {L"en", L"ABC-1234-5", L"AAA-9999-X", L"ABC12345"},
630 {L"en", L"ABC-1234-D", L"AAA-9999-X", L"ABC1234D"},
631 {L"en", L"A1C-1234-D", L"OOO-9999-X", L"A1C1234D"},
632 {L"en", L"A1C-1234-D", L"000-9999-X", L"A1C1234D"},
633 {L"en", L"A1C-1234-D text", L"000-9999-X 'text'", L"A1C1234D"}};
634
635 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
636 WideString result;
637 EXPECT_TRUE(fmt(tests[i].locale, tests[i].pattern)
638 ->ParseText(tests[i].input, &result));
639 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
640 }
641 }
642
TEST_F(CFGAS_StringFormatterTest,InvalidTextParse)643 TEST_F(CFGAS_StringFormatterTest, InvalidTextParse) {
644 // Input does not match mask.
645 WideString result;
646 EXPECT_FALSE(fmt(L"en", L"AAA-9999-X")->ParseText(L"123-4567-8", &result));
647 }
648
TEST_F(CFGAS_StringFormatterTest,TextFormat)649 TEST_F(CFGAS_StringFormatterTest, TextFormat) {
650 struct {
651 const wchar_t* locale;
652 const wchar_t* input;
653 const wchar_t* pattern;
654 const wchar_t* output;
655 } tests[] = {
656 {L"en", L"K1S5K2", L"A9A 9A9", L"K1S 5K2"},
657 {L"en", L"K1S5K2", L"text(fr){A9A 9A9}", L"K1S 5K2"},
658 {L"en", L"6135551212", L"'+1 ('9\u002399') '999-9999",
659 L"+1 (6#13) 555-1212"},
660 {L"en", L"6135551212", L"999.999.9999", L"613.555.1212"},
661 {L"en", L"6135551212", L"999\u0023999\u002A9999", L"613#555*1212"},
662 {L"en", L"K1#5K2", L"00X OO9", L"K1# 5K2"},
663 };
664
665 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
666 WideString result;
667 EXPECT_TRUE(fmt(tests[i].locale, tests[i].pattern)
668 ->FormatText(tests[i].input, &result));
669 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
670 }
671 }
672
TEST_F(CFGAS_StringFormatterTest,NullParse)673 TEST_F(CFGAS_StringFormatterTest, NullParse) {
674 struct {
675 const wchar_t* locale;
676 const wchar_t* input;
677 const wchar_t* pattern;
678 } tests[] = {
679 {L"en", L"", L"null{}"},
680 {L"en", L"No data", L"null{'No data'}"},
681 };
682
683 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
684 EXPECT_TRUE(
685 fmt(tests[i].locale, tests[i].pattern)->ParseNull(tests[i].input))
686 << " TEST: " << i;
687 }
688 }
689
TEST_F(CFGAS_StringFormatterTest,NullFormat)690 TEST_F(CFGAS_StringFormatterTest, NullFormat) {
691 struct {
692 const wchar_t* locale;
693 const wchar_t* pattern;
694 const wchar_t* output;
695 } tests[] = {{L"en", L"null{'n/a'}", L"n/a"}, {L"en", L"null{}", L""}};
696
697 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
698 WideString result;
699 EXPECT_TRUE(fmt(tests[i].locale, tests[i].pattern)->FormatNull(&result));
700 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
701 }
702 }
703
TEST_F(CFGAS_StringFormatterTest,ZeroParse)704 TEST_F(CFGAS_StringFormatterTest, ZeroParse) {
705 struct {
706 const wchar_t* locale;
707 const wchar_t* input;
708 const wchar_t* pattern;
709 } tests[] = {{L"en", L"", L"zero{}"},
710 {L"en", L"9", L"zero{9}"},
711 {L"en", L"a", L"zero{'a'}"}};
712
713 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
714 EXPECT_TRUE(
715 fmt(tests[i].locale, tests[i].pattern)->ParseZero(tests[i].input))
716 << " TEST: " << i;
717 }
718 }
719
TEST_F(CFGAS_StringFormatterTest,ZeroFormat)720 TEST_F(CFGAS_StringFormatterTest, ZeroFormat) {
721 struct {
722 const wchar_t* locale;
723 const wchar_t* input;
724 const wchar_t* pattern;
725 const wchar_t* output;
726 } tests[] = {// TODO(dsinclair): The zero format can take a number specifier
727 // which we don't take into account.
728 // {L"en", L"", L"zero {9}", L""},
729 // {L"en", L"0", L"zero {9}", L"0"},
730 // {L"en", L"0.0", L"zero{9}", L"0"},
731 {L"en", L"0", L"zero{}", L""}};
732
733 for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
734 WideString result;
735 EXPECT_TRUE(fmt(tests[i].locale, tests[i].pattern)->FormatZero(&result));
736 EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
737 }
738 }
739
TEST_F(CFGAS_StringFormatterTest,GetCategory)740 TEST_F(CFGAS_StringFormatterTest, GetCategory) {
741 EXPECT_EQ(FX_LOCALECATEGORY_Unknown,
742 fmt(L"en", L"'just text'")->GetCategory());
743 EXPECT_EQ(FX_LOCALECATEGORY_Null, fmt(L"en", L"null{}")->GetCategory());
744 EXPECT_EQ(FX_LOCALECATEGORY_Zero, fmt(L"en", L"zero{}")->GetCategory());
745 EXPECT_EQ(FX_LOCALECATEGORY_Num, fmt(L"en", L"num{}")->GetCategory());
746 EXPECT_EQ(FX_LOCALECATEGORY_Text, fmt(L"en", L"text{}")->GetCategory());
747 EXPECT_EQ(FX_LOCALECATEGORY_DateTime,
748 fmt(L"en", L"datetime{}")->GetCategory());
749 EXPECT_EQ(FX_LOCALECATEGORY_Time, fmt(L"en", L"time{}")->GetCategory());
750 EXPECT_EQ(FX_LOCALECATEGORY_Date, fmt(L"en", L"date{}")->GetCategory());
751 EXPECT_EQ(FX_LOCALECATEGORY_DateTime,
752 fmt(L"en", L"time{} date{}")->GetCategory());
753 EXPECT_EQ(FX_LOCALECATEGORY_DateTime,
754 fmt(L"en", L"date{} time{}")->GetCategory());
755 EXPECT_EQ(FX_LOCALECATEGORY_Num, fmt(L"en", L"num(en_GB){}")->GetCategory());
756 EXPECT_EQ(FX_LOCALECATEGORY_Date, fmt(L"en", L"date.long{}")->GetCategory());
757 }
758