]>
Commit | Line | Data |
---|---|---|
1d09f67e TL |
1 | // Licensed to the Apache Software Foundation (ASF) under one |
2 | // or more contributor license agreements. See the NOTICE file | |
3 | // distributed with this work for additional information | |
4 | // regarding copyright ownership. The ASF licenses this file | |
5 | // to you under the Apache License, Version 2.0 (the | |
6 | // "License"); you may not use this file except in compliance | |
7 | // with the License. You may obtain a copy of the License at | |
8 | // | |
9 | // http://www.apache.org/licenses/LICENSE-2.0 | |
10 | // | |
11 | // Unless required by applicable law or agreed to in writing, | |
12 | // software distributed under the License is distributed on an | |
13 | // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY | |
14 | // KIND, either express or implied. See the License for the | |
15 | // specific language governing permissions and limitations | |
16 | // under the License. | |
17 | ||
18 | #include "arrow/pretty_print.h" | |
19 | ||
20 | #include <gtest/gtest.h> | |
21 | ||
22 | #include <cstdint> | |
23 | #include <cstring> | |
24 | #include <limits> | |
25 | #include <memory> | |
26 | #include <sstream> | |
27 | #include <string> | |
28 | #include <vector> | |
29 | ||
30 | #include "arrow/array.h" | |
31 | #include "arrow/table.h" | |
32 | #include "arrow/testing/gtest_util.h" | |
33 | #include "arrow/type.h" | |
34 | #include "arrow/util/key_value_metadata.h" | |
35 | ||
36 | namespace arrow { | |
37 | ||
38 | class TestPrettyPrint : public ::testing::Test { | |
39 | public: | |
40 | void SetUp() {} | |
41 | ||
42 | void Print(const Array& array) {} | |
43 | ||
44 | private: | |
45 | std::ostringstream sink_; | |
46 | }; | |
47 | ||
48 | template <typename T> | |
49 | void CheckStream(const T& obj, const PrettyPrintOptions& options, const char* expected) { | |
50 | std::ostringstream sink; | |
51 | ASSERT_OK(PrettyPrint(obj, options, &sink)); | |
52 | std::string result = sink.str(); | |
53 | ASSERT_EQ(std::string(expected, strlen(expected)), result); | |
54 | } | |
55 | ||
56 | void CheckArray(const Array& arr, const PrettyPrintOptions& options, const char* expected, | |
57 | bool check_operator = true) { | |
58 | ARROW_SCOPED_TRACE("For datatype: ", arr.type()->ToString()); | |
59 | CheckStream(arr, options, expected); | |
60 | ||
61 | if (options.indent == 0 && check_operator) { | |
62 | std::stringstream ss; | |
63 | ss << arr; | |
64 | std::string result = std::string(expected, strlen(expected)); | |
65 | ASSERT_EQ(result, ss.str()); | |
66 | } | |
67 | } | |
68 | ||
69 | template <typename T> | |
70 | void Check(const T& obj, const PrettyPrintOptions& options, const char* expected) { | |
71 | std::string result; | |
72 | ASSERT_OK(PrettyPrint(obj, options, &result)); | |
73 | ASSERT_EQ(std::string(expected, strlen(expected)), result); | |
74 | } | |
75 | ||
76 | template <typename TYPE, typename C_TYPE> | |
77 | void CheckPrimitive(const std::shared_ptr<DataType>& type, | |
78 | const PrettyPrintOptions& options, const std::vector<bool>& is_valid, | |
79 | const std::vector<C_TYPE>& values, const char* expected, | |
80 | bool check_operator = true) { | |
81 | std::shared_ptr<Array> array; | |
82 | ArrayFromVector<TYPE, C_TYPE>(type, is_valid, values, &array); | |
83 | CheckArray(*array, options, expected, check_operator); | |
84 | } | |
85 | ||
86 | template <typename TYPE, typename C_TYPE> | |
87 | void CheckPrimitive(const PrettyPrintOptions& options, const std::vector<bool>& is_valid, | |
88 | const std::vector<C_TYPE>& values, const char* expected, | |
89 | bool check_operator = true) { | |
90 | CheckPrimitive<TYPE, C_TYPE>(TypeTraits<TYPE>::type_singleton(), options, is_valid, | |
91 | values, expected, check_operator); | |
92 | } | |
93 | ||
94 | TEST_F(TestPrettyPrint, PrimitiveType) { | |
95 | std::vector<bool> is_valid = {true, true, false, true, false}; | |
96 | ||
97 | std::vector<int32_t> values = {0, 1, 2, 3, 4}; | |
98 | static const char* expected = R"expected([ | |
99 | 0, | |
100 | 1, | |
101 | null, | |
102 | 3, | |
103 | null | |
104 | ])expected"; | |
105 | CheckPrimitive<Int32Type, int32_t>({0, 10}, is_valid, values, expected); | |
106 | ||
107 | static const char* expected_na = R"expected([ | |
108 | 0, | |
109 | 1, | |
110 | NA, | |
111 | 3, | |
112 | NA | |
113 | ])expected"; | |
114 | CheckPrimitive<Int32Type, int32_t>({0, 10, 2, "NA"}, is_valid, values, expected_na, | |
115 | false); | |
116 | ||
117 | static const char* ex_in2 = R"expected( [ | |
118 | 0, | |
119 | 1, | |
120 | null, | |
121 | 3, | |
122 | null | |
123 | ])expected"; | |
124 | CheckPrimitive<Int32Type, int32_t>({2, 10}, is_valid, values, ex_in2); | |
125 | static const char* ex_in2_w2 = R"expected( [ | |
126 | 0, | |
127 | 1, | |
128 | ... | |
129 | 3, | |
130 | null | |
131 | ])expected"; | |
132 | CheckPrimitive<Int32Type, int32_t>({2, 2}, is_valid, values, ex_in2_w2); | |
133 | ||
134 | std::vector<double> values2 = {0., 1., 2., 3., 4.}; | |
135 | static const char* ex2 = R"expected([ | |
136 | 0, | |
137 | 1, | |
138 | null, | |
139 | 3, | |
140 | null | |
141 | ])expected"; | |
142 | CheckPrimitive<DoubleType, double>({0, 10}, is_valid, values2, ex2); | |
143 | static const char* ex2_in2 = R"expected( [ | |
144 | 0, | |
145 | 1, | |
146 | null, | |
147 | 3, | |
148 | null | |
149 | ])expected"; | |
150 | CheckPrimitive<DoubleType, double>({2, 10}, is_valid, values2, ex2_in2); | |
151 | ||
152 | std::vector<std::string> values3 = {"foo", "bar", "", "baz", ""}; | |
153 | static const char* ex3 = R"expected([ | |
154 | "foo", | |
155 | "bar", | |
156 | null, | |
157 | "baz", | |
158 | null | |
159 | ])expected"; | |
160 | CheckPrimitive<StringType, std::string>({0, 10}, is_valid, values3, ex3); | |
161 | CheckPrimitive<LargeStringType, std::string>({0, 10}, is_valid, values3, ex3); | |
162 | static const char* ex3_in2 = R"expected( [ | |
163 | "foo", | |
164 | "bar", | |
165 | null, | |
166 | "baz", | |
167 | null | |
168 | ])expected"; | |
169 | CheckPrimitive<StringType, std::string>({2, 10}, is_valid, values3, ex3_in2); | |
170 | CheckPrimitive<LargeStringType, std::string>({2, 10}, is_valid, values3, ex3_in2); | |
171 | } | |
172 | ||
173 | TEST_F(TestPrettyPrint, PrimitiveTypeNoNewlines) { | |
174 | std::vector<bool> is_valid = {true, true, false, true, false}; | |
175 | std::vector<int32_t> values = {0, 1, 2, 3, 4}; | |
176 | ||
177 | PrettyPrintOptions options{}; | |
178 | options.skip_new_lines = true; | |
179 | options.window = 4; | |
180 | ||
181 | const char* expected = "[0,1,null,3,null]"; | |
182 | CheckPrimitive<Int32Type, int32_t>(options, is_valid, values, expected, false); | |
183 | ||
184 | // With ellipsis | |
185 | is_valid.insert(is_valid.end(), 20, true); | |
186 | is_valid.insert(is_valid.end(), {true, false, true}); | |
187 | values.insert(values.end(), 20, 99); | |
188 | values.insert(values.end(), {44, 43, 42}); | |
189 | ||
190 | expected = "[0,1,null,3,...,99,44,null,42]"; | |
191 | CheckPrimitive<Int32Type, int32_t>(options, is_valid, values, expected, false); | |
192 | } | |
193 | ||
194 | TEST_F(TestPrettyPrint, Int8) { | |
195 | static const char* expected = R"expected([ | |
196 | 0, | |
197 | 127, | |
198 | -128 | |
199 | ])expected"; | |
200 | CheckPrimitive<Int8Type, int8_t>({0, 10}, {true, true, true}, {0, 127, -128}, expected); | |
201 | } | |
202 | ||
203 | TEST_F(TestPrettyPrint, UInt8) { | |
204 | static const char* expected = R"expected([ | |
205 | 0, | |
206 | 255 | |
207 | ])expected"; | |
208 | CheckPrimitive<UInt8Type, uint8_t>({0, 10}, {true, true}, {0, 255}, expected); | |
209 | } | |
210 | ||
211 | TEST_F(TestPrettyPrint, Int64) { | |
212 | static const char* expected = R"expected([ | |
213 | 0, | |
214 | 9223372036854775807, | |
215 | -9223372036854775808 | |
216 | ])expected"; | |
217 | CheckPrimitive<Int64Type, int64_t>( | |
218 | {0, 10}, {true, true, true}, {0, 9223372036854775807LL, -9223372036854775807LL - 1}, | |
219 | expected); | |
220 | } | |
221 | ||
222 | TEST_F(TestPrettyPrint, UInt64) { | |
223 | static const char* expected = R"expected([ | |
224 | 0, | |
225 | 9223372036854775803, | |
226 | 18446744073709551615 | |
227 | ])expected"; | |
228 | CheckPrimitive<UInt64Type, uint64_t>( | |
229 | {0, 10}, {true, true, true}, {0, 9223372036854775803ULL, 18446744073709551615ULL}, | |
230 | expected); | |
231 | } | |
232 | ||
233 | TEST_F(TestPrettyPrint, DateTimeTypes) { | |
234 | std::vector<bool> is_valid = {true, true, false, true, false}; | |
235 | ||
236 | { | |
237 | std::vector<int32_t> values = {0, 1, 2, 31, 4}; | |
238 | static const char* expected = R"expected([ | |
239 | 1970-01-01, | |
240 | 1970-01-02, | |
241 | null, | |
242 | 1970-02-01, | |
243 | null | |
244 | ])expected"; | |
245 | CheckPrimitive<Date32Type, int32_t>({0, 10}, is_valid, values, expected); | |
246 | } | |
247 | ||
248 | { | |
249 | constexpr int64_t ms_per_day = 24 * 60 * 60 * 1000; | |
250 | std::vector<int64_t> values = {0 * ms_per_day, 1 * ms_per_day, 2 * ms_per_day, | |
251 | 31 * ms_per_day, 4 * ms_per_day}; | |
252 | static const char* expected = R"expected([ | |
253 | 1970-01-01, | |
254 | 1970-01-02, | |
255 | null, | |
256 | 1970-02-01, | |
257 | null | |
258 | ])expected"; | |
259 | CheckPrimitive<Date64Type, int64_t>({0, 10}, is_valid, values, expected); | |
260 | } | |
261 | ||
262 | { | |
263 | std::vector<int64_t> values = { | |
264 | 0, 1, 2, 678 + 1000000 * (5 + 60 * (4 + 60 * (3 + 24 * int64_t(1)))), 4}; | |
265 | static const char* expected = R"expected([ | |
266 | 1970-01-01 00:00:00.000000, | |
267 | 1970-01-01 00:00:00.000001, | |
268 | null, | |
269 | 1970-01-02 03:04:05.000678, | |
270 | null | |
271 | ])expected"; | |
272 | CheckPrimitive<TimestampType, int64_t>(timestamp(TimeUnit::MICRO, "Transylvania"), | |
273 | {0, 10}, is_valid, values, expected); | |
274 | } | |
275 | ||
276 | { | |
277 | std::vector<int32_t> values = {1, 62, 2, 3 + 60 * (2 + 60 * 1), 4}; | |
278 | static const char* expected = R"expected([ | |
279 | 00:00:01, | |
280 | 00:01:02, | |
281 | null, | |
282 | 01:02:03, | |
283 | null | |
284 | ])expected"; | |
285 | CheckPrimitive<Time32Type, int32_t>(time32(TimeUnit::SECOND), {0, 10}, is_valid, | |
286 | values, expected); | |
287 | } | |
288 | ||
289 | { | |
290 | std::vector<int64_t> values = { | |
291 | 0, 1, 2, 678 + int64_t(1000000000) * (5 + 60 * (4 + 60 * 3)), 4}; | |
292 | static const char* expected = R"expected([ | |
293 | 00:00:00.000000000, | |
294 | 00:00:00.000000001, | |
295 | null, | |
296 | 03:04:05.000000678, | |
297 | null | |
298 | ])expected"; | |
299 | CheckPrimitive<Time64Type, int64_t>(time64(TimeUnit::NANO), {0, 10}, is_valid, values, | |
300 | expected); | |
301 | } | |
302 | } | |
303 | ||
304 | TEST_F(TestPrettyPrint, TestIntervalTypes) { | |
305 | std::vector<bool> is_valid = {true, true, false, true, false}; | |
306 | ||
307 | { | |
308 | std::vector<DayTimeIntervalType::DayMilliseconds> values = { | |
309 | {1, 2}, {-3, 4}, {}, {}, {}}; | |
310 | static const char* expected = R"expected([ | |
311 | 1d2ms, | |
312 | -3d4ms, | |
313 | null, | |
314 | 0d0ms, | |
315 | null | |
316 | ])expected"; | |
317 | CheckPrimitive<DayTimeIntervalType, DayTimeIntervalType::DayMilliseconds>( | |
318 | {0, 10}, is_valid, values, expected); | |
319 | } | |
320 | { | |
321 | std::vector<MonthDayNanoIntervalType::MonthDayNanos> values = { | |
322 | {1, 2, 3}, {-3, 4, -5}, {}, {}, {}}; | |
323 | static const char* expected = R"expected([ | |
324 | 1M2d3ns, | |
325 | -3M4d-5ns, | |
326 | null, | |
327 | 0M0d0ns, | |
328 | null | |
329 | ])expected"; | |
330 | CheckPrimitive<MonthDayNanoIntervalType, MonthDayNanoIntervalType::MonthDayNanos>( | |
331 | {0, 10}, is_valid, values, expected); | |
332 | } | |
333 | } | |
334 | ||
335 | TEST_F(TestPrettyPrint, DateTimeTypesWithOutOfRangeValues) { | |
336 | // Our vendored date library allows years within [-32767, 32767], | |
337 | // which limits the range of values which can be displayed. | |
338 | const int32_t min_int32 = std::numeric_limits<int32_t>::min(); | |
339 | const int32_t max_int32 = std::numeric_limits<int32_t>::max(); | |
340 | const int64_t min_int64 = std::numeric_limits<int64_t>::min(); | |
341 | const int64_t max_int64 = std::numeric_limits<int64_t>::max(); | |
342 | ||
343 | const int32_t min_date32 = -12687428; | |
344 | const int32_t max_date32 = 11248737; | |
345 | const int64_t min_date64 = 86400000LL * min_date32; | |
346 | const int64_t max_date64 = 86400000LL * (max_date32 + 1) - 1; | |
347 | ||
348 | const int32_t min_time32_seconds = 0; | |
349 | const int32_t max_time32_seconds = 86399; | |
350 | const int32_t min_time32_millis = 0; | |
351 | const int32_t max_time32_millis = 86399999; | |
352 | const int64_t min_time64_micros = 0; | |
353 | const int64_t max_time64_micros = 86399999999LL; | |
354 | const int64_t min_time64_nanos = 0; | |
355 | const int64_t max_time64_nanos = 86399999999999LL; | |
356 | ||
357 | const int64_t min_timestamp_seconds = -1096193779200LL; | |
358 | const int64_t max_timestamp_seconds = 971890963199LL; | |
359 | const int64_t min_timestamp_millis = min_timestamp_seconds * 1000; | |
360 | const int64_t max_timestamp_millis = max_timestamp_seconds * 1000 + 999; | |
361 | const int64_t min_timestamp_micros = min_timestamp_millis * 1000; | |
362 | const int64_t max_timestamp_micros = max_timestamp_millis * 1000 + 999; | |
363 | ||
364 | std::vector<bool> is_valid = {false, false, false, false, true, | |
365 | true, true, true, true, true}; | |
366 | ||
367 | // Dates | |
368 | { | |
369 | std::vector<int32_t> values = {min_int32, max_int32, min_date32 - 1, max_date32 + 1, | |
370 | min_int32, max_int32, min_date32 - 1, max_date32 + 1, | |
371 | min_date32, max_date32}; | |
372 | static const char* expected = R"expected([ | |
373 | null, | |
374 | null, | |
375 | null, | |
376 | null, | |
377 | <value out of range: -2147483648>, | |
378 | <value out of range: 2147483647>, | |
379 | <value out of range: -12687429>, | |
380 | <value out of range: 11248738>, | |
381 | -32767-01-01, | |
382 | 32767-12-31 | |
383 | ])expected"; | |
384 | CheckPrimitive<Date32Type, int32_t>({0, 10}, is_valid, values, expected); | |
385 | } | |
386 | { | |
387 | std::vector<int64_t> values = {min_int64, max_int64, min_date64 - 1, max_date64 + 1, | |
388 | min_int64, max_int64, min_date64 - 1, max_date64 + 1, | |
389 | min_date64, max_date64}; | |
390 | static const char* expected = R"expected([ | |
391 | null, | |
392 | null, | |
393 | null, | |
394 | null, | |
395 | <value out of range: -9223372036854775808>, | |
396 | <value out of range: 9223372036854775807>, | |
397 | <value out of range: -1096193779200001>, | |
398 | <value out of range: 971890963200000>, | |
399 | -32767-01-01, | |
400 | 32767-12-31 | |
401 | ])expected"; | |
402 | CheckPrimitive<Date64Type, int64_t>({0, 10}, is_valid, values, expected); | |
403 | } | |
404 | ||
405 | // Times | |
406 | { | |
407 | std::vector<int32_t> values = {min_int32, | |
408 | max_int32, | |
409 | min_time32_seconds - 1, | |
410 | max_time32_seconds + 1, | |
411 | min_int32, | |
412 | max_int32, | |
413 | min_time32_seconds - 1, | |
414 | max_time32_seconds + 1, | |
415 | min_time32_seconds, | |
416 | max_time32_seconds}; | |
417 | static const char* expected = R"expected([ | |
418 | null, | |
419 | null, | |
420 | null, | |
421 | null, | |
422 | <value out of range: -2147483648>, | |
423 | <value out of range: 2147483647>, | |
424 | <value out of range: -1>, | |
425 | <value out of range: 86400>, | |
426 | 00:00:00, | |
427 | 23:59:59 | |
428 | ])expected"; | |
429 | CheckPrimitive<Time32Type, int32_t>(time32(TimeUnit::SECOND), {0, 10}, is_valid, | |
430 | values, expected); | |
431 | } | |
432 | { | |
433 | std::vector<int32_t> values = { | |
434 | min_int32, max_int32, min_time32_millis - 1, max_time32_millis + 1, | |
435 | min_int32, max_int32, min_time32_millis - 1, max_time32_millis + 1, | |
436 | min_time32_millis, max_time32_millis}; | |
437 | static const char* expected = R"expected([ | |
438 | null, | |
439 | null, | |
440 | null, | |
441 | null, | |
442 | <value out of range: -2147483648>, | |
443 | <value out of range: 2147483647>, | |
444 | <value out of range: -1>, | |
445 | <value out of range: 86400000>, | |
446 | 00:00:00.000, | |
447 | 23:59:59.999 | |
448 | ])expected"; | |
449 | CheckPrimitive<Time32Type, int32_t>(time32(TimeUnit::MILLI), {0, 10}, is_valid, | |
450 | values, expected); | |
451 | } | |
452 | { | |
453 | std::vector<int64_t> values = { | |
454 | min_int64, max_int64, min_time64_micros - 1, max_time64_micros + 1, | |
455 | min_int64, max_int64, min_time64_micros - 1, max_time64_micros + 1, | |
456 | min_time64_micros, max_time64_micros}; | |
457 | static const char* expected = R"expected([ | |
458 | null, | |
459 | null, | |
460 | null, | |
461 | null, | |
462 | <value out of range: -9223372036854775808>, | |
463 | <value out of range: 9223372036854775807>, | |
464 | <value out of range: -1>, | |
465 | <value out of range: 86400000000>, | |
466 | 00:00:00.000000, | |
467 | 23:59:59.999999 | |
468 | ])expected"; | |
469 | CheckPrimitive<Time64Type, int64_t>(time64(TimeUnit::MICRO), {0, 10}, is_valid, | |
470 | values, expected); | |
471 | } | |
472 | { | |
473 | std::vector<int64_t> values = { | |
474 | min_int64, max_int64, min_time64_nanos - 1, max_time64_nanos + 1, | |
475 | min_int64, max_int64, min_time64_nanos - 1, max_time64_nanos + 1, | |
476 | min_time64_nanos, max_time64_nanos}; | |
477 | static const char* expected = R"expected([ | |
478 | null, | |
479 | null, | |
480 | null, | |
481 | null, | |
482 | <value out of range: -9223372036854775808>, | |
483 | <value out of range: 9223372036854775807>, | |
484 | <value out of range: -1>, | |
485 | <value out of range: 86400000000000>, | |
486 | 00:00:00.000000000, | |
487 | 23:59:59.999999999 | |
488 | ])expected"; | |
489 | CheckPrimitive<Time64Type, int64_t>(time64(TimeUnit::NANO), {0, 10}, is_valid, values, | |
490 | expected); | |
491 | } | |
492 | ||
493 | // Timestamps | |
494 | { | |
495 | std::vector<int64_t> values = {min_int64, | |
496 | max_int64, | |
497 | min_timestamp_seconds - 1, | |
498 | max_timestamp_seconds + 1, | |
499 | min_int64, | |
500 | max_int64, | |
501 | min_timestamp_seconds - 1, | |
502 | max_timestamp_seconds + 1, | |
503 | min_timestamp_seconds, | |
504 | max_timestamp_seconds}; | |
505 | static const char* expected = R"expected([ | |
506 | null, | |
507 | null, | |
508 | null, | |
509 | null, | |
510 | <value out of range: -9223372036854775808>, | |
511 | <value out of range: 9223372036854775807>, | |
512 | <value out of range: -1096193779201>, | |
513 | <value out of range: 971890963200>, | |
514 | -32767-01-01 00:00:00, | |
515 | 32767-12-31 23:59:59 | |
516 | ])expected"; | |
517 | CheckPrimitive<TimestampType, int64_t>(timestamp(TimeUnit::SECOND), {0, 10}, is_valid, | |
518 | values, expected); | |
519 | } | |
520 | { | |
521 | std::vector<int64_t> values = {min_int64, | |
522 | max_int64, | |
523 | min_timestamp_millis - 1, | |
524 | max_timestamp_millis + 1, | |
525 | min_int64, | |
526 | max_int64, | |
527 | min_timestamp_millis - 1, | |
528 | max_timestamp_millis + 1, | |
529 | min_timestamp_millis, | |
530 | max_timestamp_millis}; | |
531 | static const char* expected = R"expected([ | |
532 | null, | |
533 | null, | |
534 | null, | |
535 | null, | |
536 | <value out of range: -9223372036854775808>, | |
537 | <value out of range: 9223372036854775807>, | |
538 | <value out of range: -1096193779200001>, | |
539 | <value out of range: 971890963200000>, | |
540 | -32767-01-01 00:00:00.000, | |
541 | 32767-12-31 23:59:59.999 | |
542 | ])expected"; | |
543 | CheckPrimitive<TimestampType, int64_t>(timestamp(TimeUnit::MILLI), {0, 10}, is_valid, | |
544 | values, expected); | |
545 | } | |
546 | { | |
547 | std::vector<int64_t> values = {min_int64, | |
548 | max_int64, | |
549 | min_timestamp_micros - 1, | |
550 | max_timestamp_micros + 1, | |
551 | min_int64, | |
552 | max_int64, | |
553 | min_timestamp_micros - 1, | |
554 | max_timestamp_micros + 1, | |
555 | min_timestamp_micros, | |
556 | max_timestamp_micros}; | |
557 | static const char* expected = R"expected([ | |
558 | null, | |
559 | null, | |
560 | null, | |
561 | null, | |
562 | <value out of range: -9223372036854775808>, | |
563 | <value out of range: 9223372036854775807>, | |
564 | <value out of range: -1096193779200000001>, | |
565 | <value out of range: 971890963200000000>, | |
566 | -32767-01-01 00:00:00.000000, | |
567 | 32767-12-31 23:59:59.999999 | |
568 | ])expected"; | |
569 | CheckPrimitive<TimestampType, int64_t>(timestamp(TimeUnit::MICRO), {0, 10}, is_valid, | |
570 | values, expected); | |
571 | } | |
572 | // Note that while the values below are legal and correct, they used to | |
573 | // trigger an internal signed overflow inside the vendored "date" library | |
574 | // (https://github.com/HowardHinnant/date/issues/696). | |
575 | { | |
576 | std::vector<int64_t> values = {min_int64, max_int64}; | |
577 | static const char* expected = R"expected([ | |
578 | 1677-09-21 00:12:43.145224192, | |
579 | 2262-04-11 23:47:16.854775807 | |
580 | ])expected"; | |
581 | CheckPrimitive<TimestampType, int64_t>(timestamp(TimeUnit::NANO), {0, 10}, | |
582 | {true, true}, values, expected); | |
583 | } | |
584 | } | |
585 | ||
586 | TEST_F(TestPrettyPrint, StructTypeBasic) { | |
587 | auto simple_1 = field("one", int32()); | |
588 | auto simple_2 = field("two", int32()); | |
589 | auto simple_struct = struct_({simple_1, simple_2}); | |
590 | ||
591 | auto array = ArrayFromJSON(simple_struct, "[[11, 22]]"); | |
592 | ||
593 | static const char* ex = R"expected(-- is_valid: all not null | |
594 | -- child 0 type: int32 | |
595 | [ | |
596 | 11 | |
597 | ] | |
598 | -- child 1 type: int32 | |
599 | [ | |
600 | 22 | |
601 | ])expected"; | |
602 | CheckStream(*array, {0, 10}, ex); | |
603 | ||
604 | static const char* ex_2 = R"expected( -- is_valid: all not null | |
605 | -- child 0 type: int32 | |
606 | [ | |
607 | 11 | |
608 | ] | |
609 | -- child 1 type: int32 | |
610 | [ | |
611 | 22 | |
612 | ])expected"; | |
613 | CheckStream(*array, {2, 10}, ex_2); | |
614 | } | |
615 | ||
616 | TEST_F(TestPrettyPrint, StructTypeAdvanced) { | |
617 | auto simple_1 = field("one", int32()); | |
618 | auto simple_2 = field("two", int32()); | |
619 | auto simple_struct = struct_({simple_1, simple_2}); | |
620 | ||
621 | auto array = ArrayFromJSON(simple_struct, "[[11, 22], null, [null, 33]]"); | |
622 | ||
623 | static const char* ex = R"expected(-- is_valid: | |
624 | [ | |
625 | true, | |
626 | false, | |
627 | true | |
628 | ] | |
629 | -- child 0 type: int32 | |
630 | [ | |
631 | 11, | |
632 | 0, | |
633 | null | |
634 | ] | |
635 | -- child 1 type: int32 | |
636 | [ | |
637 | 22, | |
638 | 0, | |
639 | 33 | |
640 | ])expected"; | |
641 | CheckStream(*array, {0, 10}, ex); | |
642 | } | |
643 | ||
644 | TEST_F(TestPrettyPrint, BinaryType) { | |
645 | std::vector<bool> is_valid = {true, true, false, true, true, true}; | |
646 | std::vector<std::string> values = {"foo", "bar", "", "baz", "", "\xff"}; | |
647 | static const char* ex = "[\n 666F6F,\n 626172,\n null,\n 62617A,\n ,\n FF\n]"; | |
648 | CheckPrimitive<BinaryType, std::string>({0}, is_valid, values, ex); | |
649 | CheckPrimitive<LargeBinaryType, std::string>({0}, is_valid, values, ex); | |
650 | static const char* ex_in2 = | |
651 | " [\n 666F6F,\n 626172,\n null,\n 62617A,\n ,\n FF\n ]"; | |
652 | CheckPrimitive<BinaryType, std::string>({2}, is_valid, values, ex_in2); | |
653 | CheckPrimitive<LargeBinaryType, std::string>({2}, is_valid, values, ex_in2); | |
654 | } | |
655 | ||
656 | TEST_F(TestPrettyPrint, BinaryNoNewlines) { | |
657 | std::vector<bool> is_valid = {true, true, false, true, true, true}; | |
658 | std::vector<std::string> values = {"foo", "bar", "", "baz", "", "\xff"}; | |
659 | ||
660 | PrettyPrintOptions options{}; | |
661 | options.skip_new_lines = true; | |
662 | ||
663 | const char* expected = "[666F6F,626172,null,62617A,,FF]"; | |
664 | CheckPrimitive<BinaryType, std::string>(options, is_valid, values, expected, false); | |
665 | ||
666 | // With ellipsis | |
667 | options.window = 2; | |
668 | expected = "[666F6F,626172,...,,FF]"; | |
669 | CheckPrimitive<BinaryType, std::string>(options, is_valid, values, expected, false); | |
670 | } | |
671 | ||
672 | TEST_F(TestPrettyPrint, ListType) { | |
673 | auto list_type = list(int64()); | |
674 | ||
675 | static const char* ex = R"expected([ | |
676 | [ | |
677 | null | |
678 | ], | |
679 | [], | |
680 | null, | |
681 | [ | |
682 | 4, | |
683 | 6, | |
684 | 7 | |
685 | ], | |
686 | [ | |
687 | 2, | |
688 | 3 | |
689 | ] | |
690 | ])expected"; | |
691 | static const char* ex_2 = R"expected( [ | |
692 | [ | |
693 | null | |
694 | ], | |
695 | [], | |
696 | null, | |
697 | [ | |
698 | 4, | |
699 | 6, | |
700 | 7 | |
701 | ], | |
702 | [ | |
703 | 2, | |
704 | 3 | |
705 | ] | |
706 | ])expected"; | |
707 | static const char* ex_3 = R"expected([ | |
708 | [ | |
709 | null | |
710 | ], | |
711 | ... | |
712 | [ | |
713 | 2, | |
714 | 3 | |
715 | ] | |
716 | ])expected"; | |
717 | ||
718 | auto array = ArrayFromJSON(list_type, "[[null], [], null, [4, 6, 7], [2, 3]]"); | |
719 | CheckArray(*array, {0, 10}, ex); | |
720 | CheckArray(*array, {2, 10}, ex_2); | |
721 | CheckStream(*array, {0, 1}, ex_3); | |
722 | ||
723 | list_type = large_list(int64()); | |
724 | array = ArrayFromJSON(list_type, "[[null], [], null, [4, 6, 7], [2, 3]]"); | |
725 | CheckArray(*array, {0, 10}, ex); | |
726 | CheckArray(*array, {2, 10}, ex_2); | |
727 | CheckStream(*array, {0, 1}, ex_3); | |
728 | } | |
729 | ||
730 | TEST_F(TestPrettyPrint, ListTypeNoNewlines) { | |
731 | auto list_type = list(int64()); | |
732 | auto empty_array = ArrayFromJSON(list_type, "[]"); | |
733 | auto array = ArrayFromJSON(list_type, "[[null], [], null, [4, 5, 6, 7, 8], [2, 3]]"); | |
734 | ||
735 | PrettyPrintOptions options{}; | |
736 | options.skip_new_lines = true; | |
737 | options.null_rep = "NA"; | |
738 | CheckArray(*empty_array, options, "[]", false); | |
739 | CheckArray(*array, options, "[[NA],[],NA,[4,5,6,7,8],[2,3]]", false); | |
740 | ||
741 | options.window = 2; | |
742 | CheckArray(*empty_array, options, "[]", false); | |
743 | CheckArray(*array, options, "[[NA],[],...,[4,5,...,7,8],[2,3]]", false); | |
744 | } | |
745 | ||
746 | TEST_F(TestPrettyPrint, MapType) { | |
747 | auto map_type = map(utf8(), int64()); | |
748 | auto array = ArrayFromJSON(map_type, R"([ | |
749 | [["joe", 0], ["mark", null]], | |
750 | null, | |
751 | [["cap", 8]], | |
752 | [] | |
753 | ])"); | |
754 | ||
755 | static const char* ex = R"expected([ | |
756 | keys: | |
757 | [ | |
758 | "joe", | |
759 | "mark" | |
760 | ] | |
761 | values: | |
762 | [ | |
763 | 0, | |
764 | null | |
765 | ], | |
766 | null, | |
767 | keys: | |
768 | [ | |
769 | "cap" | |
770 | ] | |
771 | values: | |
772 | [ | |
773 | 8 | |
774 | ], | |
775 | keys: | |
776 | [] | |
777 | values: | |
778 | [] | |
779 | ])expected"; | |
780 | CheckArray(*array, {0, 10}, ex); | |
781 | } | |
782 | ||
783 | TEST_F(TestPrettyPrint, FixedSizeListType) { | |
784 | auto list_type = fixed_size_list(int32(), 3); | |
785 | auto array = ArrayFromJSON(list_type, | |
786 | "[[null, 0, 1], [2, 3, null], null, [4, 6, 7], [8, 9, 5]]"); | |
787 | ||
788 | CheckArray(*array, {0, 10}, R"expected([ | |
789 | [ | |
790 | null, | |
791 | 0, | |
792 | 1 | |
793 | ], | |
794 | [ | |
795 | 2, | |
796 | 3, | |
797 | null | |
798 | ], | |
799 | null, | |
800 | [ | |
801 | 4, | |
802 | 6, | |
803 | 7 | |
804 | ], | |
805 | [ | |
806 | 8, | |
807 | 9, | |
808 | 5 | |
809 | ] | |
810 | ])expected"); | |
811 | CheckStream(*array, {0, 1}, R"expected([ | |
812 | [ | |
813 | null, | |
814 | ... | |
815 | 1 | |
816 | ], | |
817 | ... | |
818 | [ | |
819 | 8, | |
820 | ... | |
821 | 5 | |
822 | ] | |
823 | ])expected"); | |
824 | } | |
825 | ||
826 | TEST_F(TestPrettyPrint, FixedSizeBinaryType) { | |
827 | std::vector<bool> is_valid = {true, true, false, true, false}; | |
828 | ||
829 | auto type = fixed_size_binary(3); | |
830 | auto array = ArrayFromJSON(type, "[\"foo\", \"bar\", null, \"baz\"]"); | |
831 | ||
832 | static const char* ex = "[\n 666F6F,\n 626172,\n null,\n 62617A\n]"; | |
833 | CheckArray(*array, {0, 10}, ex); | |
834 | static const char* ex_2 = " [\n 666F6F,\n ...\n 62617A\n ]"; | |
835 | CheckArray(*array, {2, 1}, ex_2); | |
836 | } | |
837 | ||
838 | TEST_F(TestPrettyPrint, DecimalTypes) { | |
839 | int32_t p = 19; | |
840 | int32_t s = 4; | |
841 | ||
842 | for (auto type : {decimal128(p, s), decimal256(p, s)}) { | |
843 | auto array = ArrayFromJSON(type, "[\"123.4567\", \"456.7891\", null]"); | |
844 | ||
845 | static const char* ex = "[\n 123.4567,\n 456.7891,\n null\n]"; | |
846 | CheckArray(*array, {0}, ex); | |
847 | } | |
848 | } | |
849 | ||
850 | TEST_F(TestPrettyPrint, DictionaryType) { | |
851 | std::vector<bool> is_valid = {true, true, false, true, true, true}; | |
852 | ||
853 | std::shared_ptr<Array> dict; | |
854 | std::vector<std::string> dict_values = {"foo", "bar", "baz"}; | |
855 | ArrayFromVector<StringType, std::string>(dict_values, &dict); | |
856 | std::shared_ptr<DataType> dict_type = dictionary(int16(), utf8()); | |
857 | ||
858 | std::shared_ptr<Array> indices; | |
859 | std::vector<int16_t> indices_values = {1, 2, -1, 0, 2, 0}; | |
860 | ArrayFromVector<Int16Type, int16_t>(is_valid, indices_values, &indices); | |
861 | auto arr = std::make_shared<DictionaryArray>(dict_type, indices, dict); | |
862 | ||
863 | static const char* expected = R"expected( | |
864 | -- dictionary: | |
865 | [ | |
866 | "foo", | |
867 | "bar", | |
868 | "baz" | |
869 | ] | |
870 | -- indices: | |
871 | [ | |
872 | 1, | |
873 | 2, | |
874 | null, | |
875 | 0, | |
876 | 2, | |
877 | 0 | |
878 | ])expected"; | |
879 | ||
880 | CheckArray(*arr, {0}, expected); | |
881 | } | |
882 | ||
883 | TEST_F(TestPrettyPrint, ChunkedArrayPrimitiveType) { | |
884 | auto array = ArrayFromJSON(int32(), "[0, 1, null, 3, null]"); | |
885 | ChunkedArray chunked_array(array); | |
886 | ||
887 | static const char* expected = R"expected([ | |
888 | [ | |
889 | 0, | |
890 | 1, | |
891 | null, | |
892 | 3, | |
893 | null | |
894 | ] | |
895 | ])expected"; | |
896 | CheckStream(chunked_array, {0}, expected); | |
897 | ||
898 | ChunkedArray chunked_array_2({array, array}); | |
899 | ||
900 | static const char* expected_2 = R"expected([ | |
901 | [ | |
902 | 0, | |
903 | 1, | |
904 | null, | |
905 | 3, | |
906 | null | |
907 | ], | |
908 | [ | |
909 | 0, | |
910 | 1, | |
911 | null, | |
912 | 3, | |
913 | null | |
914 | ] | |
915 | ])expected"; | |
916 | ||
917 | CheckStream(chunked_array_2, {0}, expected_2); | |
918 | } | |
919 | ||
920 | TEST_F(TestPrettyPrint, TablePrimitive) { | |
921 | std::shared_ptr<Field> int_field = field("column", int32()); | |
922 | auto array = ArrayFromJSON(int_field->type(), "[0, 1, null, 3, null]"); | |
923 | auto column = std::make_shared<ChunkedArray>(ArrayVector({array})); | |
924 | std::shared_ptr<Schema> table_schema = schema({int_field}); | |
925 | std::shared_ptr<Table> table = Table::Make(table_schema, {column}); | |
926 | ||
927 | static const char* expected = R"expected(column: int32 | |
928 | ---- | |
929 | column: | |
930 | [ | |
931 | [ | |
932 | 0, | |
933 | 1, | |
934 | null, | |
935 | 3, | |
936 | null | |
937 | ] | |
938 | ] | |
939 | )expected"; | |
940 | CheckStream(*table, {0}, expected); | |
941 | } | |
942 | ||
943 | TEST_F(TestPrettyPrint, SchemaWithDictionary) { | |
944 | std::vector<bool> is_valid = {true, true, false, true, true, true}; | |
945 | ||
946 | std::shared_ptr<Array> dict; | |
947 | std::vector<std::string> dict_values = {"foo", "bar", "baz"}; | |
948 | ArrayFromVector<StringType, std::string>(dict_values, &dict); | |
949 | ||
950 | auto simple = field("one", int32()); | |
951 | auto simple_dict = field("two", dictionary(int16(), utf8())); | |
952 | auto list_of_dict = field("three", list(simple_dict)); | |
953 | auto struct_with_dict = field("four", struct_({simple, simple_dict})); | |
954 | ||
955 | auto sch = schema({simple, simple_dict, list_of_dict, struct_with_dict}); | |
956 | ||
957 | static const char* expected = R"expected(one: int32 | |
958 | two: dictionary<values=string, indices=int16, ordered=0> | |
959 | three: list<two: dictionary<values=string, indices=int16, ordered=0>> | |
960 | child 0, two: dictionary<values=string, indices=int16, ordered=0> | |
961 | four: struct<one: int32, two: dictionary<values=string, indices=int16, ordered=0>> | |
962 | child 0, one: int32 | |
963 | child 1, two: dictionary<values=string, indices=int16, ordered=0>)expected"; | |
964 | ||
965 | PrettyPrintOptions options; | |
966 | Check(*sch, options, expected); | |
967 | } | |
968 | ||
969 | TEST_F(TestPrettyPrint, SchemaWithNotNull) { | |
970 | auto simple = field("one", int32()); | |
971 | auto non_null = field("two", int32(), false); | |
972 | auto list_simple = field("three", list(int32())); | |
973 | auto list_non_null = field("four", list(int32()), false); | |
974 | auto list_non_null2 = field("five", list(field("item", int32(), false))); | |
975 | ||
976 | auto sch = schema({simple, non_null, list_simple, list_non_null, list_non_null2}); | |
977 | ||
978 | static const char* expected = R"expected(one: int32 | |
979 | two: int32 not null | |
980 | three: list<item: int32> | |
981 | child 0, item: int32 | |
982 | four: list<item: int32> not null | |
983 | child 0, item: int32 | |
984 | five: list<item: int32 not null> | |
985 | child 0, item: int32 not null)expected"; | |
986 | ||
987 | PrettyPrintOptions options; | |
988 | Check(*sch, options, expected); | |
989 | } | |
990 | ||
991 | TEST_F(TestPrettyPrint, SchemaWithMetadata) { | |
992 | // ARROW-7063 | |
993 | auto metadata1 = key_value_metadata({"foo1"}, {"bar1"}); | |
994 | auto metadata2 = key_value_metadata({"foo2"}, {"bar2"}); | |
995 | auto metadata3 = key_value_metadata( | |
996 | {"foo3", "lorem"}, | |
997 | {"bar3", | |
998 | R"(Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nulla accumsan vel | |
999 | turpis et mollis. Aliquam tincidunt arcu id tortor blandit blandit. Donec | |
1000 | eget leo quis lectus scelerisque varius. Class aptent taciti sociosqu ad | |
1001 | litora torquent per conubia nostra, per inceptos himenaeos. Praesent | |
1002 | faucibus, diam eu volutpat iaculis, tellus est porta ligula, a efficitur | |
1003 | turpis nulla facilisis quam. Aliquam vitae lorem erat. Proin a dolor ac libero | |
1004 | dignissim mollis vitae eu mauris. Quisque posuere tellus vitae massa | |
1005 | pellentesque sagittis. Aenean feugiat, diam ac dignissim fermentum, lorem | |
1006 | sapien commodo massa, vel volutpat orci nisi eu justo. Nulla non blandit | |
1007 | sapien. Quisque pretium vestibulum urna eu vehicula.)"}); | |
1008 | auto my_schema = schema( | |
1009 | {field("one", int32(), true, metadata1), field("two", utf8(), false, metadata2)}, | |
1010 | metadata3); | |
1011 | ||
1012 | PrettyPrintOptions options; | |
1013 | static const char* expected = R"(one: int32 | |
1014 | -- field metadata -- | |
1015 | foo1: 'bar1' | |
1016 | two: string not null | |
1017 | -- field metadata -- | |
1018 | foo2: 'bar2' | |
1019 | -- schema metadata -- | |
1020 | foo3: 'bar3' | |
1021 | lorem: 'Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nulla ac' + 737)"; | |
1022 | Check(*my_schema, options, expected); | |
1023 | ||
1024 | static const char* expected_verbose = R"(one: int32 | |
1025 | -- field metadata -- | |
1026 | foo1: 'bar1' | |
1027 | two: string not null | |
1028 | -- field metadata -- | |
1029 | foo2: 'bar2' | |
1030 | -- schema metadata -- | |
1031 | foo3: 'bar3' | |
1032 | lorem: 'Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nulla accumsan vel | |
1033 | turpis et mollis. Aliquam tincidunt arcu id tortor blandit blandit. Donec | |
1034 | eget leo quis lectus scelerisque varius. Class aptent taciti sociosqu ad | |
1035 | litora torquent per conubia nostra, per inceptos himenaeos. Praesent | |
1036 | faucibus, diam eu volutpat iaculis, tellus est porta ligula, a efficitur | |
1037 | turpis nulla facilisis quam. Aliquam vitae lorem erat. Proin a dolor ac libero | |
1038 | dignissim mollis vitae eu mauris. Quisque posuere tellus vitae massa | |
1039 | pellentesque sagittis. Aenean feugiat, diam ac dignissim fermentum, lorem | |
1040 | sapien commodo massa, vel volutpat orci nisi eu justo. Nulla non blandit | |
1041 | sapien. Quisque pretium vestibulum urna eu vehicula.')"; | |
1042 | options.truncate_metadata = false; | |
1043 | Check(*my_schema, options, expected_verbose); | |
1044 | ||
1045 | // Metadata that exactly fits | |
1046 | auto metadata4 = | |
1047 | key_value_metadata({"key"}, {("valuexxxxxxxxxxxxxxxxxxxxxxxxxxxxx" | |
1048 | "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx")}); | |
1049 | my_schema = schema({field("f0", int32())}, metadata4); | |
1050 | static const char* expected_fits = R"(f0: int32 | |
1051 | -- schema metadata -- | |
1052 | key: 'valuexxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx')"; | |
1053 | options.truncate_metadata = false; | |
1054 | Check(*my_schema, options, expected_fits); | |
1055 | ||
1056 | // A large key | |
1057 | auto metadata5 = key_value_metadata({"0123456789012345678901234567890123456789"}, | |
1058 | {("valuexxxxxxxxxxxxxxxxxxxxxxxxxxxxx" | |
1059 | "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx")}); | |
1060 | my_schema = schema({field("f0", int32())}, metadata5); | |
1061 | static const char* expected_big_key = R"(f0: int32 | |
1062 | -- schema metadata -- | |
1063 | 0123456789012345678901234567890123456789: 'valuexxxxxxxxxxxxxxxxxxxxxxxxx' + 40)"; | |
1064 | options.truncate_metadata = true; | |
1065 | Check(*my_schema, options, expected_big_key); | |
1066 | } | |
1067 | ||
1068 | TEST_F(TestPrettyPrint, SchemaIndentation) { | |
1069 | // ARROW-6159 | |
1070 | auto simple = field("one", int32()); | |
1071 | auto non_null = field("two", int32(), false); | |
1072 | auto sch = schema({simple, non_null}); | |
1073 | ||
1074 | static const char* expected = R"expected( one: int32 | |
1075 | two: int32 not null)expected"; | |
1076 | ||
1077 | PrettyPrintOptions options(/*indent=*/4); | |
1078 | Check(*sch, options, expected); | |
1079 | } | |
1080 | ||
1081 | } // namespace arrow |