Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.8/site-packages/pandas/core/dtypes/inference.py: 52%

1""" basic inference routines """ 

2 

3from __future__ import annotations 

4 

5from collections import abc 

6from numbers import Number 

7import re 

8from typing import Pattern 

9 

10import numpy as np 

11 

12from pandas._libs import lib 

13 

14is_bool = lib.is_bool 

15 

16is_integer = lib.is_integer 

17 

18is_float = lib.is_float 

19 

20is_complex = lib.is_complex 

21 

22is_scalar = lib.is_scalar 

23 

24is_decimal = lib.is_decimal 

25 

26is_interval = lib.is_interval 

27 

28is_list_like = lib.is_list_like 

29 

30is_iterator = lib.is_iterator 

31 

32 

33def is_number(obj) -> bool: 

34 """ 

35 Check if the object is a number. 

36 

37 Returns True when the object is a number, and False if is not. 

38 

39 Parameters 

40 ---------- 

41 obj : any type 

42 The object to check if is a number. 

43 

44 Returns 

45 ------- 

46 bool 

47 Whether `obj` is a number or not. 

48 

49 See Also 

50 -------- 

51 api.types.is_integer: Checks a subgroup of numbers. 

52 

53 Examples 

54 -------- 

55 >>> from pandas.api.types import is_number 

56 >>> is_number(1) 

57 True 

58 >>> is_number(7.15) 

59 True 

60 

61 Booleans are valid because they are int subclass. 

62 

63 >>> is_number(False) 

64 True 

65 

66 >>> is_number("foo") 

67 False 

68 >>> is_number("5") 

69 False 

70 """ 

71 return isinstance(obj, (Number, np.number)) 

72 

73 

74def iterable_not_string(obj) -> bool: 

75 """ 

76 Check if the object is an iterable but not a string. 

77 

78 Parameters 

79 ---------- 

80 obj : The object to check. 

81 

82 Returns 

83 ------- 

84 is_iter_not_string : bool 

85 Whether `obj` is a non-string iterable. 

86 

87 Examples 

88 -------- 

89 >>> iterable_not_string([1, 2, 3]) 

90 True 

91 >>> iterable_not_string("foo") 

92 False 

93 >>> iterable_not_string(1) 

94 False 

95 """ 

96 return isinstance(obj, abc.Iterable) and not isinstance(obj, str) 

97 

98 

99def is_file_like(obj) -> bool: 

100 """ 

101 Check if the object is a file-like object. 

102 

103 For objects to be considered file-like, they must 

104 be an iterator AND have either a `read` and/or `write` 

105 method as an attribute. 

106 

107 Note: file-like objects must be iterable, but 

108 iterable objects need not be file-like. 

109 

110 Parameters 

111 ---------- 

112 obj : The object to check 

113 

114 Returns 

115 ------- 

116 bool 

117 Whether `obj` has file-like properties. 

118 

119 Examples 

120 -------- 

121 >>> import io 

122 >>> from pandas.api.types import is_file_like 

123 >>> buffer = io.StringIO("data") 

124 >>> is_file_like(buffer) 

125 True 

126 >>> is_file_like([1, 2, 3]) 

127 False 

128 """ 

129 if not (hasattr(obj, "read") or hasattr(obj, "write")): 

130 return False 

131 

132 return bool(hasattr(obj, "__iter__")) 

133 

134 

135def is_re(obj) -> bool: 

136 """ 

137 Check if the object is a regex pattern instance. 

138 

139 Parameters 

140 ---------- 

141 obj : The object to check 

142 

143 Returns 

144 ------- 

145 bool 

146 Whether `obj` is a regex pattern. 

147 

148 Examples 

149 -------- 

150 >>> from pandas.api.types import is_re 

151 >>> import re 

152 >>> is_re(re.compile(".*")) 

153 True 

154 >>> is_re("foo") 

155 False 

156 """ 

157 return isinstance(obj, Pattern) 

158 

159 

160def is_re_compilable(obj) -> bool: 

161 """ 

162 Check if the object can be compiled into a regex pattern instance. 

163 

164 Parameters 

165 ---------- 

166 obj : The object to check 

167 

168 Returns 

169 ------- 

170 bool 

171 Whether `obj` can be compiled as a regex pattern. 

172 

173 Examples 

174 -------- 

175 >>> from pandas.api.types import is_re_compilable 

176 >>> is_re_compilable(".*") 

177 True 

178 >>> is_re_compilable(1) 

179 False 

180 """ 

181 try: 

182 re.compile(obj) 

183 except TypeError: 

184 return False 

185 else: 

186 return True 

187 

188 

189def is_array_like(obj) -> bool: 

190 """ 

191 Check if the object is array-like. 

192 

193 For an object to be considered array-like, it must be list-like and 

194 have a `dtype` attribute. 

195 

196 Parameters 

197 ---------- 

198 obj : The object to check 

199 

200 Returns 

201 ------- 

202 is_array_like : bool 

203 Whether `obj` has array-like properties. 

204 

205 Examples 

206 -------- 

207 >>> is_array_like(np.array([1, 2, 3])) 

208 True 

209 >>> is_array_like(pd.Series(["a", "b"])) 

210 True 

211 >>> is_array_like(pd.Index(["2016-01-01"])) 

212 True 

213 >>> is_array_like([1, 2, 3]) 

214 False 

215 >>> is_array_like(("a", "b")) 

216 False 

217 """ 

218 return is_list_like(obj) and hasattr(obj, "dtype") 

219 

220 

221def is_nested_list_like(obj) -> bool: 

222 """ 

223 Check if the object is list-like, and that all of its elements 

224 are also list-like. 

225 

226 Parameters 

227 ---------- 

228 obj : The object to check 

229 

230 Returns 

231 ------- 

232 is_list_like : bool 

233 Whether `obj` has list-like properties. 

234 

235 Examples 

236 -------- 

237 >>> is_nested_list_like([[1, 2, 3]]) 

238 True 

239 >>> is_nested_list_like([{1, 2, 3}, {1, 2, 3}]) 

240 True 

241 >>> is_nested_list_like(["foo"]) 

242 False 

243 >>> is_nested_list_like([]) 

244 False 

245 >>> is_nested_list_like([[1, 2, 3], 1]) 

246 False 

247 

248 Notes 

249 ----- 

250 This won't reliably detect whether a consumable iterator (e. g. 

251 a generator) is a nested-list-like without consuming the iterator. 

252 To avoid consuming it, we always return False if the outer container 

253 doesn't define `__len__`. 

254 

255 See Also 

256 -------- 

257 is_list_like 

258 """ 

259 return ( 

260 is_list_like(obj) 

261 and hasattr(obj, "__len__") 

262 and len(obj) > 0 

263 and all(is_list_like(item) for item in obj) 

264 ) 

265 

266 

267def is_dict_like(obj) -> bool: 

268 """ 

269 Check if the object is dict-like. 

270 

271 Parameters 

272 ---------- 

273 obj : The object to check 

274 

275 Returns 

276 ------- 

277 bool 

278 Whether `obj` has dict-like properties. 

279 

280 Examples 

281 -------- 

282 >>> from pandas.api.types import is_dict_like 

283 >>> is_dict_like({1: 2}) 

284 True 

285 >>> is_dict_like([1, 2, 3]) 

286 False 

287 >>> is_dict_like(dict) 

288 False 

289 >>> is_dict_like(dict()) 

290 True 

291 """ 

292 dict_like_attrs = ("__getitem__", "keys", "__contains__") 

293 return ( 

294 all(hasattr(obj, attr) for attr in dict_like_attrs) 

295 # [GH 25196] exclude classes 

296 and not isinstance(obj, type) 

297 ) 

298 

299 

300def is_named_tuple(obj) -> bool: 

301 """ 

302 Check if the object is a named tuple. 

303 

304 Parameters 

305 ---------- 

306 obj : The object to check 

307 

308 Returns 

309 ------- 

310 bool 

311 Whether `obj` is a named tuple. 

312 

313 Examples 

314 -------- 

315 >>> from collections import namedtuple 

316 >>> from pandas.api.types import is_named_tuple 

317 >>> Point = namedtuple("Point", ["x", "y"]) 

318 >>> p = Point(1, 2) 

319 >>> 

320 >>> is_named_tuple(p) 

321 True 

322 >>> is_named_tuple((1, 2)) 

323 False 

324 """ 

325 return isinstance(obj, abc.Sequence) and hasattr(obj, "_fields") 

326 

327 

328def is_hashable(obj) -> bool: 

329 """ 

330 Return True if hash(obj) will succeed, False otherwise. 

331 

332 Some types will pass a test against collections.abc.Hashable but fail when 

333 they are actually hashed with hash(). 

334 

335 Distinguish between these and other types by trying the call to hash() and 

336 seeing if they raise TypeError. 

337 

338 Returns 

339 ------- 

340 bool 

341 

342 Examples 

343 -------- 

344 >>> import collections 

345 >>> from pandas.api.types import is_hashable 

346 >>> a = ([],) 

347 >>> isinstance(a, collections.abc.Hashable) 

348 True 

349 >>> is_hashable(a) 

350 False 

351 """ 

352 # Unfortunately, we can't use isinstance(obj, collections.abc.Hashable), 

353 # which can be faster than calling hash. That is because numpy scalars 

354 # fail this test. 

355 

356 # Reconsider this decision once this numpy bug is fixed: 

357 # https://github.com/numpy/numpy/issues/5562 

358 

359 try: 

360 hash(obj) 

361 except TypeError: 

362 return False 

363 else: 

364 return True 

365 

366 

367def is_sequence(obj) -> bool: 

368 """ 

369 Check if the object is a sequence of objects. 

370 String types are not included as sequences here. 

371 

372 Parameters 

373 ---------- 

374 obj : The object to check 

375 

376 Returns 

377 ------- 

378 is_sequence : bool 

379 Whether `obj` is a sequence of objects. 

380 

381 Examples 

382 -------- 

383 >>> l = [1, 2, 3] 

384 >>> 

385 >>> is_sequence(l) 

386 True 

387 >>> is_sequence(iter(l)) 

388 False 

389 """ 

390 try: 

391 iter(obj) # Can iterate over it. 

392 len(obj) # Has a length associated with it. 

393 return not isinstance(obj, (str, bytes)) 

394 except (TypeError, AttributeError): 

395 return False 

396 

397 

398def is_dataclass(item): 

399 """ 

400 Checks if the object is a data-class instance 

401 

402 Parameters 

403 ---------- 

404 item : object 

405 

406 Returns 

407 -------- 

408 is_dataclass : bool 

409 True if the item is an instance of a data-class, 

410 will return false if you pass the data class itself 

411 

412 Examples 

413 -------- 

414 >>> from dataclasses import dataclass 

415 >>> @dataclass 

416 ... class Point: 

417 ... x: int 

418 ... y: int 

419 

420 >>> is_dataclass(Point) 

421 False 

422 >>> is_dataclass(Point(0,2)) 

423 True 

424 

425 """ 

426 try: 

427 import dataclasses 

428 

429 return dataclasses.is_dataclass(item) and not isinstance(item, type) 

430 except ImportError: 

431 return False