-
Notifications
You must be signed in to change notification settings - Fork 11
/
netcdfarray.py
601 lines (431 loc) · 17.1 KB
/
netcdfarray.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
import netCDF4
import numpy
from . import abstract
from .numpyarray import NumpyArray
class NetCDFArray(abstract.Array):
"""An underlying array stored in a netCDF file.
.. versionadded:: (cfdm) 1.7.0
"""
def __init__(
self,
filename=None,
ncvar=None,
varid=None,
group=None,
dtype=None,
ndim=None,
shape=None,
size=None,
mask=True,
units=False,
calendar=False,
missing_values=None,
source=None,
copy=True,
):
"""**Initialisation**
:Parameters:
filename: `str`
The name of the netCDF file containing the array.
ncvar: `str`, optional
The name of the netCDF variable containing the
array. Required unless *varid* is set.
varid: `int`, optional
The UNIDATA netCDF interface ID of the variable
containing the array. Required if *ncvar* is not set,
ignored if *ncvar* is set.
group: `None` or sequence of `str`, optional
Specify the netCDF4 group to which the netCDF variable
belongs. By default, or if *group* is `None` or an
empty sequence, it assumed to be in the root
group. The last element in the sequence is the name of
the group in which the variable lies, with other
elements naming any parent groups (excluding the root
group).
*Parameter example:*
To specify that a variable is in the root group:
``group=()`` or ``group=None``
*Parameter example:*
To specify that a variable is in the group '/forecasts':
``group=['forecasts']``
*Parameter example:*
To specify that a variable is in the group
'/forecasts/model2': ``group=['forecasts', 'model2']``
.. versionadded:: (cfdm) 1.8.6.0
dtype: `numpy.dtype`
The data type of the array in the netCDF file. May be
`None` if the numpy data-type is not known (which can be
the case for netCDF string types, for example).
shape: `tuple`
The array dimension sizes in the netCDF file.
size: `int`
Number of elements in the array in the netCDF file.
ndim: `int`
The number of array dimensions in the netCDF file.
mask: `bool`
If True (the default) then mask by convention when
reading data from disk.
A netCDF array is masked depending on the values of any of
the netCDF variable attributes ``valid_min``,
``valid_max``, ``valid_range``, ``_FillValue`` and
``missing_value``.
.. versionadded:: (cfdm) 1.8.2
units: `str` or `None`, optional
The units of the netCDF variable. Set to `None` to
indicate that there are no units. If unset then the
units will be set during the first `__getitem__` call.
.. versionadded:: (cfdm) 1.10.0.1
calendar: `str` or `None`, optional
The calendar of the netCDF variable. By default, or if
set to `None`, then the CF default calendar is
assumed, if applicable. If unset then the calendar
will be set during the first `__getitem__` call.
.. versionadded:: (cfdm) 1.10.0.1
missing_values: `dict`, optional
The missing value indicators defined by the netCDF
variable attributes. See `get_missing_values` for
details.
.. versionadded:: (cfdm) 1.10.0.3
{{init source: optional}}
.. versionadded:: (cfdm) 1.10.0.0
{{init copy: `bool`, optional}}
.. versionadded:: (cfdm) 1.10.0.0
"""
super().__init__(source=source, copy=copy)
if source is not None:
try:
shape = source._get_component("shape", None)
except AttributeError:
shape = None
try:
filename = source._get_component("filename", None)
except AttributeError:
filename = None
try:
ncvar = source._get_component("ncvar", None)
except AttributeError:
ncvar = None
try:
varid = source._get_component("varid", None)
except AttributeError:
varid = None
try:
group = source._get_component("group", None)
except AttributeError:
group = None
try:
dtype = source._get_component("dtype", None)
except AttributeError:
dtype = None
try:
mask = source._get_component("mask", True)
except AttributeError:
mask = True
try:
units = source._get_component("units", False)
except AttributeError:
units = False
try:
calendar = source._get_component("calendar", False)
except AttributeError:
calendar = False
try:
missing_values = source._get_component("missing_values", None)
except AttributeError:
missing_values = None
if shape is not None:
self._set_component("shape", shape, copy=False)
if filename is not None:
self._set_component("filename", filename, copy=False)
if ncvar is not None:
self._set_component("ncvar", ncvar, copy=False)
if varid is not None:
self._set_component("varid", varid, copy=False)
if missing_values is not None:
self._set_component(
"missing_values", missing_values.copy(), copy=False
)
self._set_component("group", group, copy=False)
self._set_component("dtype", dtype, copy=False)
self._set_component("mask", mask, copy=False)
self._set_component("units", units, copy=False)
self._set_component("calendar", calendar, copy=False)
# By default, close the netCDF file after data array access
self._set_component("close", True, copy=False)
def __getitem__(self, indices):
"""Returns a subspace of the array as a numpy array.
x.__getitem__(indices) <==> x[indices]
The indices that define the subspace must be either `Ellipsis` or
a sequence that contains an index for each dimension. In the
latter case, each dimension's index must either be a `slice`
object or a sequence of two or more integers.
Indexing is similar to numpy indexing. The only difference to
numpy indexing (given the restrictions on the type of indices
allowed) is:
* When two or more dimension's indices are sequences of integers
then these indices work independently along each dimension
(similar to the way vector subscripts work in Fortran).
.. versionadded:: (cfdm) 1.7.0
"""
netcdf = self.open()
dataset = netcdf
# Traverse the group structure, if there is one (CF>=1.8).
group = self.get_group()
if group:
for g in group[:-1]:
netcdf = netcdf.groups[g]
netcdf = netcdf.groups[group[-1]]
ncvar = self.get_ncvar()
mask = self.get_mask()
if ncvar is not None:
# Get the variable by netCDF name
variable = netcdf.variables[ncvar]
variable.set_auto_mask(mask)
array = variable[indices]
else:
# Get the variable by netCDF ID
varid = self.get_varid()
for variable in netcdf.variables.values():
if variable._varid == varid:
variable.set_auto_mask(mask)
array = variable[indices]
break
# Set the units, if they haven't been set already.
self._set_units(variable)
self.close(dataset)
del netcdf, dataset
string_type = isinstance(array, str)
if string_type:
# --------------------------------------------------------
# A netCDF string type scalar variable comes out as Python
# str object, so convert it to a numpy array.
# --------------------------------------------------------
array = numpy.array(array, dtype=f"S{len(array)}")
if not self.ndim:
# Hmm netCDF4 has a thing for making scalar size 1 , 1d
array = array.squeeze()
kind = array.dtype.kind
if not string_type and kind in "SU":
# == 'S' and array.ndim > (self.ndim -
# getattr(self, 'gathered', 0) -
# getattr(self, 'ragged', 0)):
# --------------------------------------------------------
# Collapse (by concatenation) the outermost (fastest
# varying) dimension of char array into
# memory. E.g. [['a','b','c']] becomes ['abc']
# --------------------------------------------------------
if kind == "U":
array = array.astype("S")
array = netCDF4.chartostring(array)
shape = array.shape
array = numpy.array([x.rstrip() for x in array.flat], dtype="S")
array = numpy.reshape(array, shape)
array = numpy.ma.masked_where(array == b"", array)
elif not string_type and kind == "O":
# --------------------------------------------------------
# A netCDF string type N-d (N>=1) variable comes out as a
# numpy object array, so convert it to numpy string array.
# --------------------------------------------------------
array = array.astype("S") # , copy=False)
# --------------------------------------------------------
# netCDF4 does not auto-mask VLEN variable, so do it here.
# --------------------------------------------------------
array = numpy.ma.where(array == b"", numpy.ma.masked, array)
return array
def __repr__(self):
"""Called by the `repr` built-in function.
x.__repr__() <==> repr(x)
"""
return f"<{self.__class__.__name__}{self.shape}: {self}>"
def __str__(self):
"""Called by the `str` built-in function.
x.__str__() <==> str(x)
"""
return f"{self.get_filename(None)}, {self.get_address()}"
def _set_units(self, var):
"""The units and calendar properties.
These are set from the netCDF variable attributes, but only if
they have already not been defined, either during {{class}}
instantiation or by a previous call to `_set_units`.
.. versionadded:: (cfdm) 1.10.0.1
:Parameters:
var: `netCDF4.Variable`
The variable containing the units and calendar
definitions.
:Returns:
`tuple`
The units and calendar values, either of which may be
`None`.
"""
# Note: Can't use None as the default since it is a valid
# `units` or 'calendar' value that indicates that the
# attribute has not been set in the dataset.
units = self._get_component("units", False)
if units is False:
try:
units = var.getncattr("units")
except AttributeError:
units = None
self._set_component("units", units, copy=False)
calendar = self._get_component("calendar", False)
if calendar is False:
try:
calendar = var.getncattr("calendar")
except AttributeError:
calendar = None
self._set_component("calendar", calendar, copy=False)
return units, calendar
@property
def array(self):
"""Return an independent numpy array containing the data.
.. versionadded:: (cfdm) 1.7.0
:Returns:
`numpy.ndarray`
An independent numpy array of the data.
**Examples**
>>> n = numpy.asanyarray(a)
>>> isinstance(n, numpy.ndarray)
True
"""
return self[...]
@property
def dtype(self):
"""Data-type of the data elements.
.. versionadded:: (cfdm) 1.7.0
"""
return self._get_component("dtype")
@property
def file_address(self):
"""The file name and address.
.. versionadded:: (cfdm) 1.10.0.0
:Returns:
`tuple`
The file name and file address.
**Examples**
>>> a.file_address()
('file.nc', 'latitude')
"""
pointer = self._get_component("ncvar", None)
if pointer is None:
pointer = self.get_varid()
return (self.get_filename(None), pointer)
@property
def shape(self):
"""Tuple of array dimension sizes.
.. versionadded:: (cfdm) 1.7.0
"""
return self._get_component("shape")
def get_address(self):
"""The address in the file of the variable.
Either the netCDF variable name, or else the UNIDATA netCDF
interface ID.
.. versionadded:: (cfdm) 1.10.0.1
.. seealso:: `get_filename`, `get_varid`
:Returns:
`str` or `None`
The address, or `None` if there isn't one.
"""
address = self.get_ncvar()
if address is None:
address = self.get_varid()
return address
def get_group(self):
"""The netCDF4 group structure of the netCDF variable.
.. versionadded:: (cfdm) 1.8.6.0
**Examples**
>>> b = a.get_group()
"""
return self._get_component("group")
def get_mask(self):
"""Whether or not to automatically mask the data.
.. versionadded:: (cfdm) 1.8.2
**Examples**
>>> b = a.get_mask()
"""
return self._get_component("mask")
def get_missing_values(self):
"""The missing value indicators from the netCDF variable.
.. versionadded:: (cfdm) 1.10.0.3
:Returns:
`dict` or `None`
The missing value indicators from the netCDF variable,
keyed by their netCDF attribute names. An empty
dictionary signifies that no missing values are given
in the file. `None` signifies that the missing values
have not been set.
**Examples**
>>> a.get_missing_values()
None
>>> b.get_missing_values()
{}
>>> c.get_missing_values()
{'missing_value': 1e20, 'valid_range': (-10, 20)}
>>> d.get_missing_values()
{'valid_min': -999}
"""
out = self._get_component("missing_values", None)
if out is None:
return
return out.copy()
def get_ncvar(self):
"""The name of the netCDF variable containing the array.
.. versionadded:: (cfdm) 1.7.0
**Examples**
>>> print(a.netcdf)
'tas'
>>> print(a.varid)
None
>>> print(a.netcdf)
None
>>> print(a.varid)
4
"""
return self._get_component("ncvar")
def get_varid(self):
"""The UNIDATA netCDF interface ID of the array's variable.
.. versionadded:: (cfdm) 1.7.0
**Examples**
>>> print(a.netcdf)
'tas'
>>> print(a.varid)
None
>>> print(a.netcdf)
None
>>> print(a.varid)
4
"""
return self._get_component("varid", None)
def close(self, netcdf):
"""Close the dataset containing the data.
.. versionadded:: (cfdm) 1.7.0
:Parameters:
netcdf: `netCDF4.Dataset`
The netCDF dataset to be be closed.
:Returns:
`None`
"""
if self._get_component("close"):
netcdf.close()
def open(self):
"""Returns an open dataset containing the data array.
.. versionadded:: (cfdm) 1.7.0
:Returns:
`netCDF4.Dataset`
**Examples**
>>> netcdf = a.open()
>>> variable = netcdf.variables[a.get_ncvar()]
>>> variable.getncattr('standard_name')
'eastward_wind'
"""
try:
return netCDF4.Dataset(self.get_filename(None), "r")
except RuntimeError as error:
raise RuntimeError(f"{error}: {self.get_filename(None)}")
def to_memory(self):
"""Bring data on disk into memory.
.. versionadded:: (cfdm) 1.7.0
:Returns:
`NumpyArray`
The new with all of its data in memory.
"""
return NumpyArray(self[...])