1 | """ |
---|
2 | NXcanSAS data reader for reading HDF5 formatted CanSAS files. |
---|
3 | """ |
---|
4 | |
---|
5 | import h5py |
---|
6 | import numpy as np |
---|
7 | import re |
---|
8 | import os |
---|
9 | import sys |
---|
10 | |
---|
11 | from ..data_info import plottable_1D, plottable_2D,\ |
---|
12 | Data1D, Data2D, DataInfo, Process, Aperture, Collimation, \ |
---|
13 | TransmissionSpectrum, Detector |
---|
14 | from ..loader_exceptions import FileContentsException, DefaultReaderException |
---|
15 | from ..file_reader_base_class import FileReader, decode |
---|
16 | |
---|
17 | try: |
---|
18 | basestring |
---|
19 | except NameError: # CRUFT: python 2 support |
---|
20 | basestring = str |
---|
21 | |
---|
22 | |
---|
23 | def h5attr(node, key, default=None): |
---|
24 | return decode(node.attrs.get(key, default)) |
---|
25 | |
---|
26 | |
---|
27 | class Reader(FileReader): |
---|
28 | """ |
---|
29 | A class for reading in NXcanSAS data files. The current implementation has |
---|
30 | been tested to load data generated by multiple facilities, all of which are |
---|
31 | known to produce NXcanSAS standards compliant data. Any number of data sets |
---|
32 | may be present within the file and any dimensionality of data may be used. |
---|
33 | Currently 1D and 2D SAS data sets are supported, but should be immediately |
---|
34 | extensible to SESANS data. |
---|
35 | |
---|
36 | Any number of SASdata groups may be present in a SASentry and the data |
---|
37 | within each SASdata group can be a single 1D I(Q), multi-framed 1D I(Q), |
---|
38 | 2D I(Qx, Qy) or multi-framed 2D I(Qx, Qy). |
---|
39 | |
---|
40 | :Dependencies: |
---|
41 | The NXcanSAS HDF5 reader requires h5py => v2.5.0 or later. |
---|
42 | """ |
---|
43 | |
---|
44 | # CanSAS version |
---|
45 | cansas_version = 2.0 |
---|
46 | # Data type name |
---|
47 | type_name = "NXcanSAS" |
---|
48 | # Wildcards |
---|
49 | type = ["NXcanSAS HDF5 Files (*.h5)|*.h5|"] |
---|
50 | # List of allowed extensions |
---|
51 | ext = ['.h5', '.H5'] |
---|
52 | # Flag to bypass extension check |
---|
53 | allow_all = True |
---|
54 | |
---|
55 | def get_file_contents(self): |
---|
56 | """ |
---|
57 | This is the general read method that all SasView data_loaders must have. |
---|
58 | |
---|
59 | :param filename: A path for an HDF5 formatted CanSAS 2D data file. |
---|
60 | :return: List of Data1D/2D objects and/or a list of errors. |
---|
61 | """ |
---|
62 | # Reinitialize when loading a new data file to reset all class variables |
---|
63 | self.reset_state() |
---|
64 | |
---|
65 | filename = self.f_open.name |
---|
66 | self.f_open.close() # IO handled by h5py |
---|
67 | |
---|
68 | # Check that the file exists |
---|
69 | if os.path.isfile(filename): |
---|
70 | basename = os.path.basename(filename) |
---|
71 | _, extension = os.path.splitext(basename) |
---|
72 | # If the file type is not allowed, return empty list |
---|
73 | if extension in self.ext or self.allow_all: |
---|
74 | # Load the data file |
---|
75 | try: |
---|
76 | self.raw_data = h5py.File(filename, 'r') |
---|
77 | except Exception as e: |
---|
78 | if extension not in self.ext: |
---|
79 | msg = "NXcanSAS Reader could not load file {}".format( |
---|
80 | basename + extension) |
---|
81 | raise DefaultReaderException(msg) |
---|
82 | raise FileContentsException(e.message) |
---|
83 | try: |
---|
84 | # Read in all child elements of top level SASroot |
---|
85 | self.read_children(self.raw_data, []) |
---|
86 | # Add the last data set to the list of outputs |
---|
87 | self.add_data_set() |
---|
88 | except Exception as exc: |
---|
89 | raise FileContentsException(exc.message) |
---|
90 | finally: |
---|
91 | # Close the data file |
---|
92 | self.raw_data.close() |
---|
93 | |
---|
94 | for data_set in self.output: |
---|
95 | if isinstance(data_set, Data1D): |
---|
96 | if data_set.x.size < 5: |
---|
97 | exception = FileContentsException( |
---|
98 | "Fewer than 5 data points found.") |
---|
99 | data_set.errors.append(exception) |
---|
100 | |
---|
101 | def reset_state(self): |
---|
102 | """ |
---|
103 | Create the reader object and define initial states for class variables |
---|
104 | """ |
---|
105 | super(Reader, self).reset_state() |
---|
106 | self.data1d = [] |
---|
107 | self.data2d = [] |
---|
108 | self.raw_data = None |
---|
109 | self.multi_frame = False |
---|
110 | self.data_frames = [] |
---|
111 | self.data_uncertainty_frames = [] |
---|
112 | self.errors = [] |
---|
113 | self.logging = [] |
---|
114 | self.q_names = [] |
---|
115 | self.mask_name = u'' |
---|
116 | self.i_name = u'' |
---|
117 | self.i_node = u'' |
---|
118 | self.i_uncertainties_name = u'' |
---|
119 | self.q_uncertainty_names = [] |
---|
120 | self.q_resolution_names = [] |
---|
121 | self.parent_class = u'' |
---|
122 | self.detector = Detector() |
---|
123 | self.collimation = Collimation() |
---|
124 | self.aperture = Aperture() |
---|
125 | self.process = Process() |
---|
126 | self.trans_spectrum = TransmissionSpectrum() |
---|
127 | |
---|
128 | def read_children(self, data, parent_list): |
---|
129 | """ |
---|
130 | A recursive method for stepping through the hierarchical data file. |
---|
131 | |
---|
132 | :param data: h5py Group object of any kind |
---|
133 | :param parent: h5py Group parent name |
---|
134 | """ |
---|
135 | |
---|
136 | # Loop through each element of the parent and process accordingly |
---|
137 | for key in data.keys(): |
---|
138 | # Get all information for the current key |
---|
139 | value = data.get(key) |
---|
140 | class_name = h5attr(value, u'canSAS_class') |
---|
141 | if isinstance(class_name, (list, tuple, np.ndarray)): |
---|
142 | class_name = class_name[0] |
---|
143 | if class_name is None: |
---|
144 | class_name = h5attr(value, u'NX_class') |
---|
145 | if class_name is not None: |
---|
146 | class_prog = re.compile(class_name) |
---|
147 | else: |
---|
148 | class_prog = re.compile(value.name) |
---|
149 | |
---|
150 | if isinstance(value, h5py.Group): |
---|
151 | # Set parent class before recursion |
---|
152 | last_parent_class = self.parent_class |
---|
153 | self.parent_class = class_name |
---|
154 | parent_list.append(key) |
---|
155 | # If a new sasentry, store the current data sets and create |
---|
156 | # a fresh Data1D/2D object |
---|
157 | if class_prog.match(u'SASentry'): |
---|
158 | self.add_data_set(key) |
---|
159 | elif class_prog.match(u'SASdata'): |
---|
160 | self._find_data_attributes(value) |
---|
161 | self._initialize_new_data_set(value) |
---|
162 | # Recursion step to access data within the group |
---|
163 | self.read_children(value, parent_list) |
---|
164 | self.add_intermediate() |
---|
165 | # Reset parent class when returning from recursive method |
---|
166 | self.parent_class = last_parent_class |
---|
167 | parent_list.remove(key) |
---|
168 | |
---|
169 | elif isinstance(value, h5py.Dataset): |
---|
170 | # If this is a dataset, store the data appropriately |
---|
171 | data_set = value.value |
---|
172 | unit = self._get_unit(value) |
---|
173 | |
---|
174 | for data_point in data_set: |
---|
175 | if isinstance(data_point, np.ndarray): |
---|
176 | if data_point.dtype.char == 'S': |
---|
177 | data_point = decode(bytes(data_point)) |
---|
178 | else: |
---|
179 | data_point = decode(data_point) |
---|
180 | # Top Level Meta Data |
---|
181 | if key == u'definition': |
---|
182 | if isinstance(data_set, basestring): |
---|
183 | self.current_datainfo.meta_data['reader'] = data_set |
---|
184 | break |
---|
185 | else: |
---|
186 | self.current_datainfo.meta_data[ |
---|
187 | 'reader'] = data_point |
---|
188 | # Run |
---|
189 | elif key == u'run': |
---|
190 | try: |
---|
191 | run_name = h5attr(value, 'name') |
---|
192 | run_dict = {data_set: run_name} |
---|
193 | self.current_datainfo.run_name = run_dict |
---|
194 | except Exception: |
---|
195 | pass |
---|
196 | if isinstance(data_set, basestring): |
---|
197 | self.current_datainfo.run.append(data_set) |
---|
198 | break |
---|
199 | else: |
---|
200 | self.current_datainfo.run.append(data_point) |
---|
201 | # Title |
---|
202 | elif key == u'title': |
---|
203 | if isinstance(data_set, basestring): |
---|
204 | self.current_datainfo.title = data_set |
---|
205 | break |
---|
206 | else: |
---|
207 | self.current_datainfo.title = data_point |
---|
208 | # Note |
---|
209 | elif key == u'SASnote': |
---|
210 | self.current_datainfo.notes.append(data_set) |
---|
211 | break |
---|
212 | # Sample Information |
---|
213 | elif self.parent_class == u'SASsample': |
---|
214 | self.process_sample(data_point, key) |
---|
215 | # Instrumental Information |
---|
216 | elif (key == u'name' |
---|
217 | and self.parent_class == u'SASinstrument'): |
---|
218 | self.current_datainfo.instrument = data_point |
---|
219 | # Detector |
---|
220 | elif self.parent_class == u'SASdetector': |
---|
221 | self.process_detector(data_point, key, unit) |
---|
222 | # Collimation |
---|
223 | elif self.parent_class == u'SAScollimation': |
---|
224 | self.process_collimation(data_point, key, unit) |
---|
225 | # Aperture |
---|
226 | elif self.parent_class == u'SASaperture': |
---|
227 | self.process_aperture(data_point, key) |
---|
228 | # Process Information |
---|
229 | elif self.parent_class == u'SASprocess': # CanSAS 2.0 |
---|
230 | self.process_process(data_point, key) |
---|
231 | # Source |
---|
232 | elif self.parent_class == u'SASsource': |
---|
233 | self.process_source(data_point, key, unit) |
---|
234 | # Everything else goes in meta_data |
---|
235 | elif self.parent_class == u'SASdata': |
---|
236 | if isinstance(self.current_dataset, plottable_2D): |
---|
237 | self.process_2d_data_object(data_set, key, unit) |
---|
238 | else: |
---|
239 | self.process_1d_data_object(data_set, key, unit) |
---|
240 | |
---|
241 | break |
---|
242 | elif self.parent_class == u'SAStransmission_spectrum': |
---|
243 | self.process_trans_spectrum(data_set, key) |
---|
244 | break |
---|
245 | else: |
---|
246 | new_key = self._create_unique_key( |
---|
247 | self.current_datainfo.meta_data, key) |
---|
248 | self.current_datainfo.meta_data[new_key] = data_point |
---|
249 | |
---|
250 | else: |
---|
251 | # I don't know if this reachable code |
---|
252 | self.errors.append("ShouldNeverHappenException") |
---|
253 | |
---|
254 | def process_1d_data_object(self, data_set, key, unit): |
---|
255 | """ |
---|
256 | SASdata processor method for 1d data items |
---|
257 | :param data_set: data from HDF5 file |
---|
258 | :param key: canSAS_class attribute |
---|
259 | :param unit: unit attribute |
---|
260 | """ |
---|
261 | if key == self.i_name: |
---|
262 | if self.multi_frame: |
---|
263 | for x in range(0, data_set.shape[0]): |
---|
264 | self.data_frames.append(data_set[x].flatten()) |
---|
265 | else: |
---|
266 | self.current_dataset.y = data_set.flatten() |
---|
267 | self.current_dataset.yaxis("Intensity", unit) |
---|
268 | elif key == self.i_uncertainties_name: |
---|
269 | if self.multi_frame: |
---|
270 | for x in range(0, data_set.shape[0]): |
---|
271 | self.data_uncertainty_frames.append(data_set[x].flatten()) |
---|
272 | self.current_dataset.dy = data_set.flatten() |
---|
273 | elif key in self.q_names: |
---|
274 | self.current_dataset.xaxis("Q", unit) |
---|
275 | self.current_dataset.x = data_set.flatten() |
---|
276 | elif key in self.q_resolution_names: |
---|
277 | if (len(self.q_resolution_names) > 1 |
---|
278 | and np.where(self.q_resolution_names == key)[0] == 0): |
---|
279 | self.current_dataset.dxw = data_set.flatten() |
---|
280 | elif (len(self.q_resolution_names) > 1 |
---|
281 | and np.where(self.q_resolution_names == key)[0] == 1): |
---|
282 | self.current_dataset.dxl = data_set.flatten() |
---|
283 | else: |
---|
284 | self.current_dataset.dx = data_set.flatten() |
---|
285 | elif key in self.q_uncertainty_names: |
---|
286 | if (len(self.q_uncertainty_names) > 1 |
---|
287 | and np.where(self.q_uncertainty_names == key)[0] == 0): |
---|
288 | self.current_dataset.dxw = data_set.flatten() |
---|
289 | elif (len(self.q_uncertainty_names) > 1 |
---|
290 | and np.where(self.q_uncertainty_names == key)[0] == 1): |
---|
291 | self.current_dataset.dxl = data_set.flatten() |
---|
292 | else: |
---|
293 | self.current_dataset.dx = data_set.flatten() |
---|
294 | elif key == self.mask_name: |
---|
295 | self.current_dataset.mask = data_set.flatten() |
---|
296 | elif key == u'wavelength': |
---|
297 | self.current_datainfo.source.wavelength = data_set[0] |
---|
298 | self.current_datainfo.source.wavelength_unit = unit |
---|
299 | |
---|
300 | def process_2d_data_object(self, data_set, key, unit): |
---|
301 | if key == self.i_name: |
---|
302 | self.current_dataset.data = data_set |
---|
303 | self.current_dataset.zaxis("Intensity", unit) |
---|
304 | elif key == self.i_uncertainties_name: |
---|
305 | self.current_dataset.err_data = data_set.flatten() |
---|
306 | elif key in self.q_names: |
---|
307 | self.current_dataset.xaxis("Q_x", unit) |
---|
308 | self.current_dataset.yaxis("Q_y", unit) |
---|
309 | if self.q_names[0] == self.q_names[1]: |
---|
310 | # All q data in a single array |
---|
311 | self.current_dataset.qx_data = data_set[0] |
---|
312 | self.current_dataset.qy_data = data_set[1] |
---|
313 | elif self.q_names.index(key) == 0: |
---|
314 | self.current_dataset.qx_data = data_set |
---|
315 | elif self.q_names.index(key) == 1: |
---|
316 | self.current_dataset.qy_data = data_set |
---|
317 | elif key in self.q_uncertainty_names or key in self.q_resolution_names: |
---|
318 | if ((self.q_uncertainty_names[0] == self.q_uncertainty_names[1]) or |
---|
319 | (self.q_resolution_names[0] == self.q_resolution_names[1])): |
---|
320 | # All q data in a single array |
---|
321 | self.current_dataset.dqx_data = data_set[0].flatten() |
---|
322 | self.current_dataset.dqy_data = data_set[1].flatten() |
---|
323 | elif (self.q_uncertainty_names.index(key) == 0 or |
---|
324 | self.q_resolution_names.index(key) == 0): |
---|
325 | self.current_dataset.dqx_data = data_set.flatten() |
---|
326 | elif (self.q_uncertainty_names.index(key) == 1 or |
---|
327 | self.q_resolution_names.index(key) == 1): |
---|
328 | self.current_dataset.dqy_data = data_set.flatten() |
---|
329 | self.current_dataset.yaxis("Q_y", unit) |
---|
330 | elif key == self.mask_name: |
---|
331 | self.current_dataset.mask = data_set.flatten() |
---|
332 | elif key == u'Qy': |
---|
333 | self.current_dataset.yaxis("Q_y", unit) |
---|
334 | self.current_dataset.qy_data = data_set.flatten() |
---|
335 | elif key == u'Qydev': |
---|
336 | self.current_dataset.dqy_data = data_set.flatten() |
---|
337 | elif key == u'Qx': |
---|
338 | self.current_dataset.xaxis("Q_x", unit) |
---|
339 | self.current_dataset.qx_data = data_set.flatten() |
---|
340 | elif key == u'Qxdev': |
---|
341 | self.current_dataset.dqx_data = data_set.flatten() |
---|
342 | |
---|
343 | def process_trans_spectrum(self, data_set, key): |
---|
344 | """ |
---|
345 | SAStransmission_spectrum processor |
---|
346 | :param data_set: data from HDF5 file |
---|
347 | :param key: canSAS_class attribute |
---|
348 | """ |
---|
349 | if key == u'T': |
---|
350 | self.trans_spectrum.transmission = data_set.flatten() |
---|
351 | elif key == u'Tdev': |
---|
352 | self.trans_spectrum.transmission_deviation = data_set.flatten() |
---|
353 | elif key == u'lambda': |
---|
354 | self.trans_spectrum.wavelength = data_set.flatten() |
---|
355 | |
---|
356 | def process_sample(self, data_point, key): |
---|
357 | """ |
---|
358 | SASsample processor |
---|
359 | :param data_point: Single point from an HDF5 data file |
---|
360 | :param key: class name data_point was taken from |
---|
361 | """ |
---|
362 | if key == u'Title': |
---|
363 | self.current_datainfo.sample.name = data_point |
---|
364 | elif key == u'name': |
---|
365 | self.current_datainfo.sample.name = data_point |
---|
366 | elif key == u'ID': |
---|
367 | self.current_datainfo.sample.name = data_point |
---|
368 | elif key == u'thickness': |
---|
369 | self.current_datainfo.sample.thickness = data_point |
---|
370 | elif key == u'temperature': |
---|
371 | self.current_datainfo.sample.temperature = data_point |
---|
372 | elif key == u'transmission': |
---|
373 | self.current_datainfo.sample.transmission = data_point |
---|
374 | elif key == u'x_position': |
---|
375 | self.current_datainfo.sample.position.x = data_point |
---|
376 | elif key == u'y_position': |
---|
377 | self.current_datainfo.sample.position.y = data_point |
---|
378 | elif key == u'pitch': |
---|
379 | self.current_datainfo.sample.orientation.x = data_point |
---|
380 | elif key == u'yaw': |
---|
381 | self.current_datainfo.sample.orientation.y = data_point |
---|
382 | elif key == u'roll': |
---|
383 | self.current_datainfo.sample.orientation.z = data_point |
---|
384 | elif key == u'details': |
---|
385 | self.current_datainfo.sample.details.append(data_point) |
---|
386 | |
---|
387 | def process_detector(self, data_point, key, unit): |
---|
388 | """ |
---|
389 | SASdetector processor |
---|
390 | :param data_point: Single point from an HDF5 data file |
---|
391 | :param key: class name data_point was taken from |
---|
392 | :param unit: unit attribute from data set |
---|
393 | """ |
---|
394 | if key == u'name': |
---|
395 | self.detector.name = data_point |
---|
396 | elif key == u'SDD': |
---|
397 | self.detector.distance = float(data_point) |
---|
398 | self.detector.distance_unit = unit |
---|
399 | elif key == u'slit_length': |
---|
400 | self.detector.slit_length = float(data_point) |
---|
401 | self.detector.slit_length_unit = unit |
---|
402 | elif key == u'x_position': |
---|
403 | self.detector.offset.x = float(data_point) |
---|
404 | self.detector.offset_unit = unit |
---|
405 | elif key == u'y_position': |
---|
406 | self.detector.offset.y = float(data_point) |
---|
407 | self.detector.offset_unit = unit |
---|
408 | elif key == u'pitch': |
---|
409 | self.detector.orientation.x = float(data_point) |
---|
410 | self.detector.orientation_unit = unit |
---|
411 | elif key == u'roll': |
---|
412 | self.detector.orientation.z = float(data_point) |
---|
413 | self.detector.orientation_unit = unit |
---|
414 | elif key == u'yaw': |
---|
415 | self.detector.orientation.y = float(data_point) |
---|
416 | self.detector.orientation_unit = unit |
---|
417 | elif key == u'beam_center_x': |
---|
418 | self.detector.beam_center.x = float(data_point) |
---|
419 | self.detector.beam_center_unit = unit |
---|
420 | elif key == u'beam_center_y': |
---|
421 | self.detector.beam_center.y = float(data_point) |
---|
422 | self.detector.beam_center_unit = unit |
---|
423 | elif key == u'x_pixel_size': |
---|
424 | self.detector.pixel_size.x = float(data_point) |
---|
425 | self.detector.pixel_size_unit = unit |
---|
426 | elif key == u'y_pixel_size': |
---|
427 | self.detector.pixel_size.y = float(data_point) |
---|
428 | self.detector.pixel_size_unit = unit |
---|
429 | |
---|
430 | def process_collimation(self, data_point, key, unit): |
---|
431 | """ |
---|
432 | SAScollimation processor |
---|
433 | :param data_point: Single point from an HDF5 data file |
---|
434 | :param key: class name data_point was taken from |
---|
435 | :param unit: unit attribute from data set |
---|
436 | """ |
---|
437 | if key == u'distance': |
---|
438 | self.collimation.length = data_point |
---|
439 | self.collimation.length_unit = unit |
---|
440 | elif key == u'name': |
---|
441 | self.collimation.name = data_point |
---|
442 | |
---|
443 | def process_aperture(self, data_point, key): |
---|
444 | """ |
---|
445 | SASaperture processor |
---|
446 | :param data_point: Single point from an HDF5 data file |
---|
447 | :param key: class name data_point was taken from |
---|
448 | """ |
---|
449 | if key == u'shape': |
---|
450 | self.aperture.shape = data_point |
---|
451 | elif key == u'x_gap': |
---|
452 | self.aperture.size.x = data_point |
---|
453 | elif key == u'y_gap': |
---|
454 | self.aperture.size.y = data_point |
---|
455 | |
---|
456 | def process_source(self, data_point, key, unit): |
---|
457 | """ |
---|
458 | SASsource processor |
---|
459 | :param data_point: Single point from an HDF5 data file |
---|
460 | :param key: class name data_point was taken from |
---|
461 | :param unit: unit attribute from data set |
---|
462 | """ |
---|
463 | if key == u'incident_wavelength': |
---|
464 | self.current_datainfo.source.wavelength = data_point |
---|
465 | self.current_datainfo.source.wavelength_unit = unit |
---|
466 | elif key == u'wavelength_max': |
---|
467 | self.current_datainfo.source.wavelength_max = data_point |
---|
468 | self.current_datainfo.source.wavelength_max_unit = unit |
---|
469 | elif key == u'wavelength_min': |
---|
470 | self.current_datainfo.source.wavelength_min = data_point |
---|
471 | self.current_datainfo.source.wavelength_min_unit = unit |
---|
472 | elif key == u'incident_wavelength_spread': |
---|
473 | self.current_datainfo.source.wavelength_spread = data_point |
---|
474 | self.current_datainfo.source.wavelength_spread_unit = unit |
---|
475 | elif key == u'beam_size_x': |
---|
476 | self.current_datainfo.source.beam_size.x = data_point |
---|
477 | self.current_datainfo.source.beam_size_unit = unit |
---|
478 | elif key == u'beam_size_y': |
---|
479 | self.current_datainfo.source.beam_size.y = data_point |
---|
480 | self.current_datainfo.source.beam_size_unit = unit |
---|
481 | elif key == u'beam_shape': |
---|
482 | self.current_datainfo.source.beam_shape = data_point |
---|
483 | elif key == u'radiation': |
---|
484 | self.current_datainfo.source.radiation = data_point |
---|
485 | |
---|
486 | def process_process(self, data_point, key): |
---|
487 | """ |
---|
488 | SASprocess processor |
---|
489 | :param data_point: Single point from an HDF5 data file |
---|
490 | :param key: class name data_point was taken from |
---|
491 | """ |
---|
492 | term_match = re.compile(u'^term[0-9]+$') |
---|
493 | if key == u'Title': # CanSAS 2.0 |
---|
494 | self.process.name = data_point |
---|
495 | elif key == u'name': # NXcanSAS |
---|
496 | self.process.name = data_point |
---|
497 | elif key == u'description': |
---|
498 | self.process.description = data_point |
---|
499 | elif key == u'date': |
---|
500 | self.process.date = data_point |
---|
501 | elif term_match.match(key): |
---|
502 | self.process.term.append(data_point) |
---|
503 | else: |
---|
504 | self.process.notes.append(data_point) |
---|
505 | |
---|
506 | def add_intermediate(self): |
---|
507 | """ |
---|
508 | This method stores any intermediate objects within the final data set |
---|
509 | after fully reading the set. |
---|
510 | |
---|
511 | :param parent: The NXclass name for the h5py Group object that just |
---|
512 | finished being processed |
---|
513 | """ |
---|
514 | |
---|
515 | if self.parent_class == u'SASprocess': |
---|
516 | self.current_datainfo.process.append(self.process) |
---|
517 | self.process = Process() |
---|
518 | elif self.parent_class == u'SASdetector': |
---|
519 | self.current_datainfo.detector.append(self.detector) |
---|
520 | self.detector = Detector() |
---|
521 | elif self.parent_class == u'SAStransmission_spectrum': |
---|
522 | self.current_datainfo.trans_spectrum.append(self.trans_spectrum) |
---|
523 | self.trans_spectrum = TransmissionSpectrum() |
---|
524 | elif self.parent_class == u'SAScollimation': |
---|
525 | self.current_datainfo.collimation.append(self.collimation) |
---|
526 | self.collimation = Collimation() |
---|
527 | elif self.parent_class == u'SASaperture': |
---|
528 | self.collimation.aperture.append(self.aperture) |
---|
529 | self.aperture = Aperture() |
---|
530 | elif self.parent_class == u'SASdata': |
---|
531 | if isinstance(self.current_dataset, plottable_2D): |
---|
532 | self.data2d.append(self.current_dataset) |
---|
533 | elif isinstance(self.current_dataset, plottable_1D): |
---|
534 | if self.multi_frame: |
---|
535 | for x in range(0, len(self.data_frames)): |
---|
536 | self.current_dataset.y = self.data_frames[x] |
---|
537 | if len(self.data_uncertainty_frames) > x: |
---|
538 | self.current_dataset.dy = \ |
---|
539 | self.data_uncertainty_frames[x] |
---|
540 | self.data1d.append(self.current_dataset) |
---|
541 | else: |
---|
542 | self.data1d.append(self.current_dataset) |
---|
543 | |
---|
544 | def final_data_cleanup(self): |
---|
545 | """ |
---|
546 | Does some final cleanup and formatting on self.current_datainfo and |
---|
547 | all data1D and data2D objects and then combines the data and info into |
---|
548 | Data1D and Data2D objects |
---|
549 | """ |
---|
550 | # Type cast data arrays to float64 |
---|
551 | if len(self.current_datainfo.trans_spectrum) > 0: |
---|
552 | spectrum_list = [] |
---|
553 | for spectrum in self.current_datainfo.trans_spectrum: |
---|
554 | spectrum.transmission = spectrum.transmission.astype(np.float64) |
---|
555 | spectrum.transmission_deviation = \ |
---|
556 | spectrum.transmission_deviation.astype(np.float64) |
---|
557 | spectrum.wavelength = spectrum.wavelength.astype(np.float64) |
---|
558 | if len(spectrum.transmission) > 0: |
---|
559 | spectrum_list.append(spectrum) |
---|
560 | self.current_datainfo.trans_spectrum = spectrum_list |
---|
561 | |
---|
562 | # Append errors to dataset and reset class errors |
---|
563 | self.current_datainfo.errors = self.errors |
---|
564 | self.errors = [] |
---|
565 | |
---|
566 | # Combine all plottables with datainfo and append each to output |
---|
567 | # Type cast data arrays to float64 and find min/max as appropriate |
---|
568 | for dataset in self.data2d: |
---|
569 | # Calculate the actual Q matrix |
---|
570 | try: |
---|
571 | if dataset.q_data.size <= 1: |
---|
572 | dataset.q_data = np.sqrt(dataset.qx_data |
---|
573 | * dataset.qx_data |
---|
574 | + dataset.qy_data |
---|
575 | * dataset.qy_data).flatten() |
---|
576 | except: |
---|
577 | dataset.q_data = None |
---|
578 | |
---|
579 | if dataset.data.ndim == 2: |
---|
580 | dataset.y_bins = np.unique(dataset.qy_data.flatten()) |
---|
581 | dataset.x_bins = np.unique(dataset.qx_data.flatten()) |
---|
582 | dataset.data = dataset.data.flatten() |
---|
583 | dataset.qx_data = dataset.qx_data.flatten() |
---|
584 | dataset.qy_data = dataset.qy_data.flatten() |
---|
585 | |
---|
586 | try: |
---|
587 | iter(dataset.mask) |
---|
588 | dataset.mask = np.invert(np.asarray(dataset.mask, dtype=bool)) |
---|
589 | except TypeError: |
---|
590 | dataset.mask = np.ones(dataset.data.shape, dtype=bool) |
---|
591 | self.current_dataset = dataset |
---|
592 | self.send_to_output() |
---|
593 | |
---|
594 | for dataset in self.data1d: |
---|
595 | self.current_dataset = dataset |
---|
596 | self.send_to_output() |
---|
597 | |
---|
598 | def add_data_set(self, key=""): |
---|
599 | """ |
---|
600 | Adds the current_dataset to the list of outputs after preforming final |
---|
601 | processing on the data and then calls a private method to generate a |
---|
602 | new data set. |
---|
603 | |
---|
604 | :param key: NeXus group name for current tree level |
---|
605 | """ |
---|
606 | |
---|
607 | if self.current_datainfo and self.current_dataset: |
---|
608 | self.final_data_cleanup() |
---|
609 | self.data_frames = [] |
---|
610 | self.data_uncertainty_frames = [] |
---|
611 | self.data1d = [] |
---|
612 | self.data2d = [] |
---|
613 | self.current_datainfo = DataInfo() |
---|
614 | |
---|
615 | def _initialize_new_data_set(self, value=None): |
---|
616 | """ |
---|
617 | A private class method to generate a new 1D or 2D data object based on |
---|
618 | the type of data within the set. Outside methods should call |
---|
619 | add_data_set() to be sure any existing data is stored properly. |
---|
620 | |
---|
621 | :param parent_list: List of names of parent elements |
---|
622 | """ |
---|
623 | if self._is_2d_not_multi_frame(value): |
---|
624 | self.current_dataset = plottable_2D() |
---|
625 | else: |
---|
626 | x = np.array(0) |
---|
627 | y = np.array(0) |
---|
628 | self.current_dataset = plottable_1D(x, y) |
---|
629 | self.current_datainfo.filename = self.raw_data.filename |
---|
630 | |
---|
631 | @staticmethod |
---|
632 | def as_list_or_array(iterable): |
---|
633 | """ |
---|
634 | Return value as a list if not already a list or array. |
---|
635 | :param iterable: |
---|
636 | :return: |
---|
637 | """ |
---|
638 | if not (isinstance(iterable, np.ndarray) or isinstance(iterable, list)): |
---|
639 | iterable = iterable.split(",") if isinstance(iterable, basestring)\ |
---|
640 | else [iterable] |
---|
641 | return iterable |
---|
642 | |
---|
643 | def _find_data_attributes(self, value): |
---|
644 | """ |
---|
645 | A class to find the indices for Q, the name of the Qdev and Idev, and |
---|
646 | the name of the mask. |
---|
647 | :param value: SASdata/NXdata HDF5 Group |
---|
648 | """ |
---|
649 | # Initialize values to base types |
---|
650 | self.mask_name = u'' |
---|
651 | self.i_name = u'' |
---|
652 | self.i_node = u'' |
---|
653 | self.i_uncertainties_name = u'' |
---|
654 | self.q_names = [] |
---|
655 | self.q_uncertainty_names = [] |
---|
656 | self.q_resolution_names = [] |
---|
657 | # Get attributes |
---|
658 | attrs = value.attrs |
---|
659 | signal = attrs.get("signal", "I") |
---|
660 | if isinstance(signal, bytes): |
---|
661 | signal = signal.decode() |
---|
662 | i_axes = attrs.get("I_axes", ["Q"]) |
---|
663 | if isinstance(i_axes, bytes): |
---|
664 | i_axes = i_axes.decode() |
---|
665 | q_indices = attrs.get("Q_indices", [0]) |
---|
666 | if isinstance(q_indices, bytes): |
---|
667 | q_indices = q_indices.decode() |
---|
668 | i_axes = self.as_list_or_array(i_axes) |
---|
669 | keys = value.keys() |
---|
670 | # Assign attributes to appropriate class variables |
---|
671 | self.q_names = [i_axes[int(v)] for v in self.as_list_or_array(q_indices)] |
---|
672 | self.mask_name = attrs.get("mask") |
---|
673 | if isinstance(self.mask_name, bytes): |
---|
674 | self.mask_name = self.mask_name.decode() |
---|
675 | self.i_name = signal |
---|
676 | self.i_node = value.get(self.i_name) |
---|
677 | |
---|
678 | for item in self.q_names: |
---|
679 | if item in keys: |
---|
680 | q_vals = value.get(item) |
---|
681 | if q_vals.attrs.get("uncertainties") is not None: |
---|
682 | self.q_uncertainty_names = q_vals.attrs.get("uncertainties") |
---|
683 | elif q_vals.attrs.get("uncertainty") is not None: |
---|
684 | self.q_uncertainty_names = q_vals.attrs.get("uncertainty") |
---|
685 | if isinstance(self.q_uncertainty_names, basestring): |
---|
686 | self.q_uncertainty_names = self.q_uncertainty_names.split(",") |
---|
687 | if q_vals.attrs.get("resolutions") is not None: |
---|
688 | self.q_resolution_names = q_vals.attrs.get("resolutions") |
---|
689 | if isinstance(self.q_resolution_names, basestring): |
---|
690 | self.q_resolution_names = self.q_resolution_names.split(",") |
---|
691 | if self.i_name in keys: |
---|
692 | i_vals = value.get(self.i_name) |
---|
693 | self.i_uncertainties_name = i_vals.attrs.get("uncertainties") |
---|
694 | if self.i_uncertainties_name is None: |
---|
695 | self.i_uncertainties_name = i_vals.attrs.get("uncertainty") |
---|
696 | if isinstance(self.i_uncertainties_name, bytes): |
---|
697 | key = self.i_uncertainties_name |
---|
698 | self.i_uncertainties_name = key.decode() if isinstance(key, bytes) else key |
---|
699 | |
---|
700 | def _is_2d_not_multi_frame(self, value, i_base="", q_base=""): |
---|
701 | """ |
---|
702 | A private class to determine if the data set is 1d or 2d. |
---|
703 | |
---|
704 | :param value: Nexus/NXcanSAS data group |
---|
705 | :param basename: Approximate name of an entry to search for |
---|
706 | :return: True if 2D, otherwise false |
---|
707 | """ |
---|
708 | i_basename = i_base if i_base != "" else self.i_name |
---|
709 | i_vals = value.get(i_basename) |
---|
710 | q_basename = q_base if q_base != "" else self.q_names |
---|
711 | q_vals = value.get(q_basename[0]) |
---|
712 | self.multi_frame = (i_vals is not None and q_vals is not None |
---|
713 | and len(i_vals.shape) != 1 |
---|
714 | and len(q_vals.shape) == 1) |
---|
715 | return (i_vals is not None and len(i_vals.shape) != 1 |
---|
716 | and not self.multi_frame) |
---|
717 | |
---|
718 | def _create_unique_key(self, dictionary, name, numb=0): |
---|
719 | """ |
---|
720 | Create a unique key value for any dictionary to prevent overwriting |
---|
721 | Recurses until a unique key value is found. |
---|
722 | |
---|
723 | :param dictionary: A dictionary with any number of entries |
---|
724 | :param name: The index of the item to be added to dictionary |
---|
725 | :param numb: The number to be appended to the name, starts at 0 |
---|
726 | :return: The new name for the dictionary entry |
---|
727 | """ |
---|
728 | if dictionary.get(name) is not None: |
---|
729 | numb += 1 |
---|
730 | name = name.split("_")[0] |
---|
731 | name += "_{0}".format(numb) |
---|
732 | name = self._create_unique_key(dictionary, name, numb) |
---|
733 | return name |
---|
734 | |
---|
735 | def _get_unit(self, value): |
---|
736 | """ |
---|
737 | Find the unit for a particular value within the h5py dictionary |
---|
738 | |
---|
739 | :param value: attribute dictionary for a particular value set |
---|
740 | :return: unit for the value passed to the method |
---|
741 | """ |
---|
742 | unit = h5attr(value, u'units') |
---|
743 | if unit is None: |
---|
744 | unit = h5attr(value, u'unit') |
---|
745 | return unit |
---|