source: sasview/src/sas/sascalc/dataloader/readers/cansas_reader_HDF5.py @ b76cf0d

ESS_GUIESS_GUI_DocsESS_GUI_batch_fittingESS_GUI_bumps_abstractionESS_GUI_iss1116ESS_GUI_iss879ESS_GUI_iss959ESS_GUI_openclESS_GUI_orderingESS_GUI_sync_sascalc
Last change on this file since b76cf0d was b76cf0d, checked in by Piotr Rozyczko <rozyczko@…>, 8 years ago

Improve efficiency of CanSAS 2.0 reader

When it came to reading the Q and I data values, the reader was looping
through every single data point in the hdf5 file and appending it to an
array. This caused problems when loading very large files, rendering
SasView? unresponsive. Now, the reader saves the entire array from the
file, instead of looping through each point.
  • Property mode set to 100644
File size: 20.2 KB
Line 
1"""
2    CanSAS 2D data reader for reading HDF5 formatted CanSAS files.
3"""
4
5import h5py
6import numpy as np
7import re
8import os
9import sys
10
11from sas.sascalc.dataloader.data_info import plottable_1D, plottable_2D, Data1D, Data2D, DataInfo, Process, Aperture
12from sas.sascalc.dataloader.data_info import Collimation, TransmissionSpectrum, Detector
13from sas.sascalc.dataloader.data_info import combine_data_info_with_plottable
14
15
16
17class Reader():
18    """
19    A class for reading in CanSAS v2.0 data files. The existing iteration opens Mantid generated HDF5 formatted files
20    with file extension .h5/.H5. Any number of data sets may be present within the file and any dimensionality of data
21    may be used. Currently 1D and 2D SAS data sets are supported, but future implementations will include 1D and 2D
22    SESANS data.
23
24    Any number of SASdata sets may be present in a SASentry and the data within can be either 1D I(Q) or 2D I(Qx, Qy).
25
26    :Dependencies:
27        The CanSAS HDF5 reader requires h5py => v2.5.0 or later.
28    """
29
30    ## CanSAS version
31    cansas_version = 2.0
32    ## Logged warnings or messages
33    logging = None
34    ## List of errors for the current data set
35    errors = None
36    ## Raw file contents to be processed
37    raw_data = None
38    ## Data info currently being read in
39    current_datainfo = None
40    ## SASdata set currently being read in
41    current_dataset = None
42    ## List of plottable1D objects that should be linked to the current_datainfo
43    data1d = None
44    ## List of plottable2D objects that should be linked to the current_datainfo
45    data2d = None
46    ## Data type name
47    type_name = "CanSAS 2.0"
48    ## Wildcards
49    type = ["CanSAS 2.0 HDF5 Files (*.h5)|*.h5"]
50    ## List of allowed extensions
51    ext = ['.h5', '.H5']
52    ## Flag to bypass extension check
53    allow_all = False
54    ## List of files to return
55    output = None
56
57    def read(self, filename):
58        """
59        This is the general read method that all SasView data_loaders must have.
60
61        :param filename: A path for an HDF5 formatted CanSAS 2D data file.
62        :return: List of Data1D/2D objects and/or a list of errors.
63        """
64        ## Reinitialize the class when loading a new data file to reset all class variables
65        self.reset_class_variables()
66        ## Check that the file exists
67        if os.path.isfile(filename):
68            basename = os.path.basename(filename)
69            _, extension = os.path.splitext(basename)
70            # If the file type is not allowed, return empty list
71            if extension in self.ext or self.allow_all:
72                ## Load the data file
73                self.raw_data = h5py.File(filename, 'r')
74                ## Read in all child elements of top level SASroot
75                self.read_children(self.raw_data, [])
76                ## Add the last data set to the list of outputs
77                self.add_data_set()
78        ## Return data set(s)
79        return self.output
80
81    def reset_class_variables(self):
82        """
83        Create the reader object and define initial states for class variables
84        """
85        self.current_datainfo = None
86        self.current_dataset = None
87        self.data1d = []
88        self.data2d = []
89        self.raw_data = None
90        self.errors = set()
91        self.logging = []
92        self.output = []
93        self.parent_class = u''
94        self.detector = Detector()
95        self.collimation = Collimation()
96        self.aperture = Aperture()
97        self.process = Process()
98        self.trans_spectrum = TransmissionSpectrum()
99
100    def read_children(self, data, parent_list):
101        """
102        A recursive method for stepping through the hierarchical data file.
103
104        :param data: h5py Group object of any kind
105        :param parent: h5py Group parent name
106        """
107
108        ## Loop through each element of the parent and process accordingly
109        for key in data.keys():
110            ## Get all information for the current key
111            value = data.get(key)
112            if value.attrs.get(u'canSAS_class') is not None:
113                class_name = value.attrs.get(u'canSAS_class')
114            else:
115                class_name = value.attrs.get(u'NX_class')
116            if class_name is not None:
117                class_prog = re.compile(class_name)
118            else:
119                class_prog = re.compile(value.name)
120
121            if isinstance(value, h5py.Group):
122                self.parent_class = class_name
123                parent_list.append(key)
124                ## If this is a new sasentry, store the current data sets and create a fresh Data1D/2D object
125                if class_prog.match(u'SASentry'):
126                    self.add_data_set(key)
127                elif class_prog.match(u'SASdata'):
128                    self._initialize_new_data_set(parent_list)
129                ## Recursion step to access data within the group
130                self.read_children(value, parent_list)
131                self.add_intermediate()
132                parent_list.remove(key)
133
134            elif isinstance(value, h5py.Dataset):
135                ## If this is a dataset, store the data appropriately
136                data_set = data[key][:]
137                unit = self._get_unit(value)
138
139                ## I and Q Data
140                if key == u'I':
141                    if type(self.current_dataset) is plottable_2D:
142                        self.current_dataset.data = data_set.flatten()
143                        self.current_dataset.zaxis("Intensity", unit)
144                    else:
145                        self.current_dataset.y = data_set.flatten()
146                        self.current_dataset.yaxis("Intensity", unit)
147                    continue
148                elif key == u'Idev':
149                    if type(self.current_dataset) is plottable_2D:
150                        self.current_dataset.err_data = data_set.flatten()
151                    else:
152                        self.current_dataset.dy = data_set.flatten()
153                    continue
154                elif key == u'Q':
155                    self.current_dataset.xaxis("Q", unit)
156                    if type(self.current_dataset) is plottable_2D:
157                        self.current_dataset.q = data_set.flatten()
158                    else:
159                        self.current_dataset.x = data_set.flatten()
160                    continue
161                elif key == u'Qy':
162                    self.current_dataset.yaxis("Q_y", unit)
163                    self.current_dataset.qy_data = data_set.flatten()
164                    continue
165                elif key == u'Qydev':
166                    self.current_dataset.dqy_data = data_set.flatten()
167                    continue
168                elif key == u'Qx':
169                    self.current_dataset.xaxis("Q_x", unit)
170                    self.current_dataset.qx_data = data_set.flatten()
171                    continue
172                elif key == u'Qxdev':
173                    self.current_dataset.dqx_data = data_set.flatten()
174                    continue
175                elif key == u'Mask':
176                    self.current_dataset.mask = data_set.flatten()
177                    continue
178
179                for data_point in data_set:
180                    ## Top Level Meta Data
181                    if key == u'definition':
182                        self.current_datainfo.meta_data['reader'] = data_point
183                    elif key == u'run':
184                        self.current_datainfo.run.append(data_point)
185                    elif key == u'title':
186                        self.current_datainfo.title = data_point
187                    elif key == u'SASnote':
188                        self.current_datainfo.notes.append(data_point)
189
190                    ## Sample Information
191                    elif key == u'Title' and self.parent_class == u'SASsample':
192                        self.current_datainfo.sample.name = data_point
193                    elif key == u'thickness' and self.parent_class == u'SASsample':
194                        self.current_datainfo.sample.thickness = data_point
195                    elif key == u'temperature' and self.parent_class == u'SASsample':
196                        self.current_datainfo.sample.temperature = data_point
197
198                    ## Instrumental Information
199                    elif key == u'name' and self.parent_class == u'SASinstrument':
200                        self.current_datainfo.instrument = data_point
201                    elif key == u'name' and self.parent_class == u'SASdetector':
202                        self.detector.name = data_point
203                    elif key == u'SDD' and self.parent_class == u'SASdetector':
204                        self.detector.distance = float(data_point)
205                        self.detector.distance_unit = unit
206                    elif key == u'SSD' and self.parent_class == u'SAScollimation':
207                        self.collimation.length = data_point
208                        self.collimation.length_unit = unit
209                    elif key == u'name' and self.parent_class == u'SAScollimation':
210                        self.collimation.name = data_point
211
212                    ## Process Information
213                    elif key == u'name' and self.parent_class == u'SASprocess':
214                        self.process.name = data_point
215                    elif key == u'Title' and self.parent_class == u'SASprocess':
216                        self.process.name = data_point
217                    elif key == u'description' and self.parent_class == u'SASprocess':
218                        self.process.description = data_point
219                    elif key == u'date' and self.parent_class == u'SASprocess':
220                        self.process.date = data_point
221                    elif self.parent_class == u'SASprocess':
222                        self.process.notes.append(data_point)
223
224                    ## Transmission Spectrum
225                    elif key == u'T' and self.parent_class == u'SAStransmission_spectrum':
226                        self.trans_spectrum.transmission.append(data_point)
227                    elif key == u'Tdev' and self.parent_class == u'SAStransmission_spectrum':
228                        self.trans_spectrum.transmission_deviation.append(data_point)
229                    elif key == u'lambda' and self.parent_class == u'SAStransmission_spectrum':
230                        self.trans_spectrum.wavelength.append(data_point)
231
232                    ## Other Information
233                    elif key == u'wavelength' and self.parent_class == u'SASdata':
234                        self.current_datainfo.source.wavelength = data_point
235                        self.current_datainfo.source.wavelength.unit = unit
236                    elif key == u'radiation' and self.parent_class == u'SASsource':
237                        self.current_datainfo.source.radiation = data_point
238                    elif key == u'transmission' and self.parent_class == u'SASdata':
239                        self.current_datainfo.sample.transmission = data_point
240
241                    ## Everything else goes in meta_data
242                    else:
243                        new_key = self._create_unique_key(self.current_datainfo.meta_data, key)
244                        self.current_datainfo.meta_data[new_key] = data_point
245
246            else:
247                ## I don't know if this reachable code
248                self.errors.add("ShouldNeverHappenException")
249
250    def add_intermediate(self):
251        """
252        This method stores any intermediate objects within the final data set after fully reading the set.
253
254        :param parent: The NXclass name for the h5py Group object that just finished being processed
255        """
256
257        if self.parent_class == u'SASprocess':
258            self.current_datainfo.process.append(self.process)
259            self.process = Process()
260        elif self.parent_class == u'SASdetector':
261            self.current_datainfo.detector.append(self.detector)
262            self.detector = Detector()
263        elif self.parent_class == u'SAStransmission_spectrum':
264            self.current_datainfo.trans_spectrum.append(self.trans_spectrum)
265            self.trans_spectrum = TransmissionSpectrum()
266        elif self.parent_class == u'SAScollimation':
267            self.current_datainfo.collimation.append(self.collimation)
268            self.collimation = Collimation()
269        elif self.parent_class == u'SASaperture':
270            self.collimation.aperture.append(self.aperture)
271            self.aperture = Aperture()
272        elif self.parent_class == u'SASdata':
273            if type(self.current_dataset) is plottable_2D:
274                self.data2d.append(self.current_dataset)
275            elif type(self.current_dataset) is plottable_1D:
276                self.data1d.append(self.current_dataset)
277
278    def final_data_cleanup(self):
279        """
280        Does some final cleanup and formatting on self.current_datainfo and all data1D and data2D objects and then
281        combines the data and info into Data1D and Data2D objects
282        """
283
284        ## Type cast data arrays to float64
285        if len(self.current_datainfo.trans_spectrum) > 0:
286            spectrum_list = []
287            for spectrum in self.current_datainfo.trans_spectrum:
288                spectrum.transmission = np.delete(spectrum.transmission, [0])
289                spectrum.transmission = spectrum.transmission.astype(np.float64)
290                spectrum.transmission_deviation = np.delete(spectrum.transmission_deviation, [0])
291                spectrum.transmission_deviation = spectrum.transmission_deviation.astype(np.float64)
292                spectrum.wavelength = np.delete(spectrum.wavelength, [0])
293                spectrum.wavelength = spectrum.wavelength.astype(np.float64)
294                if len(spectrum.transmission) > 0:
295                    spectrum_list.append(spectrum)
296            self.current_datainfo.trans_spectrum = spectrum_list
297
298        ## Append errors to dataset and reset class errors
299        self.current_datainfo.errors = self.errors
300        self.errors.clear()
301
302        ## Combine all plottables with datainfo and append each to output
303        ## Type cast data arrays to float64 and find min/max as appropriate
304        for dataset in self.data2d:
305            dataset.data = np.delete(dataset.data, [0])
306            dataset.data = dataset.data.astype(np.float64)
307            dataset.err_data = np.delete(dataset.err_data, [0])
308            dataset.err_data = dataset.err_data.astype(np.float64)
309            dataset.mask = np.delete(dataset.mask, [0])
310            if dataset.qx_data is not None:
311                dataset.qx_data = np.delete(dataset.qx_data, [0])
312                dataset.xmin = np.min(dataset.qx_data)
313                dataset.xmax = np.max(dataset.qx_data)
314                dataset.qx_data = dataset.qx_data.astype(np.float64)
315            if dataset.dqx_data is not None:
316                dataset.dqx_data = np.delete(dataset.dqx_data, [0])
317                dataset.dqx_data = dataset.dqx_data.astype(np.float64)
318            if dataset.qy_data is not None:
319                dataset.qy_data = np.delete(dataset.qy_data, [0])
320                dataset.ymin = np.min(dataset.qy_data)
321                dataset.ymax = np.max(dataset.qy_data)
322                dataset.qy_data = dataset.qy_data.astype(np.float64)
323            if dataset.dqy_data is not None:
324                dataset.dqy_data = np.delete(dataset.dqy_data, [0])
325                dataset.dqy_data = dataset.dqy_data.astype(np.float64)
326            if dataset.q_data is not None:
327                dataset.q_data = np.delete(dataset.q_data, [0])
328                dataset.q_data = dataset.q_data.astype(np.float64)
329            zeros = np.ones(dataset.data.size, dtype=bool)
330            try:
331                for i in range (0, dataset.mask.size - 1):
332                    zeros[i] = dataset.mask[i]
333            except:
334                self.errors.add(sys.exc_value)
335            dataset.mask = zeros
336            ## Calculate the actual Q matrix
337            try:
338                if dataset.q_data.size <= 1:
339                    dataset.q_data = np.sqrt(dataset.qx_data * dataset.qx_data + dataset.qy_data * dataset.qy_data)
340            except:
341                dataset.q_data = None
342            final_dataset = combine_data_info_with_plottable(dataset, self.current_datainfo)
343            self.output.append(final_dataset)
344
345        for dataset in self.data1d:
346            if dataset.x is not None:
347                dataset.x = np.delete(dataset.x, [0])
348                dataset.x = dataset.x.astype(np.float64)
349                dataset.xmin = np.min(dataset.x)
350                dataset.xmax = np.max(dataset.x)
351            if dataset.y is not None:
352                dataset.y = np.delete(dataset.y, [0])
353                dataset.y = dataset.y.astype(np.float64)
354                dataset.ymin = np.min(dataset.y)
355                dataset.ymax = np.max(dataset.y)
356            if dataset.dx is not None:
357                dataset.dx = np.delete(dataset.dx, [0])
358                dataset.dx = dataset.dx.astype(np.float64)
359            if dataset.dxl is not None:
360                dataset.dxl = np.delete(dataset.dxl, [0])
361                dataset.dxl = dataset.dxl.astype(np.float64)
362            if dataset.dxw is not None:
363                dataset.dxw = np.delete(dataset.dxw, [0])
364                dataset.dxw = dataset.dxw.astype(np.float64)
365            if dataset.dy is not None:
366                dataset.dy = np.delete(dataset.dy, [0])
367                dataset.dy = dataset.dy.astype(np.float64)
368            final_dataset = combine_data_info_with_plottable(dataset, self.current_datainfo)
369            self.output.append(final_dataset)
370
371    def add_data_set(self, key=""):
372        """
373        Adds the current_dataset to the list of outputs after preforming final processing on the data and then calls a
374        private method to generate a new data set.
375
376        :param key: NeXus group name for current tree level
377        """
378
379        if self.current_datainfo and self.current_dataset:
380            self.final_data_cleanup()
381        self.data1d = []
382        self.data2d = []
383        self.current_datainfo = DataInfo()
384
385    def _initialize_new_data_set(self, parent_list = None):
386        """
387        A private class method to generate a new 1D or 2D data object based on the type of data within the set.
388        Outside methods should call add_data_set() to be sure any existing data is stored properly.
389
390        :param parent_list: List of names of parent elements
391        """
392
393        if parent_list is None:
394            parent_list = []
395        if self._find_intermediate(parent_list, "Qx"):
396            self.current_dataset = plottable_2D()
397        else:
398            x = np.array(0)
399            y = np.array(0)
400            self.current_dataset = plottable_1D(x, y)
401        self.current_datainfo.filename = self.raw_data.filename
402
403    def _find_intermediate(self, parent_list, basename=""):
404        """
405        A private class used to find an entry by either using a direct key or knowing the approximate basename.
406
407        :param parent_list: List of parents to the current level in the HDF5 file
408        :param basename: Approximate name of an entry to search for
409        :return:
410        """
411
412        entry = False
413        key_prog = re.compile(basename)
414        top = self.raw_data
415        for parent in parent_list:
416            top = top.get(parent)
417        for key in top.keys():
418            if (key_prog.match(key)):
419                entry = True
420                break
421        return entry
422
423    def _create_unique_key(self, dictionary, name, numb=0):
424        """
425        Create a unique key value for any dictionary to prevent overwriting
426        Recurses until a unique key value is found.
427
428        :param dictionary: A dictionary with any number of entries
429        :param name: The index of the item to be added to dictionary
430        :param numb: The number to be appended to the name, starts at 0
431        :return: The new name for the dictionary entry
432        """
433        if dictionary.get(name) is not None:
434            numb += 1
435            name = name.split("_")[0]
436            name += "_{0}".format(numb)
437            name = self._create_unique_key(dictionary, name, numb)
438        return name
439
440    def _get_unit(self, value):
441        """
442        Find the unit for a particular value within the h5py dictionary
443
444        :param value: attribute dictionary for a particular value set
445        :return: unit for the value passed to the method
446        """
447        unit = value.attrs.get(u'units')
448        if unit == None:
449            unit = value.attrs.get(u'unit')
450        ## Convert the unit formats
451        if unit == "1/A":
452            unit = "A^{-1}"
453        elif unit == "1/cm":
454            unit = "cm^{-1}"
455        return unit
Note: See TracBrowser for help on using the repository browser.