ascii_reader.py @ e04f87b

ESS_GUIESS_GUI_DocsESS_GUI_batch_fittingESS_GUI_bumps_abstractionESS_GUI_iss1116ESS_GUI_iss879ESS_GUI_iss959ESS_GUI_openclESS_GUI_orderingESS_GUI_sync_sascalccostrafo411magnetic_scattrelease-4.1.1release-4.1.2release-4.2.2release_4.0.1ticket-1009ticket-1094-headlessticket-1242-2d-resolutionticket-1243ticket-1249ticket885unittest-saveload

Last change on this file since e04f87b was b699768, checked in by Piotr Rozyczko <piotr.rozyczko@…>, 9 years ago
Initial commit of the refactored SasCalc? module.
Property mode set to `100644`
File size: 15.3 KB

Line
1	"""
2	ASCII reader
3	"""
4	############################################################################
5	#This software was developed by the University of Tennessee as part of the
6	#Distributed Data Analysis of Neutron Scattering Experiments (DANSE)
7	#project funded by the US National Science Foundation.
8	#If you use DANSE applications to do scientific research that leads to
9	#publication, we ask that you acknowledge the use of the software with the
10	#following sentence:
11	#This work benefited from DANSE software developed under NSF award DMR-0520547.
12	#copyright 2008, University of Tennessee
13	#############################################################################
14
15
16	import numpy
17	import os
18	from sas.sascalc.dataloader.data_info import Data1D
19
20	# Check whether we have a converter available
21	has_converter = True
22	try:
23	from sas.sascalc.data_util.nxsunit import Converter
24	except:
25	has_converter = False
26	_ZERO = 1e-16
27
28
29	class Reader:
30	"""
31	Class to load ascii files (2, 3 or 4 columns).
32	"""
33	## File type
34	type_name = "ASCII"
35
36	## Wildcards
37	type = ["ASCII files (.txt)\|.txt",
38	"ASCII files (.dat)\|.dat",
39	"ASCII files (.abs)\|.abs",
40	"CSV files (.csv)\|.csv"]
41	## List of allowed extensions
42	ext = ['.txt', '.TXT', '.dat', '.DAT', '.abs', '.ABS', 'csv', 'CSV']
43
44	## Flag to bypass extension check
45	allow_all = True
46
47	def read(self, path):
48	"""
49	Load data file
50
51	:param path: file path
52
53	:return: Data1D object, or None
54
55	:raise RuntimeError: when the file can't be opened
56	:raise ValueError: when the length of the data vectors are inconsistent
57	"""
58	if os.path.isfile(path):
59	basename = os.path.basename(path)
60	_, extension = os.path.splitext(basename)
61	if self.allow_all or extension.lower() in self.ext:
62	try:
63	# Read in binary mode since GRASP frequently has no-ascii
64	# characters that brakes the open operation
65	input_f = open(path,'rb')
66	except:
67	raise RuntimeError, "ascii_reader: cannot open %s" % path
68	buff = input_f.read()
69	lines = buff.splitlines()
70
71	x = numpy.zeros(0)
72	y = numpy.zeros(0)
73	dy = numpy.zeros(0)
74	dx = numpy.zeros(0)
75
76	#temp. space to sort data
77	tx = numpy.zeros(0)
78	ty = numpy.zeros(0)
79	tdy = numpy.zeros(0)
80	tdx = numpy.zeros(0)
81
82	output = Data1D(x, y, dy=dy, dx=dx)
83	self.filename = output.filename = basename
84
85	data_conv_q = None
86	data_conv_i = None
87
88	if has_converter == True and output.x_unit != '1/A':
89	data_conv_q = Converter('1/A')
90	# Test it
91	data_conv_q(1.0, output.x_unit)
92
93	if has_converter == True and output.y_unit != '1/cm':
94	data_conv_i = Converter('1/cm')
95	# Test it
96	data_conv_i(1.0, output.y_unit)
97
98	# The first good line of data will define whether
99	# we have 2-column or 3-column ascii
100	has_error_dx = None
101	has_error_dy = None
102
103	#Initialize counters for data lines and header lines.
104	is_data = False # Has more than 5 lines
105	# More than "5" lines of data is considered as actual
106	# data unless that is the only data
107	mum_data_lines = 5
108	# To count # of current data candidate lines
109	i = -1
110	# To count total # of previous data candidate lines
111	i1 = -1
112	# To count # of header lines
113	j = -1
114	# Helps to count # of header lines
115	j1 = -1
116	#minimum required number of columns of data; ( <= 4).
117	lentoks = 2
118	for line in lines:
119	# Initial try for CSV (split on ,)
120	toks = line.split(',')
121	# Now try SCSV (split on ;)
122	if len(toks) < 2:
123	toks = line.split(';')
124	# Now go for whitespace
125	if len(toks) < 2:
126	toks = line.split()
127	try:
128	#Make sure that all columns are numbers.
129	for colnum in range(len(toks)):
130	float(toks[colnum])
131
132	_x = float(toks[0])
133	_y = float(toks[1])
134
135	#Reset the header line counters
136	if j == j1:
137	j = 0
138	j1 = 0
139
140	if i > 1:
141	is_data = True
142
143	if data_conv_q is not None:
144	_x = data_conv_q(_x, units=output.x_unit)
145
146	if data_conv_i is not None:
147	_y = data_conv_i(_y, units=output.y_unit)
148
149	# If we have an extra token, check
150	# whether it can be interpreted as a
151	# third column.
152	_dy = None
153	if len(toks) > 2:
154	try:
155	_dy = float(toks[2])
156
157	if data_conv_i is not None:
158	_dy = data_conv_i(_dy, units=output.y_unit)
159
160	except:
161	# The third column is not a float, skip it.
162	pass
163
164	# If we haven't set the 3rd column
165	# flag, set it now.
166	if has_error_dy == None:
167	has_error_dy = False if _dy == None else True
168
169	#Check for dx
170	_dx = None
171	if len(toks) > 3:
172	try:
173	_dx = float(toks[3])
174
175	if data_conv_i is not None:
176	_dx = data_conv_i(_dx, units=output.x_unit)
177
178	except:
179	# The 4th column is not a float, skip it.
180	pass
181
182	# If we haven't set the 3rd column
183	# flag, set it now.
184	if has_error_dx == None:
185	has_error_dx = False if _dx == None else True
186
187	#After talked with PB, we decided to take care of only
188	# 4 columns of data for now.
189	#number of columns in the current line
190	#To remember the # of columns in the current
191	#line of data
192	new_lentoks = len(toks)
193
194	#If the previous columns not equal to the current,
195	#mark the previous as non-data and reset the dependents.
196	if lentoks != new_lentoks:
197	if is_data == True:
198	break
199	else:
200	i = -1
201	i1 = 0
202	j = -1
203	j1 = -1
204
205	#Delete the previously stored lines of data candidates
206	# if is not data.
207	if i < 0 and -1 < i1 < mum_data_lines and \
208	is_data == False:
209	try:
210	x = numpy.zeros(0)
211	y = numpy.zeros(0)
212	except:
213	pass
214
215	x = numpy.append(x, _x)
216	y = numpy.append(y, _y)
217
218	if has_error_dy == True:
219	#Delete the previously stored lines of
220	# data candidates if is not data.
221	if i < 0 and -1 < i1 < mum_data_lines and \
222	is_data == False:
223	try:
224	dy = numpy.zeros(0)
225	except:
226	pass
227	dy = numpy.append(dy, _dy)
228
229	if has_error_dx == True:
230	#Delete the previously stored lines of
231	# data candidates if is not data.
232	if i < 0 and -1 < i1 < mum_data_lines and \
233	is_data == False:
234	try:
235	dx = numpy.zeros(0)
236	except:
237	pass
238	dx = numpy.append(dx, _dx)
239
240	#Same for temp.
241	#Delete the previously stored lines of data candidates
242	# if is not data.
243	if i < 0 and -1 < i1 < mum_data_lines and\
244	is_data == False:
245	try:
246	tx = numpy.zeros(0)
247	ty = numpy.zeros(0)
248	except:
249	pass
250
251	tx = numpy.append(tx, _x)
252	ty = numpy.append(ty, _y)
253
254	if has_error_dy == True:
255	#Delete the previously stored lines of
256	# data candidates if is not data.
257	if i < 0 and -1 < i1 < mum_data_lines and \
258	is_data == False:
259	try:
260	tdy = numpy.zeros(0)
261	except:
262	pass
263	tdy = numpy.append(tdy, _dy)
264	if has_error_dx == True:
265	#Delete the previously stored lines of
266	# data candidates if is not data.
267	if i < 0 and -1 < i1 < mum_data_lines and \
268	is_data == False:
269	try:
270	tdx = numpy.zeros(0)
271	except:
272	pass
273	tdx = numpy.append(tdx, _dx)
274
275	#reset i1 and flag lentoks for the next
276	if lentoks < new_lentoks:
277	if is_data == False:
278	i1 = -1
279	#To remember the # of columns on the current line
280	# for the next line of data
281	lentoks = len(toks)
282
283	#Reset # of header lines and counts #
284	# of data candidate lines
285	if j == 0 and j1 == 0:
286	i1 = i + 1
287	i += 1
288	except:
289	# It is data and meet non - number, then stop reading
290	if is_data == True:
291	break
292	lentoks = 2
293	#Counting # of header lines
294	j += 1
295	if j == j1 + 1:
296	j1 = j
297	else:
298	j = -1
299	#Reset # of lines of data candidates
300	i = -1
301
302	# Couldn't parse this line, skip it
303	pass
304
305	input_f.close()
306	# Sanity check
307	if has_error_dy == True and not len(y) == len(dy):
308	msg = "ascii_reader: y and dy have different length"
309	raise RuntimeError, msg
310	if has_error_dx == True and not len(x) == len(dx):
311	msg = "ascii_reader: y and dy have different length"
312	raise RuntimeError, msg
313	# If the data length is zero, consider this as
314	# though we were not able to read the file.
315	if len(x) == 0:
316	raise RuntimeError, "ascii_reader: could not load file"
317
318	#Let's re-order the data to make cal.
319	# curve look better some cases
320	ind = numpy.lexsort((ty, tx))
321	for i in ind:
322	x[i] = tx[ind[i]]
323	y[i] = ty[ind[i]]
324	if has_error_dy == True:
325	dy[i] = tdy[ind[i]]
326	if has_error_dx == True:
327	dx[i] = tdx[ind[i]]
328	# Zeros in dx, dy
329	if has_error_dx:
330	dx[dx == 0] = _ZERO
331	if has_error_dy:
332	dy[dy == 0] = _ZERO
333	#Data
334	output.x = x[x != 0]
335	output.y = y[x != 0]
336	output.dy = dy[x != 0] if has_error_dy == True\
337	else numpy.zeros(len(output.y))
338	output.dx = dx[x != 0] if has_error_dx == True\
339	else numpy.zeros(len(output.x))
340
341	if data_conv_q is not None:
342	output.xaxis("\\rm{Q}", output.x_unit)
343	else:
344	output.xaxis("\\rm{Q}", 'A^{-1}')
345	if data_conv_i is not None:
346	output.yaxis("\\rm{Intensity}", output.y_unit)
347	else:
348	output.yaxis("\\rm{Intensity}", "cm^{-1}")
349
350	# Store loading process information
351	output.meta_data['loader'] = self.type_name
352	if len(output.x) < 1:
353	raise RuntimeError, "%s is empty" % path
354	return output
355
356	else:
357	raise RuntimeError, "%s is not a file" % path
358	return None

Note: See TracBrowser for help on using the repository browser.

SasView

source: sasview/src/sas/sascalc/dataloader/readers/ascii_reader.py @ e04f87b

Download in other formats: