ascii_reader.py @ 9cd0baa

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 9cd0baa was 9cd0baa, checked in by Jae Cho <jhjcho@…>, 12 years ago
fixed problem with non-asciis in the file
Property mode set to `100644`
File size: 16.1 KB

Line
1
2
3	############################################################################
4	#This software was developed by the University of Tennessee as part of the
5	#Distributed Data Analysis of Neutron Scattering Experiments (DANSE)
6	#project funded by the US National Science Foundation.
7	#If you use DANSE applications to do scientific research that leads to
8	#publication, we ask that you acknowledge the use of the software with the
9	#following sentence:
10	#This work benefited from DANSE software developed under NSF award DMR-0520547.
11	#copyright 2008, University of Tennessee
12	#############################################################################
13
14
15	import numpy
16	import os
17	from sans.dataloader.data_info import Data1D
18
19	# Check whether we have a converter available
20	has_converter = True
21	try:
22	from data_util.nxsunit import Converter
23	except:
24	has_converter = False
25	_ZERO = 1e-16
26
27	class Reader:
28	"""
29	Class to load ascii files (2, 3 or 4 columns).
30	"""
31	## File type
32	type_name = "ASCII"
33
34	## Wildcards
35	type = ["ASCII files (.txt)\|.txt",
36	"ASCII files (.dat)\|.dat",
37	"ASCII files (.abs)\|.abs",
38	"CSV files (.csv)\|.csv"]
39	## List of allowed extensions
40	ext = ['.txt', '.TXT', '.dat', '.DAT', '.abs', '.ABS', 'csv', 'CSV']
41
42	## Flag to bypass extension check
43	allow_all = True
44
45	def read(self, path):
46	"""
47	Load data file
48
49	:param path: file path
50
51	:return: Data1D object, or None
52
53	:raise RuntimeError: when the file can't be opened
54	:raise ValueError: when the length of the data vectors are inconsistent
55	"""
56	if os.path.isfile(path):
57	basename = os.path.basename(path)
58	_, extension = os.path.splitext(basename)
59	if self.allow_all or extension.lower() in self.ext:
60	try:
61	# Read in binary mode since GRASP frequently has no-ascii
62	# characters that brakes the open operation
63	input_f = open(path,'rb')
64	except :
65	raise RuntimeError, "ascii_reader: cannot open %s" % path
66	buff = input_f.read()
67	lines = buff.split('\n')
68
69	#Jae could not find python universal line spliter:
70	#keep the below for now
71	# some ascii data has \r line separator,
72	# try it when the data is on only one long line
73	if len(lines) < 2 :
74	lines = buff.split('\r')
75
76	x = numpy.zeros(0)
77	y = numpy.zeros(0)
78	dy = numpy.zeros(0)
79	dx = numpy.zeros(0)
80
81	#temp. space to sort data
82	tx = numpy.zeros(0)
83	ty = numpy.zeros(0)
84	tdy = numpy.zeros(0)
85	tdx = numpy.zeros(0)
86
87	output = Data1D(x, y, dy=dy, dx=dx)
88	self.filename = output.filename = basename
89
90	data_conv_q = None
91	data_conv_i = None
92
93	if has_converter == True and output.x_unit != '1/A':
94	data_conv_q = Converter('1/A')
95	# Test it
96	data_conv_q(1.0, output.x_unit)
97
98	if has_converter == True and output.y_unit != '1/cm':
99	data_conv_i = Converter('1/cm')
100	# Test it
101	data_conv_i(1.0, output.y_unit)
102
103
104	# The first good line of data will define whether
105	# we have 2-column or 3-column ascii
106	has_error_dx = None
107	has_error_dy = None
108
109	#Initialize counters for data lines and header lines.
110	is_data = False #Has more than 5 lines
111	# More than "5" lines of data is considered as actual
112	# data unless that is the only data
113	mum_data_lines = 5
114	# To count # of current data candidate lines
115	i = -1
116	# To count total # of previous data candidate lines
117	i1 = -1
118	# To count # of header lines
119	j = -1
120	# Helps to count # of header lines
121	j1 = -1
122	#minimum required number of columns of data; ( <= 4).
123	lentoks = 2
124	for line in lines:
125	toks = line.split(',')
126	if len(toks) < 2:
127	toks = line.split()
128	try:
129	#Make sure that all columns are numbers.
130	for colnum in range(len(toks)):
131	float(toks[colnum])
132
133	_x = float(toks[0])
134	_y = float(toks[1])
135
136	#Reset the header line counters
137	if j == j1:
138	j = 0
139	j1 = 0
140
141	if i > 1:
142	is_data = True
143
144	if data_conv_q is not None:
145	_x = data_conv_q(_x, units=output.x_unit)
146
147	if data_conv_i is not None:
148	_y = data_conv_i(_y, units=output.y_unit)
149
150	# If we have an extra token, check
151	# whether it can be interpreted as a
152	# third column.
153	_dy = None
154	if len(toks) > 2:
155	try:
156	_dy = float(toks[2])
157
158	if data_conv_i is not None:
159	_dy = data_conv_i(_dy, units=output.y_unit)
160
161	except:
162	# The third column is not a float, skip it.
163	pass
164
165	# If we haven't set the 3rd column
166	# flag, set it now.
167	if has_error_dy == None:
168	has_error_dy = False if _dy == None else True
169
170	#Check for dx
171	_dx = None
172	if len(toks) > 3:
173	try:
174	_dx = float(toks[3])
175
176	if data_conv_i is not None:
177	_dx = data_conv_i(_dx, units=output.x_unit)
178
179	except:
180	# The 4th column is not a float, skip it.
181	pass
182
183	# If we haven't set the 3rd column
184	# flag, set it now.
185	if has_error_dx == None:
186	has_error_dx = False if _dx == None else True
187
188	#After talked with PB, we decided to take care of only
189	# 4 columns of data for now.
190	#number of columns in the current line
191	#To remember the # of columns in the current
192	#line of data
193	new_lentoks = len(toks)
194
195	#If the previous columns not equal to the current,
196	#mark the previous as non-data and reset the dependents.
197	if lentoks != new_lentoks :
198	if is_data == True:
199	break
200	else:
201	i = -1
202	i1 = 0
203	j = -1
204	j1 = -1
205
206
207	#Delete the previously stored lines of data candidates
208	# if is not data.
209	if i < 0 and -1 < i1 < mum_data_lines and \
210	is_data == False:
211	try:
212	x = numpy.zeros(0)
213	y = numpy.zeros(0)
214	except:
215	pass
216
217	x = numpy.append(x, _x)
218	y = numpy.append(y, _y)
219
220	if has_error_dy == True:
221	#Delete the previously stored lines of
222	# data candidates if is not data.
223	if i < 0 and -1 < i1 < mum_data_lines and \
224	is_data == False:
225	try:
226	dy = numpy.zeros(0)
227	except:
228	pass
229	dy = numpy.append(dy, _dy)
230
231	if has_error_dx == True:
232	#Delete the previously stored lines of
233	# data candidates if is not data.
234	if i < 0 and -1 < i1 < mum_data_lines and \
235	is_data == False:
236	try:
237	dx = numpy.zeros(0)
238	except:
239	pass
240	dx = numpy.append(dx, _dx)
241
242	#Same for temp.
243	#Delete the previously stored lines of data candidates
244	# if is not data.
245	if i < 0 and -1 < i1 < mum_data_lines and\
246	is_data == False:
247	try:
248	tx = numpy.zeros(0)
249	ty = numpy.zeros(0)
250	except:
251	pass
252
253	tx = numpy.append(tx, _x)
254	ty = numpy.append(ty, _y)
255
256	if has_error_dy == True:
257	#Delete the previously stored lines of
258	# data candidates if is not data.
259	if i < 0 and -1 < i1 < mum_data_lines and \
260	is_data == False:
261	try:
262	tdy = numpy.zeros(0)
263	except:
264	pass
265	tdy = numpy.append(tdy, _dy)
266	if has_error_dx == True:
267	#Delete the previously stored lines of
268	# data candidates if is not data.
269	if i < 0 and -1 < i1 < mum_data_lines and \
270	is_data == False:
271	try:
272	tdx = numpy.zeros(0)
273	except:
274	pass
275	tdx = numpy.append(tdx, _dx)
276
277	#reset i1 and flag lentoks for the next
278	if lentoks < new_lentoks:
279	if is_data == False:
280	i1 = -1
281	#To remember the # of columns on the current line
282	# for the next line of data
283	lentoks = len(toks)
284
285	#Reset # of header lines and counts #
286	# of data candidate lines
287	if j == 0 and j1 == 0:
288	i1 = i + 1
289	i += 1
290	except:
291
292	# It is data and meet non - number, then stop reading
293	if is_data == True:
294	break
295	lentoks = 2
296	#Counting # of header lines
297	j += 1
298	if j == j1 + 1:
299	j1 = j
300	else:
301	j = -1
302	#Reset # of lines of data candidates
303	i = -1
304
305	# Couldn't parse this line, skip it
306	pass
307
308	input_f.close()
309	# Sanity check
310	if has_error_dy == True and not len(y) == len(dy):
311	msg = "ascii_reader: y and dy have different length"
312	raise RuntimeError, msg
313	if has_error_dx == True and not len(x) == len(dx):
314	msg = "ascii_reader: y and dy have different length"
315	raise RuntimeError, msg
316	# If the data length is zero, consider this as
317	# though we were not able to read the file.
318	if len(x) == 0:
319	raise RuntimeError, "ascii_reader: could not load file"
320
321	#Let's re-order the data to make cal.
322	# curve look better some cases
323	ind = numpy.lexsort((ty, tx))
324	for i in ind:
325	x[i] = tx[ind[i]]
326	y[i] = ty[ind[i]]
327	if has_error_dy == True:
328	dy[i] = tdy[ind[i]]
329	if has_error_dx == True:
330	dx[i] = tdx[ind[i]]
331	# Zeros in dx, dy
332	if has_error_dx:
333	dx[dx==0] = _ZERO
334	if has_error_dy:
335	dy[dy==0] = _ZERO
336	#Data
337	output.x = x[x!=0]
338	output.y = y[x!=0]
339	output.dy = dy[x!=0] if has_error_dy == True else numpy.zeros(len(output.y))
340	output.dx = dx[x!=0] if has_error_dx == True else numpy.zeros(len(output.x))
341
342	if data_conv_q is not None:
343	output.xaxis("\\rm{Q}", output.x_unit)
344	else:
345	output.xaxis("\\rm{Q}", 'A^{-1}')
346	if data_conv_i is not None:
347	output.yaxis("\\rm{Intensity}", output.y_unit)
348	else:
349	output.yaxis("\\rm{Intensity}","cm^{-1}")
350
351	# Store loading process information
352	output.meta_data['loader'] = self.type_name
353
354	return output
355
356	else:
357	raise RuntimeError, "%s is not a file" % path
358	return None
359
360	if __name__ == "__main__":
361	reader = Reader()
362	#print reader.read("../test/test_3_columns.txt")
363	print reader.read("../test/empty.txt")
364
365
366

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

SasView

source: sasview/sansdataloader/src/sans/dataloader/readers/ascii_reader.py @ 9cd0baa

Download in other formats: