ascii_reader.py @ f1044e9

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 f1044e9 was ad8034f, checked in by Jae Cho <jhjcho@…>, 13 years ago
updated dataloader calls
Property mode set to `100644`
File size: 15.9 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	## List of allowed extensions
39	ext = ['.txt', '.TXT', '.dat', '.DAT', '.abs', '.ABS']
40
41	## Flag to bypass extension check
42	allow_all = True
43
44	def read(self, path):
45	"""
46	Load data file
47
48	:param path: file path
49
50	:return: Data1D object, or None
51
52	:raise RuntimeError: when the file can't be opened
53	:raise ValueError: when the length of the data vectors are inconsistent
54	"""
55	if os.path.isfile(path):
56	basename = os.path.basename(path)
57	_, extension = os.path.splitext(basename)
58	if self.allow_all or extension.lower() in self.ext:
59	try:
60	input_f = open(path,'r')
61	except :
62	raise RuntimeError, "ascii_reader: cannot open %s" % path
63	buff = input_f.read()
64	lines = buff.split('\n')
65
66	#Jae could not find python universal line spliter:
67	#keep the below for now
68	# some ascii data has \r line separator,
69	# try it when the data is on only one long line
70	if len(lines) < 2 :
71	lines = buff.split('\r')
72
73	x = numpy.zeros(0)
74	y = numpy.zeros(0)
75	dy = numpy.zeros(0)
76	dx = numpy.zeros(0)
77
78	#temp. space to sort data
79	tx = numpy.zeros(0)
80	ty = numpy.zeros(0)
81	tdy = numpy.zeros(0)
82	tdx = numpy.zeros(0)
83
84	output = Data1D(x, y, dy=dy, dx=dx)
85	self.filename = output.filename = basename
86
87	data_conv_q = None
88	data_conv_i = None
89
90	if has_converter == True and output.x_unit != '1/A':
91	data_conv_q = Converter('1/A')
92	# Test it
93	data_conv_q(1.0, output.x_unit)
94
95	if has_converter == True and output.y_unit != '1/cm':
96	data_conv_i = Converter('1/cm')
97	# Test it
98	data_conv_i(1.0, output.y_unit)
99
100
101	# The first good line of data will define whether
102	# we have 2-column or 3-column ascii
103	has_error_dx = None
104	has_error_dy = None
105
106	#Initialize counters for data lines and header lines.
107	is_data = False #Has more than 5 lines
108	# More than "5" lines of data is considered as actual
109	# data unless that is the only data
110	mum_data_lines = 5
111	# To count # of current data candidate lines
112	i = -1
113	# To count total # of previous data candidate lines
114	i1 = -1
115	# To count # of header lines
116	j = -1
117	# Helps to count # of header lines
118	j1 = -1
119	#minimum required number of columns of data; ( <= 4).
120	lentoks = 2
121	for line in lines:
122	toks = line.split()
123	try:
124	#Make sure that all columns are numbers.
125	for colnum in range(len(toks)):
126	float(toks[colnum])
127
128	_x = float(toks[0])
129	_y = float(toks[1])
130
131	#Reset the header line counters
132	if j == j1:
133	j = 0
134	j1 = 0
135
136	if i > 1:
137	is_data = True
138
139	if data_conv_q is not None:
140	_x = data_conv_q(_x, units=output.x_unit)
141
142	if data_conv_i is not None:
143	_y = data_conv_i(_y, units=output.y_unit)
144
145	# If we have an extra token, check
146	# whether it can be interpreted as a
147	# third column.
148	_dy = None
149	if len(toks) > 2:
150	try:
151	_dy = float(toks[2])
152
153	if data_conv_i is not None:
154	_dy = data_conv_i(_dy, units=output.y_unit)
155
156	except:
157	# The third column is not a float, skip it.
158	pass
159
160	# If we haven't set the 3rd column
161	# flag, set it now.
162	if has_error_dy == None:
163	has_error_dy = False if _dy == None else True
164
165	#Check for dx
166	_dx = None
167	if len(toks) > 3:
168	try:
169	_dx = float(toks[3])
170
171	if data_conv_i is not None:
172	_dx = data_conv_i(_dx, units=output.x_unit)
173
174	except:
175	# The 4th column is not a float, skip it.
176	pass
177
178	# If we haven't set the 3rd column
179	# flag, set it now.
180	if has_error_dx == None:
181	has_error_dx = False if _dx == None else True
182
183	#After talked with PB, we decided to take care of only
184	# 4 columns of data for now.
185	#number of columns in the current line
186	#To remember the # of columns in the current
187	#line of data
188	new_lentoks = len(toks)
189
190	#If the previous columns not equal to the current,
191	#mark the previous as non-data and reset the dependents.
192	if lentoks != new_lentoks :
193	if is_data == True:
194	break
195	else:
196	i = -1
197	i1 = 0
198	j = -1
199	j1 = -1
200
201
202	#Delete the previously stored lines of data candidates
203	# if is not data.
204	if i < 0 and -1 < i1 < mum_data_lines and \
205	is_data == False:
206	try:
207	x = numpy.zeros(0)
208	y = numpy.zeros(0)
209	except:
210	pass
211
212	x = numpy.append(x, _x)
213	y = numpy.append(y, _y)
214
215	if has_error_dy == True:
216	#Delete the previously stored lines of
217	# data candidates if is not data.
218	if i < 0 and -1 < i1 < mum_data_lines and \
219	is_data == False:
220	try:
221	dy = numpy.zeros(0)
222	except:
223	pass
224	dy = numpy.append(dy, _dy)
225
226	if has_error_dx == True:
227	#Delete the previously stored lines of
228	# data candidates if is not data.
229	if i < 0 and -1 < i1 < mum_data_lines and \
230	is_data == False:
231	try:
232	dx = numpy.zeros(0)
233	except:
234	pass
235	dx = numpy.append(dx, _dx)
236
237	#Same for temp.
238	#Delete the previously stored lines of data candidates
239	# if is not data.
240	if i < 0 and -1 < i1 < mum_data_lines and\
241	is_data == False:
242	try:
243	tx = numpy.zeros(0)
244	ty = numpy.zeros(0)
245	except:
246	pass
247
248	tx = numpy.append(tx, _x)
249	ty = numpy.append(ty, _y)
250
251	if has_error_dy == True:
252	#Delete the previously stored lines of
253	# data candidates if is not data.
254	if i < 0 and -1 < i1 < mum_data_lines and \
255	is_data == False:
256	try:
257	tdy = numpy.zeros(0)
258	except:
259	pass
260	tdy = numpy.append(tdy, _dy)
261	if has_error_dx == True:
262	#Delete the previously stored lines of
263	# data candidates if is not data.
264	if i < 0 and -1 < i1 < mum_data_lines and \
265	is_data == False:
266	try:
267	tdx = numpy.zeros(0)
268	except:
269	pass
270	tdx = numpy.append(tdx, _dx)
271
272	#reset i1 and flag lentoks for the next
273	if lentoks < new_lentoks:
274	if is_data == False:
275	i1 = -1
276	#To remember the # of columns on the current line
277	# for the next line of data
278	lentoks = len(toks)
279
280	#Reset # of header lines and counts #
281	# of data candidate lines
282	if j == 0 and j1 == 0:
283	i1 = i + 1
284	i += 1
285	except:
286
287	# It is data and meet non - number, then stop reading
288	if is_data == True:
289	break
290	lentoks = 2
291	#Counting # of header lines
292	j += 1
293	if j == j1 + 1:
294	j1 = j
295	else:
296	j = -1
297	#Reset # of lines of data candidates
298	i = -1
299
300	# Couldn't parse this line, skip it
301	pass
302
303	input_f.close()
304	# Sanity check
305	if has_error_dy == True and not len(y) == len(dy):
306	msg = "ascii_reader: y and dy have different length"
307	raise RuntimeError, msg
308	if has_error_dx == True and not len(x) == len(dx):
309	msg = "ascii_reader: y and dy have different length"
310	raise RuntimeError, msg
311	# If the data length is zero, consider this as
312	# though we were not able to read the file.
313	if len(x) == 0:
314	raise RuntimeError, "ascii_reader: could not load file"
315
316	#Let's re-order the data to make cal.
317	# curve look better some cases
318	ind = numpy.lexsort((ty, tx))
319	for i in ind:
320	x[i] = tx[ind[i]]
321	y[i] = ty[ind[i]]
322	if has_error_dy == True:
323	dy[i] = tdy[ind[i]]
324	if has_error_dx == True:
325	dx[i] = tdx[ind[i]]
326	# Zeros in dx, dy
327	if has_error_dx:
328	dx[dx==0] = _ZERO
329	if has_error_dy:
330	dy[dy==0] = _ZERO
331	#Data
332	output.x = x[x!=0]
333	output.y = y[x!=0]
334	output.dy = dy[x!=0] if has_error_dy == True else numpy.zeros(len(output.y))
335	output.dx = dx[x!=0] if has_error_dx == True else numpy.zeros(len(output.x))
336
337	if data_conv_q is not None:
338	output.xaxis("\\rm{Q}", output.x_unit)
339	else:
340	output.xaxis("\\rm{Q}", 'A^{-1}')
341	if data_conv_i is not None:
342	output.yaxis("\\rm{Intensity}", output.y_unit)
343	else:
344	output.yaxis("\\rm{Intensity}","cm^{-1}")
345
346	# Store loading process information
347	output.meta_data['loader'] = self.type_name
348
349	return output
350
351	else:
352	raise RuntimeError, "%s is not a file" % path
353	return None
354
355	if __name__ == "__main__":
356	reader = Reader()
357	#print reader.read("../test/test_3_columns.txt")
358	print reader.read("../test/empty.txt")
359
360
361

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

SasView

source: sasview/sansdataloader/src/sans/dataloader/readers/ascii_reader.py @ f1044e9

Download in other formats: