ascii_reader.py @ 6450d8c

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 6450d8c was fca90f82, checked in by Jae Cho <jhjcho@…>, 14 years ago
put zeros for dx,dy when no errs are given: asked by Mathieu
Property mode set to `100644`
File size: 15.0 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 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
26	class Reader:
27	"""
28	Class to load ascii files (2, 3 or 4 columns).
29	"""
30	## File type
31	type_name = "ASCII"
32
33	## Wildcards
34	type = ["ASCII files (.txt)\|.txt",
35	"ASCII files (.dat)\|.dat",
36	"ASCII files (.abs)\|.abs"]
37	## List of allowed extensions
38	ext=['.txt', '.TXT', '.dat', '.DAT', '.abs', '.ABS']
39
40	## Flag to bypass extension check
41	allow_all = True
42
43	def read(self, path):
44	"""
45	Load data file
46
47	:param path: file path
48
49	:return: Data1D object, or None
50
51	:raise RuntimeError: when the file can't be opened
52	:raise ValueError: when the length of the data vectors are inconsistent
53	"""
54	if os.path.isfile(path):
55	basename = os.path.basename(path)
56	root, extension = os.path.splitext(basename)
57	if self.allow_all or extension.lower() in self.ext:
58	try:
59	input_f = open(path,'r')
60	except :
61	raise RuntimeError, "ascii_reader: cannot open %s" % path
62	buff = input_f.read()
63	lines = buff.split('\n')
64
65	#Jae could not find python universal line spliter: keep the below for now
66	# some ascii data has \r line separator, try it when the data is on only one long line
67	if len(lines) < 2 :
68	lines = buff.split('\r')
69
70	x = numpy.zeros(0)
71	y = numpy.zeros(0)
72	dy = numpy.zeros(0)
73	dx = numpy.zeros(0)
74
75	#temp. space to sort data
76	tx = numpy.zeros(0)
77	ty = numpy.zeros(0)
78	tdy = numpy.zeros(0)
79	tdx = numpy.zeros(0)
80
81	output = Data1D(x, y, dy=dy, dx=dx)
82	self.filename = output.filename = basename
83
84	data_conv_q = None
85	data_conv_i = None
86
87	if has_converter == True and output.x_unit != '1/A':
88	data_conv_q = Converter('1/A')
89	# Test it
90	data_conv_q(1.0, output.x_unit)
91
92	if has_converter == True and output.y_unit != '1/cm':
93	data_conv_i = Converter('1/cm')
94	# Test it
95	data_conv_i(1.0, output.y_unit)
96
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	mum_data_lines = 5 # More than "5" lines of data is considered as actual data unless that is the only data
106
107	i=-1 # To count # of current data candidate lines
108	i1=-1 # To count total # of previous data candidate lines
109	j=-1 # To count # of header lines
110	j1=-1 # Helps to count # of header lines
111	lentoks = 2 # minimum required number of columns of data; ( <= 4).
112
113	for line in lines:
114	toks = line.split()
115
116	try:
117	#Make sure that all columns are numbers.
118	for colnum in range(len(toks)):
119	float(toks[colnum])
120
121	_x = float(toks[0])
122	_y = float(toks[1])
123
124	#Reset the header line counters
125	if j == j1:
126	j = 0
127	j1 = 0
128
129	if i > 1:
130	is_data = True
131
132	if data_conv_q is not None:
133	_x = data_conv_q(_x, units=output.x_unit)
134
135	if data_conv_i is not None:
136	_y = data_conv_i(_y, units=output.y_unit)
137
138	# If we have an extra token, check
139	# whether it can be interpreted as a
140	# third column.
141	_dy = None
142	if len(toks)>2:
143	try:
144	_dy = float(toks[2])
145
146	if data_conv_i is not None:
147	_dy = data_conv_i(_dy, units=output.y_unit)
148
149	except:
150	# The third column is not a float, skip it.
151	pass
152
153	# If we haven't set the 3rd column
154	# flag, set it now.
155	if has_error_dy == None:
156	has_error_dy = False if _dy == None else True
157
158	#Check for dx
159	_dx = None
160	if len(toks)>3:
161	try:
162	_dx = float(toks[3])
163
164	if data_conv_i is not None:
165	_dx = data_conv_i(_dx, units=output.x_unit)
166
167	except:
168	# The 4th column is not a float, skip it.
169	pass
170
171	# If we haven't set the 3rd column
172	# flag, set it now.
173	if has_error_dx == None:
174	has_error_dx = False if _dx == None else True
175
176	#After talked with PB, we decided to take care of only 4 columns of data for now.
177	#number of columns in the current line
178	#To remember the # of columns in the current line of data
179	new_lentoks = len(toks)
180
181	#If the previous columns not equal to the current, mark the previous as non-data and reset the dependents.
182	if lentoks != new_lentoks :
183	if is_data == True:
184	break
185	else:
186	i = -1
187	i1 = 0
188	j = -1
189	j1 = -1
190
191
192	#Delete the previously stored lines of data candidates if is not data.
193	if i < 0 and -1< i1 < mum_data_lines and is_data == False:
194	try:
195	x= numpy.zeros(0)
196	y= numpy.zeros(0)
197
198	except:
199	pass
200
201	x = numpy.append(x, _x)
202	y = numpy.append(y, _y)
203
204	if has_error_dy == True:
205	#Delete the previously stored lines of data candidates if is not data.
206	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
207	try:
208	dy = numpy.zeros(0)
209	except:
210	pass
211	dy = numpy.append(dy, _dy)
212
213	if has_error_dx == True:
214	#Delete the previously stored lines of data candidates if is not data.
215	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
216	try:
217	dx = numpy.zeros(0)
218	except:
219	pass
220	dx = numpy.append(dx, _dx)
221
222	#Same for temp.
223	#Delete the previously stored lines of data candidates if is not data.
224	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
225	try:
226	tx = numpy.zeros(0)
227	ty = numpy.zeros(0)
228	except:
229	pass
230
231	tx = numpy.append(tx, _x)
232	ty = numpy.append(ty, _y)
233
234	if has_error_dy == True:
235	#Delete the previously stored lines of data candidates if is not data.
236	if i < 0 and -1<i1 < mum_data_lines and is_data== False:
237	try:
238	tdy = numpy.zeros(0)
239	except:
240	pass
241	tdy = numpy.append(tdy, _dy)
242	if has_error_dx == True:
243	#Delete the previously stored lines of data candidates if is not data.
244	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
245	try:
246	tdx = numpy.zeros(0)
247	except:
248	pass
249	tdx = numpy.append(tdx, _dx)
250
251	#reset i1 and flag lentoks for the next
252	if lentoks < new_lentoks :
253	if is_data == False:
254	i1 = -1
255	#To remember the # of columns on the current line for the next line of data
256	lentoks = len(toks)
257
258	#Reset # of header lines and counts # of data candidate lines
259	if j == 0 and j1 ==0:
260	i1 = i + 1
261	i+=1
262
263	except:
264
265	# It is data and meet non - number, then stop reading
266	if is_data == True:
267	break
268	lentoks = 2
269	#Counting # of header lines
270	j+=1
271	if j == j1+1:
272	j1 = j
273	else:
274	j = -1
275	#Reset # of lines of data candidates
276	i = -1
277
278	# Couldn't parse this line, skip it
279	pass
280
281
282	input_f.close()
283	# Sanity check
284	if has_error_dy == True and not len(y) == len(dy):
285	raise RuntimeError, "ascii_reader: y and dy have different length"
286	if has_error_dx == True and not len(x) == len(dx):
287	raise RuntimeError, "ascii_reader: y and dy have different length"
288
289	# If the data length is zero, consider this as
290	# though we were not able to read the file.
291	if len(x)==0:
292	raise RuntimeError, "ascii_reader: could not load file"
293
294	#Let's re-order the data to make cal. curve look better some cases
295	ind = numpy.lexsort((ty,tx))
296	for i in ind:
297	x[i] = tx[ind[i]]
298	y[i] = ty[ind[i]]
299	if has_error_dy == True:
300	dy[i] = tdy[ind[i]]
301	if has_error_dx == True:
302	dx[i] = tdx[ind[i]]
303
304
305	#Data
306	output.x = x
307	output.y = y
308	output.dy = dy if has_error_dy == True else numpy.zeros(len(y))
309	output.dx = dx if has_error_dx == True else numpy.zeros(len(x))
310
311	if data_conv_q is not None:
312	output.xaxis("\\rm{Q}", output.x_unit)
313	else:
314	output.xaxis("\\rm{Q}", 'A^{-1}')
315	if data_conv_i is not None:
316	output.yaxis("\\rm{Intensity}", output.y_unit)
317	else:
318	output.yaxis("\\rm{Intensity}","cm^{-1}")
319
320	# Store loading process information
321	output.meta_data['loader'] = self.type_name
322
323	return output
324
325	else:
326	raise RuntimeError, "%s is not a file" % path
327	return None
328
329	if __name__ == "__main__":
330	reader = Reader()
331	#print reader.read("../test/test_3_columns.txt")
332	print reader.read("../test/empty.txt")
333
334
335

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

SasView

source: sasview/DataLoader/readers/ascii_reader.py @ 6450d8c

Download in other formats: