ascii_reader.py @ a96d246

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 a96d246 was e082e2c, checked in by Mathieu Doucet <doucetm@…>, 15 years ago
dataloader: allow all file extensions to be read by the ascii reader.
Property mode set to `100644`
File size: 14.3 KB

Line
1	"""
2	This software was developed by the University of Tennessee as part of the
3	Distributed Data Analysis of Neutron Scattering Experiments (DANSE)
4	project funded by the US National Science Foundation.
5
6	See the license text in license.txt
7
8	copyright 2008, University of Tennessee
9	"""
10
11	import numpy
12	import os
13	from DataLoader.data_info import Data1D
14
15	# Check whether we have a converter available
16	has_converter = True
17	try:
18	from data_util.nxsunit import Converter
19	except:
20	has_converter = False
21
22	class Reader:
23	"""
24	Class to load ascii files (2, 3 or 4 columns)
25	"""
26	## File type
27	type_name = "ASCII"
28
29	## Wildcards
30	type = ["ASCII files (.txt)\|.txt",
31	"ASCII files (.dat)\|.dat",
32	"ASCII files (.abs)\|.abs"]
33	## List of allowed extensions
34	ext=['.txt', '.TXT', '.dat', '.DAT', '.abs', '.ABS']
35
36	## Flag to bypass extension check
37	allow_all = True
38
39	def read(self, path):
40	"""
41	Load data file
42
43	@param path: file path
44	@return: Data1D object, or None
45	@raise RuntimeError: when the file can't be opened
46	@raise ValueError: when the length of the data vectors are inconsistent
47	"""
48	if os.path.isfile(path):
49	basename = os.path.basename(path)
50	root, extension = os.path.splitext(basename)
51	if self.allow_all or extension.lower() in self.ext:
52	try:
53	input_f = open(path,'r')
54	except :
55	raise RuntimeError, "ascii_reader: cannot open %s" % path
56	buff = input_f.read()
57	lines = buff.split('\n')
58
59	#Jae could not find python universal line spliter: keep the below for now
60	# some ascii data has \r line separator, try it when the data is on only one long line
61	if len(lines) < 2 :
62	lines = buff.split('\r')
63
64	x = numpy.zeros(0)
65	y = numpy.zeros(0)
66	dy = numpy.zeros(0)
67	dx = numpy.zeros(0)
68
69	#temp. space to sort data
70	tx = numpy.zeros(0)
71	ty = numpy.zeros(0)
72	tdy = numpy.zeros(0)
73	tdx = numpy.zeros(0)
74
75	output = Data1D(x, y, dy=dy, dx=dx)
76	self.filename = output.filename = basename
77
78	data_conv_q = None
79	data_conv_i = None
80
81	if has_converter == True and output.x_unit != '1/A':
82	data_conv_q = Converter('1/A')
83	# Test it
84	data_conv_q(1.0, output.x_unit)
85
86	if has_converter == True and output.y_unit != '1/cm':
87	data_conv_i = Converter('1/cm')
88	# Test it
89	data_conv_i(1.0, output.y_unit)
90
91
92	# The first good line of data will define whether
93	# we have 2-column or 3-column ascii
94	has_error_dx = None
95	has_error_dy = None
96
97	#Initialize counters for data lines and header lines.
98	is_data = False #Has more than 5 lines
99	mum_data_lines = 5 # More than "5" lines of data is considered as actual data unless that is the only data
100
101	i=-1 # To count # of current data candidate lines
102	i1=-1 # To count total # of previous data candidate lines
103	j=-1 # To count # of header lines
104	j1=-1 # Helps to count # of header lines
105	lentoks = 2 # minimum required number of columns of data; ( <= 4).
106
107	for line in lines:
108	toks = line.split()
109
110	try:
111
112	_x = float(toks[0])
113	_y = float(toks[1])
114
115	#Reset the header line counters
116	if j == j1:
117	j = 0
118	j1 = 0
119
120	if i > 1:
121	is_data = True
122
123	if data_conv_q is not None:
124	_x = data_conv_q(_x, units=output.x_unit)
125
126	if data_conv_i is not None:
127	_y = data_conv_i(_y, units=output.y_unit)
128
129	# If we have an extra token, check
130	# whether it can be interpreted as a
131	# third column.
132	_dy = None
133	if len(toks)>2:
134	try:
135	_dy = float(toks[2])
136
137	if data_conv_i is not None:
138	_dy = data_conv_i(_dy, units=output.y_unit)
139
140	except:
141	# The third column is not a float, skip it.
142	pass
143
144	# If we haven't set the 3rd column
145	# flag, set it now.
146	if has_error_dy == None:
147	has_error_dy = False if _dy == None else True
148
149	#Check for dx
150	_dx = None
151	if len(toks)>3:
152	try:
153	_dx = float(toks[3])
154
155	if data_conv_i is not None:
156	_dx = data_conv_i(_dx, units=output.x_unit)
157
158	except:
159	# The 4th column is not a float, skip it.
160	pass
161
162	# If we haven't set the 3rd column
163	# flag, set it now.
164	if has_error_dx == None:
165	has_error_dx = False if _dx == None else True
166
167	#After talked with PB, we decided to take care of only 4 columns of data for now.
168	#number of columns in the current line
169	#To remember the # of columns in the current line of data
170	new_lentoks = len(toks)
171
172	#If the previous columns not equal to the current, mark the previous as non-data and reset the dependents.
173	if lentoks != new_lentoks :
174	if is_data == True:
175	break
176	else:
177	i = -1
178	i1 = 0
179	j = -1
180	j1 = -1
181
182
183	#Delete the previously stored lines of data candidates if is not data.
184	if i < 0 and -1< i1 < mum_data_lines and is_data == False:
185	try:
186	x= numpy.zeros(0)
187	y= numpy.zeros(0)
188
189	except:
190	pass
191
192	x = numpy.append(x, _x)
193	y = numpy.append(y, _y)
194
195	if has_error_dy == True:
196	#Delete the previously stored lines of data candidates if is not data.
197	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
198	try:
199	dy = numpy.zeros(0)
200	except:
201	pass
202	dy = numpy.append(dy, _dy)
203
204	if has_error_dx == 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	dx = numpy.zeros(0)
209	except:
210	pass
211	dx = numpy.append(dx, _dx)
212
213	#Same for temp.
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	tx = numpy.zeros(0)
218	ty = numpy.zeros(0)
219	except:
220	pass
221
222	tx = numpy.append(tx, _x)
223	ty = numpy.append(ty, _y)
224
225	if has_error_dy == True:
226	#Delete the previously stored lines of data candidates if is not data.
227	if i < 0 and -1<i1 < mum_data_lines and is_data== False:
228	try:
229	tdy = numpy.zeros(0)
230	except:
231	pass
232	tdy = numpy.append(tdy, _dy)
233	if has_error_dx == True:
234	#Delete the previously stored lines of data candidates if is not data.
235	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
236	try:
237	tdx = numpy.zeros(0)
238	except:
239	pass
240	tdx = numpy.append(tdx, _dx)
241
242	#reset i1 and flag lentoks for the next
243	if lentoks < new_lentoks :
244	if is_data == False:
245	i1 = -1
246	#To remember the # of columns on the current line for the next line of data
247	lentoks = len(toks)
248
249	#Reset # of header lines and counts # of data candidate lines
250	if j == 0 and j1 ==0:
251	i1 = i + 1
252	i+=1
253
254	except:
255
256	# It is data and meet non - number, then stop reading
257	if is_data == True:
258	break
259	lentoks = 2
260	#Counting # of header lines
261	j+=1
262	if j == j1+1:
263	j1 = j
264	else:
265	j = -1
266	#Reset # of lines of data candidates
267	i = -1
268
269	# Couldn't parse this line, skip it
270	pass
271
272
273
274	# Sanity check
275	if has_error_dy == True and not len(y) == len(dy):
276	raise RuntimeError, "ascii_reader: y and dy have different length"
277	if has_error_dx == True and not len(x) == len(dx):
278	raise RuntimeError, "ascii_reader: y and dy have different length"
279
280	# If the data length is zero, consider this as
281	# though we were not able to read the file.
282	if len(x)==0:
283	raise RuntimeError, "ascii_reader: could not load file"
284
285	#Let's re-order the data to make cal. curve look better some cases
286	ind = numpy.lexsort((ty,tx))
287	for i in ind:
288	x[i] = tx[ind[i]]
289	y[i] = ty[ind[i]]
290	if has_error_dy == True:
291	dy[i] = tdy[ind[i]]
292	if has_error_dx == True:
293	dx[i] = tdx[ind[i]]
294
295
296	#Data
297	output.x = x
298	output.y = y
299	output.dy = dy if has_error_dy == True else None
300	output.dx = dx if has_error_dx == True else None
301
302	if data_conv_q is not None:
303	output.xaxis("\\rm{Q}", output.x_unit)
304	else:
305	output.xaxis("\\rm{Q}", 'A^{-1}')
306	if data_conv_i is not None:
307	output.yaxis("\\rm{Intensity}", output.y_unit)
308	else:
309	output.yaxis("\\rm{Intensity}","cm^{-1}")
310	return output
311
312	else:
313	raise RuntimeError, "%s is not a file" % path
314	return None
315
316	if __name__ == "__main__":
317	reader = Reader()
318	#print reader.read("../test/test_3_columns.txt")
319	print reader.read("../test/empty.txt")
320
321
322

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

SasView

source: sasview/DataLoader/readers/ascii_reader.py @ a96d246

Download in other formats: