ascii_reader.py @ d9dbb76

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 d9dbb76 was 0e5e586, checked in by Jae Cho <jhjcho@…>, 15 years ago
changed min. required # of lines (5 lines now) and enable to distinguish # of columns (even more than 4 col.)
Property mode set to `100644`
File size: 14.2 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	def read(self, path):
37	"""
38	Load data file
39
40	@param path: file path
41	@return: Data1D object, or None
42	@raise RuntimeError: when the file can't be opened
43	@raise ValueError: when the length of the data vectors are inconsistent
44	"""
45	if os.path.isfile(path):
46	basename = os.path.basename(path)
47	root, extension = os.path.splitext(basename)
48	if extension.lower() in self.ext:
49	try:
50	input_f = open(path,'r')
51	except :
52	raise RuntimeError, "ascii_reader: cannot open %s" % path
53	buff = input_f.read()
54	lines = buff.split('\n')
55
56	#Jae could not find python universal line spliter: keep the below for now
57	# some ascii data has \r line separator, try it when the data is on only one long line
58	if len(lines) < 2 :
59	lines = buff.split('\r')
60
61	x = numpy.zeros(0)
62	y = numpy.zeros(0)
63	dy = numpy.zeros(0)
64	dx = numpy.zeros(0)
65
66	#temp. space to sort data
67	tx = numpy.zeros(0)
68	ty = numpy.zeros(0)
69	tdy = numpy.zeros(0)
70	tdx = numpy.zeros(0)
71
72	output = Data1D(x, y, dy=dy, dx=dx)
73	self.filename = output.filename = basename
74
75	data_conv_q = None
76	data_conv_i = None
77
78	if has_converter == True and output.x_unit != '1/A':
79	data_conv_q = Converter('1/A')
80	# Test it
81	data_conv_q(1.0, output.x_unit)
82
83	if has_converter == True and output.y_unit != '1/cm':
84	data_conv_i = Converter('1/cm')
85	# Test it
86	data_conv_i(1.0, output.y_unit)
87
88
89	# The first good line of data will define whether
90	# we have 2-column or 3-column ascii
91	has_error_dx = None
92	has_error_dy = None
93
94	#Initialize counters for data lines and header lines.
95	is_data = False #Has more than 5 lines
96	mum_data_lines = 5 # More than "5" lines of data is considered as actual data unless that is the only data
97
98	i=-1 # To count # of current data candidate lines
99	i1=-1 # To count total # of previous data candidate lines
100	j=-1 # To count # of header lines
101	j1=-1 # Helps to count # of header lines
102	lentoks = 2 # minimum required number of columns of data; ( <= 4).
103
104	for line in lines:
105	toks = line.split()
106
107	try:
108
109	_x = float(toks[0])
110	_y = float(toks[1])
111
112	#Reset the header line counters
113	if j == j1:
114	j = 0
115	j1 = 0
116
117	if i > 1:
118	is_data = True
119
120	if data_conv_q is not None:
121	_x = data_conv_q(_x, units=output.x_unit)
122
123	if data_conv_i is not None:
124	_y = data_conv_i(_y, units=output.y_unit)
125
126	# If we have an extra token, check
127	# whether it can be interpreted as a
128	# third column.
129	_dy = None
130	if len(toks)>2:
131	try:
132	_dy = float(toks[2])
133
134	if data_conv_i is not None:
135	_dy = data_conv_i(_dy, units=output.y_unit)
136
137	except:
138	# The third column is not a float, skip it.
139	pass
140
141	# If we haven't set the 3rd column
142	# flag, set it now.
143	if has_error_dy == None:
144	has_error_dy = False if _dy == None else True
145
146	#Check for dx
147	_dx = None
148	if len(toks)>3:
149	try:
150	_dx = float(toks[3])
151
152	if data_conv_i is not None:
153	_dx = data_conv_i(_dx, units=output.x_unit)
154
155	except:
156	# The 4th column is not a float, skip it.
157	pass
158
159	# If we haven't set the 3rd column
160	# flag, set it now.
161	if has_error_dx == None:
162	has_error_dx = False if _dx == None else True
163
164	#After talked with PB, we decided to take care of only 4 columns of data for now.
165	#number of columns in the current line
166	#To remember the # of columns in the current line of data
167	new_lentoks = len(toks)
168
169	#If the previous columns not equal to the current, mark the previous as non-data and reset the dependents.
170	if lentoks != new_lentoks :
171	if is_data == True:
172	break
173	else:
174	i = -1
175	i1 = 0
176	j = -1
177	j1 = -1
178
179
180	#Delete the previously stored lines of data candidates if is not data.
181	if i < 0 and -1< i1 < mum_data_lines and is_data == False:
182	try:
183	x= numpy.zeros(0)
184	y= numpy.zeros(0)
185
186	except:
187	pass
188
189	x = numpy.append(x, _x)
190	y = numpy.append(y, _y)
191
192	if has_error_dy == True:
193	#Delete the previously stored lines of data candidates if is not data.
194	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
195	try:
196	dy = numpy.zeros(0)
197	except:
198	pass
199	dy = numpy.append(dy, _dy)
200
201	if has_error_dx == True:
202	#Delete the previously stored lines of data candidates if is not data.
203	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
204	try:
205	dx = numpy.zeros(0)
206	except:
207	pass
208	dx = numpy.append(dx, _dx)
209
210	#Same for temp.
211	#Delete the previously stored lines of data candidates if is not data.
212	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
213	try:
214	tx = numpy.zeros(0)
215	ty = numpy.zeros(0)
216	except:
217	pass
218
219	tx = numpy.append(tx, _x)
220	ty = numpy.append(ty, _y)
221
222	if has_error_dy == True:
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	tdy = numpy.zeros(0)
227	except:
228	pass
229	tdy = numpy.append(tdy, _dy)
230	if has_error_dx == True:
231	#Delete the previously stored lines of data candidates if is not data.
232	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
233	try:
234	tdx = numpy.zeros(0)
235	except:
236	pass
237	tdx = numpy.append(tdx, _dx)
238
239	#reset i1 and flag lentoks for the next
240	if lentoks < new_lentoks :
241	if is_data == False:
242	i1 = -1
243	#To remember the # of columns on the current line for the next line of data
244	lentoks = len(toks)
245
246	#Reset # of header lines and counts # of data candidate lines
247	if j == 0 and j1 ==0:
248	i1 = i + 1
249	i+=1
250
251	except:
252
253	# It is data and meet non - number, then stop reading
254	if is_data == True:
255	break
256	lentoks = 2
257	#Counting # of header lines
258	j+=1
259	if j == j1+1:
260	j1 = j
261	else:
262	j = -1
263	#Reset # of lines of data candidates
264	i = -1
265
266	# Couldn't parse this line, skip it
267	pass
268
269
270
271	# Sanity check
272	if has_error_dy == True and not len(y) == len(dy):
273	raise RuntimeError, "ascii_reader: y and dy have different length"
274	if has_error_dx == True and not len(x) == len(dx):
275	raise RuntimeError, "ascii_reader: y and dy have different length"
276
277	# If the data length is zero, consider this as
278	# though we were not able to read the file.
279	if len(x)==0:
280	raise RuntimeError, "ascii_reader: could not load file"
281
282	#Let's re-order the data to make cal. curve look better some cases
283	ind = numpy.lexsort((ty,tx))
284	for i in ind:
285	x[i] = tx[ind[i]]
286	y[i] = ty[ind[i]]
287	if has_error_dy == True:
288	dy[i] = tdy[ind[i]]
289	if has_error_dx == True:
290	dx[i] = tdx[ind[i]]
291
292
293	#Data
294	output.x = x
295	output.y = y
296	output.dy = dy if has_error_dy == True else None
297	output.dx = dx if has_error_dx == True else None
298
299	if data_conv_q is not None:
300	output.xaxis("\\rm{Q}", output.x_unit)
301	else:
302	output.xaxis("\\rm{Q}", 'A^{-1}')
303	if data_conv_i is not None:
304	output.yaxis("\\rm{Intensity}", output.y_unit)
305	else:
306	output.yaxis("\\rm{Intensity}","cm^{-1}")
307	return output
308
309	else:
310	raise RuntimeError, "%s is not a file" % path
311	return None
312
313	if __name__ == "__main__":
314	reader = Reader()
315	#print reader.read("../test/test_3_columns.txt")
316	print reader.read("../test/empty.txt")
317
318
319

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

SasView

source: sasview/DataLoader/readers/ascii_reader.py @ d9dbb76

Download in other formats: