ascii_reader.py @ c9795f6

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 c9795f6 was 3617aa2, checked in by Mathieu Doucet <doucetm@…>, 16 years ago
DataLoader?: Added a new unit test
Property mode set to `100644`
File size: 14.4 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 or 3 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 it has only one 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 3 lines
96	mum_data_lines = 3 # More than "3" 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	#To reject the line when reader meets less columns of data
113	if lentoks == 3:
114	_dy = float(toks[2])
115	elif lentoks == 4:
116	_dx = float(toks[3])
117
118	#Reset the header line counters
119	if j == j1:
120	j = 0
121	j1 = 0
122
123	if i > 1:
124	is_data = True
125
126	if data_conv_q is not None:
127	_x = data_conv_q(_x, units=output.x_unit)
128
129	if data_conv_i is not None:
130	_y = data_conv_i(_y, units=output.y_unit)
131
132	# If we have an extra token, check
133	# whether it can be interpreted as a
134	# third column.
135	_dy = None
136	if len(toks)>2:
137	try:
138	_dy = float(toks[2])
139
140	if data_conv_i is not None:
141	_dy = data_conv_i(_dy, units=output.y_unit)
142
143	except:
144	# The third column is not a float, skip it.
145	pass
146
147	# If we haven't set the 3rd column
148	# flag, set it now.
149	if has_error_dy == None:
150	has_error_dy = False if _dy == None else True
151
152	#Check for dx
153	_dx = None
154	if len(toks)>3:
155	try:
156	_dx = float(toks[3])
157
158	if data_conv_i is not None:
159	_dx = data_conv_i(_dx, units=output.x_unit)
160
161	except:
162	# The 4th 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_dx == None:
168	has_error_dx = False if _dx == None else True
169
170	#After talked with PB, we decided to take care of only 4 columns of data for now.
171	#number of columns in the current line
172	if len(toks)>= 4:
173	new_lentoks = 4
174	else:
175	new_lentoks = len(toks)
176
177	#If the previous columns are less than the current, mark the previous as non-data and reset the dependents.
178	if lentoks < new_lentoks :
179	if is_data == False:
180	i = -1
181	i1 = 0
182	j = -1
183	j1 = -1
184
185	#Delete the previously stored lines of data candidates if is not data.
186	if i < 0 and -1< i1 < mum_data_lines and is_data == False:
187	try:
188	x= numpy.zeros(0)
189	y= numpy.zeros(0)
190
191	except:
192	pass
193
194	x = numpy.append(x, _x)
195	y = numpy.append(y, _y)
196
197	if has_error_dy == True:
198	#Delete the previously stored lines of data candidates if is not data.
199	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
200	try:
201	dy = numpy.zeros(0)
202	except:
203	pass
204	dy = numpy.append(dy, _dy)
205
206	if has_error_dx == True:
207	#Delete the previously stored lines of data candidates if is not data.
208	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
209	try:
210	dx = numpy.zeros(0)
211	except:
212	pass
213	dx = numpy.append(dx, _dx)
214
215	#Same for temp.
216	#Delete the previously stored lines of data candidates if is not data.
217	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
218	try:
219	tx = numpy.zeros(0)
220	ty = numpy.zeros(0)
221	except:
222	pass
223
224	tx = numpy.append(tx, _x)
225	ty = numpy.append(ty, _y)
226
227	if has_error_dy == True:
228	#Delete the previously stored lines of data candidates if is not data.
229	if i < 0 and -1<i1 < mum_data_lines and is_data== False:
230	try:
231	tdy = numpy.zeros(0)
232	except:
233	pass
234	tdy = numpy.append(tdy, _dy)
235	if has_error_dx == True:
236	#Delete the previously stored lines of data candidates if is not data.
237	if i < 0 and -1< i1 < mum_data_lines and is_data== False:
238	try:
239	tdx = numpy.zeros(0)
240	except:
241	pass
242	tdx = numpy.append(tdx, _dx)
243
244	#reset i1 and flag lentoks for the next
245	if lentoks < new_lentoks :
246	if is_data == False:
247	i1 = -1
248	if len(toks)>= 4:
249	lentoks = 4
250	else:
251	lentoks = len(toks)
252
253	#Reset # of header lines and counts # of data candidate lines
254	if j == 0 and j1 ==0:
255	i1 = i + 1
256	i+=1
257
258	except:
259
260	# It is data and meet non - number, then stop reading
261	if is_data == True:
262	break
263	lentoks = 2
264	#Counting # of header lines
265	j+=1
266	if j == j1+1:
267	j1 = j
268	else:
269	j = -1
270	#Reset # of lines of data candidates
271	i = -1
272
273	# Couldn't parse this line, skip it
274	pass
275
276
277
278	# Sanity check
279	if has_error_dy == True and not len(y) == len(dy):
280	raise RuntimeError, "ascii_reader: y and dy have different length"
281	if has_error_dx == True and not len(x) == len(dx):
282	raise RuntimeError, "ascii_reader: y and dy have different length"
283
284	# If the data length is zero, consider this as
285	# though we were not able to read the file.
286	if len(x)==0:
287	raise RuntimeError, "ascii_reader: could not load file"
288
289	#Let's re-order the data to make cal. curve look better some cases
290	ind = numpy.lexsort((ty,tx))
291	for i in ind:
292	x[i] = tx[ind[i]]
293	y[i] = ty[ind[i]]
294	if has_error_dy == True:
295	dy[i] = tdy[ind[i]]
296	if has_error_dx == True:
297	dx[i] = tdx[ind[i]]
298
299
300	#Data
301	output.x = x
302	output.y = y
303	output.dy = dy if has_error_dy == True else None
304	output.dx = dx if has_error_dx == True else None
305
306	if data_conv_q is not None:
307	output.xaxis("\\rm{Q}", output.x_unit)
308	else:
309	output.xaxis("\\rm{Q}", 'A^{-1}')
310	if data_conv_i is not None:
311	output.yaxis("\\rm{Intensity}", output.y_unit)
312	else:
313	output.yaxis("\\rm{Intensity}","cm^{-1}")
314	return output
315
316	else:
317	raise RuntimeError, "%s is not a file" % path
318	return None
319
320	if __name__ == "__main__":
321	reader = Reader()
322	#print reader.read("../test/test_3_columns.txt")
323	print reader.read("../test/empty.txt")
324
325
326

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

SasView

source: sasview/DataLoader/readers/ascii_reader.py @ c9795f6

Download in other formats: