ascii_reader.py @ a36a22a

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 a36a22a was 83b81b8, checked in by Jae Cho <jhjcho@…>, 13 years ago
a little better data error control
Property mode set to `100644`
File size: 15.4 KB

Rev	Line
[7d6351e]	1	"""
	2	ASCII reader
	3	"""
[0997158f]	4	############################################################################
	5	#This software was developed by the University of Tennessee as part of the
	6	#Distributed Data Analysis of Neutron Scattering Experiments (DANSE)
	7	#project funded by the US National Science Foundation.
[7d6351e]	8	#If you use DANSE applications to do scientific research that leads to
	9	#publication, we ask that you acknowledge the use of the software with the
[0997158f]	10	#following sentence:
[7d6351e]	11	#This work benefited from DANSE software developed under NSF award DMR-0520547.
[0997158f]	12	#copyright 2008, University of Tennessee
	13	#############################################################################
	14
[8bd8ea4]	15
	16	import numpy
	17	import os
[ad8034f]	18	from sans.dataloader.data_info import Data1D
[8bd8ea4]	19
[daa56d0]	20	# Check whether we have a converter available
[99d1af6]	21	has_converter = True
	22	try:
	23	from data_util.nxsunit import Converter
	24	except:
	25	has_converter = False
[da96629]	26	_ZERO = 1e-16
[99d1af6]	27
[7d6351e]	28
[8bd8ea4]	29	class Reader:
	30	"""
[0997158f]	31	Class to load ascii files (2, 3 or 4 columns).
[8bd8ea4]	32	"""
[8780e9a]	33	## File type
[28caa03]	34	type_name = "ASCII"
	35
	36	## Wildcards
[8780e9a]	37	type = ["ASCII files (.txt)\|.txt",
[470bf7e]	38	"ASCII files (.dat)\|.dat",
[ef9d209]	39	"ASCII files (.abs)\|.abs",
	40	"CSV files (.csv)\|.csv"]
[8bd8ea4]	41	## List of allowed extensions
[7d6351e]	42	ext = ['.txt', '.TXT', '.dat', '.DAT', '.abs', '.ABS', 'csv', 'CSV']
[8bd8ea4]	43
[e082e2c]	44	## Flag to bypass extension check
	45	allow_all = True
	46
[8bd8ea4]	47	def read(self, path):
[7d6351e]	48	"""
[0997158f]	49	Load data file
	50
	51	:param path: file path
	52
	53	:return: Data1D object, or None
	54
	55	:raise RuntimeError: when the file can't be opened
	56	:raise ValueError: when the length of the data vectors are inconsistent
[8bd8ea4]	57	"""
	58	if os.path.isfile(path):
[7d6351e]	59	basename = os.path.basename(path)
[a7a5886]	60	_, extension = os.path.splitext(basename)
[e082e2c]	61	if self.allow_all or extension.lower() in self.ext:
[8bd8ea4]	62	try:
[9cd0baa]	63	# Read in binary mode since GRASP frequently has no-ascii
	64	# characters that brakes the open operation
[7d6351e]	65	input_f = open(path,'rb')
	66	except:
[8bd8ea4]	67	raise RuntimeError, "ascii_reader: cannot open %s" % path
	68	buff = input_f.read()
	69	lines = buff.split('\n')
[470bf7e]	70
[7d6351e]	71	#Jae could not find python universal line spliter:
[a7a5886]	72	#keep the below for now
	73	# some ascii data has \r line separator,
	74	# try it when the data is on only one long line
[7d6351e]	75	if len(lines) < 2 :
[470bf7e]	76	lines = buff.split('\r')
	77
[8bd8ea4]	78	x = numpy.zeros(0)
	79	y = numpy.zeros(0)
	80	dy = numpy.zeros(0)
[de1da34]	81	dx = numpy.zeros(0)
	82
	83	#temp. space to sort data
	84	tx = numpy.zeros(0)
	85	ty = numpy.zeros(0)
	86	tdy = numpy.zeros(0)
	87	tdx = numpy.zeros(0)
	88
	89	output = Data1D(x, y, dy=dy, dx=dx)
[8bd8ea4]	90	self.filename = output.filename = basename
[99d1af6]	91
	92	data_conv_q = None
	93	data_conv_i = None
	94
[ca10d8e]	95	if has_converter == True and output.x_unit != '1/A':
	96	data_conv_q = Converter('1/A')
[99d1af6]	97	# Test it
	98	data_conv_q(1.0, output.x_unit)
	99
[ca10d8e]	100	if has_converter == True and output.y_unit != '1/cm':
	101	data_conv_i = Converter('1/cm')
[99d1af6]	102	# Test it
	103	data_conv_i(1.0, output.y_unit)
	104
[8bd8ea4]	105	# The first good line of data will define whether
	106	# we have 2-column or 3-column ascii
[de1da34]	107	has_error_dx = None
	108	has_error_dy = None
[8bd8ea4]	109
[892f246]	110	#Initialize counters for data lines and header lines.
[7d6351e]	111	is_data = False # Has more than 5 lines
[a7a5886]	112	# More than "5" lines of data is considered as actual
	113	# data unless that is the only data
[7d6351e]	114	mum_data_lines = 5
[a7a5886]	115	# To count # of current data candidate lines
[7d6351e]	116	i = -1
	117	# To count total # of previous data candidate lines
	118	i1 = -1
	119	# To count # of header lines
[a7a5886]	120	j = -1
[7d6351e]	121	# Helps to count # of header lines
[a7a5886]	122	j1 = -1
[7d6351e]	123	#minimum required number of columns of data; ( <= 4).
	124	lentoks = 2
[8bd8ea4]	125	for line in lines:
[ef9d209]	126	toks = line.split(',')
	127	if len(toks) < 2:
	128	toks = line.split()
[8bd8ea4]	129	try:
[5f2d3c78]	130	#Make sure that all columns are numbers.
	131	for colnum in range(len(toks)):
	132	float(toks[colnum])
	133
[8bd8ea4]	134	_x = float(toks[0])
	135	_y = float(toks[1])
	136
[892f246]	137	#Reset the header line counters
	138	if j == j1:
	139	j = 0
	140	j1 = 0
	141
	142	if i > 1:
	143	is_data = True
[d508be9]	144
[99d1af6]	145	if data_conv_q is not None:
	146	_x = data_conv_q(_x, units=output.x_unit)
	147
	148	if data_conv_i is not None:
[7d6351e]	149	_y = data_conv_i(_y, units=output.y_unit)
[99d1af6]	150
[8bd8ea4]	151	# If we have an extra token, check
	152	# whether it can be interpreted as a
	153	# third column.
	154	_dy = None
[a7a5886]	155	if len(toks) > 2:
[8bd8ea4]	156	try:
	157	_dy = float(toks[2])
[99d1af6]	158
	159	if data_conv_i is not None:
	160	_dy = data_conv_i(_dy, units=output.y_unit)
	161
[8bd8ea4]	162	except:
	163	# The third column is not a float, skip it.
	164	pass
	165
	166	# If we haven't set the 3rd column
	167	# flag, set it now.
[de1da34]	168	if has_error_dy == None:
	169	has_error_dy = False if _dy == None else True
	170
	171	#Check for dx
	172	_dx = None
[a7a5886]	173	if len(toks) > 3:
[de1da34]	174	try:
	175	_dx = float(toks[3])
	176
	177	if data_conv_i is not None:
	178	_dx = data_conv_i(_dx, units=output.x_unit)
	179
	180	except:
	181	# The 4th column is not a float, skip it.
	182	pass
	183
	184	# If we haven't set the 3rd column
	185	# flag, set it now.
	186	if has_error_dx == None:
	187	has_error_dx = False if _dx == None else True
[892f246]	188
[7d6351e]	189	#After talked with PB, we decided to take care of only
[a7a5886]	190	# 4 columns of data for now.
[d508be9]	191	#number of columns in the current line
[7d6351e]	192	#To remember the # of columns in the current
[a7a5886]	193	#line of data
[0e5e586]	194	new_lentoks = len(toks)
[d508be9]	195
[7d6351e]	196	#If the previous columns not equal to the current,
	197	#mark the previous as non-data and reset the dependents.
	198	if lentoks != new_lentoks:
[272b107]	199	if is_data == True:
	200	break
	201	else:
[d508be9]	202	i = -1
	203	i1 = 0
	204	j = -1
	205	j1 = -1
	206
[a7a5886]	207	#Delete the previously stored lines of data candidates
	208	# if is not data.
	209	if i < 0 and -1 < i1 < mum_data_lines and \
	210	is_data == False:
[892f246]	211	try:
[a7a5886]	212	x = numpy.zeros(0)
	213	y = numpy.zeros(0)
[892f246]	214	except:
	215	pass
	216
[7d6351e]	217	x = numpy.append(x, _x)
	218	y = numpy.append(y, _y)
[892f246]	219
[de1da34]	220	if has_error_dy == True:
[a7a5886]	221	#Delete the previously stored lines of
	222	# data candidates if is not data.
	223	if i < 0 and -1 < i1 < mum_data_lines and \
	224	is_data == False:
[892f246]	225	try:
[7d6351e]	226	dy = numpy.zeros(0)
[892f246]	227	except:
[7d6351e]	228	pass
[8bd8ea4]	229	dy = numpy.append(dy, _dy)
[892f246]	230
[de1da34]	231	if has_error_dx == True:
[a7a5886]	232	#Delete the previously stored lines of
	233	# data candidates if is not data.
	234	if i < 0 and -1 < i1 < mum_data_lines and \
	235	is_data == False:
[892f246]	236	try:
[7d6351e]	237	dx = numpy.zeros(0)
[892f246]	238	except:
[7d6351e]	239	pass
[de1da34]	240	dx = numpy.append(dx, _dx)
	241
	242	#Same for temp.
[a7a5886]	243	#Delete the previously stored lines of data candidates
	244	# if is not data.
	245	if i < 0 and -1 < i1 < mum_data_lines and\
	246	is_data == False:
[892f246]	247	try:
	248	tx = numpy.zeros(0)
	249	ty = numpy.zeros(0)
	250	except:
[7d6351e]	251	pass
[892f246]	252
[7d6351e]	253	tx = numpy.append(tx, _x)
	254	ty = numpy.append(ty, _y)
[892f246]	255
[de1da34]	256	if has_error_dy == True:
[a7a5886]	257	#Delete the previously stored lines of
	258	# data candidates if is not data.
	259	if i < 0 and -1 < i1 < mum_data_lines and \
	260	is_data == False:
[892f246]	261	try:
	262	tdy = numpy.zeros(0)
	263	except:
[7d6351e]	264	pass
[de1da34]	265	tdy = numpy.append(tdy, _dy)
	266	if has_error_dx == True:
[a7a5886]	267	#Delete the previously stored lines of
	268	# data candidates if is not data.
	269	if i < 0 and -1 < i1 < mum_data_lines and \
	270	is_data == False:
[892f246]	271	try:
	272	tdx = numpy.zeros(0)
	273	except:
[7d6351e]	274	pass
[de1da34]	275	tdx = numpy.append(tdx, _dx)
[d508be9]	276
	277	#reset i1 and flag lentoks for the next
[a7a5886]	278	if lentoks < new_lentoks:
[d508be9]	279	if is_data == False:
[7d6351e]	280	i1 = -1
[a7a5886]	281	#To remember the # of columns on the current line
	282	# for the next line of data
[0e5e586]	283	lentoks = len(toks)
[8bd8ea4]	284
[7d6351e]	285	#Reset # of header lines and counts #
	286	# of data candidate lines
[a7a5886]	287	if j == 0 and j1 == 0:
[7d6351e]	288	i1 = i + 1
[a7a5886]	289	i += 1
[8bd8ea4]	290	except:
[892f246]	291
	292	# It is data and meet non - number, then stop reading
	293	if is_data == True:
[7d6351e]	294	break
[d508be9]	295	lentoks = 2
[7d6351e]	296	#Counting # of header lines
[a7a5886]	297	j += 1
	298	if j == j1 + 1:
[7d6351e]	299	j1 = j
	300	else:
[892f246]	301	j = -1
	302	#Reset # of lines of data candidates
	303	i = -1
	304
[7d6351e]	305	# Couldn't parse this line, skip it
[8bd8ea4]	306	pass
[892f246]	307
[7d6351e]	308	input_f.close()
[8bd8ea4]	309	# Sanity check
[de1da34]	310	if has_error_dy == True and not len(y) == len(dy):
[a7a5886]	311	msg = "ascii_reader: y and dy have different length"
	312	raise RuntimeError, msg
[de1da34]	313	if has_error_dx == True and not len(x) == len(dx):
[a7a5886]	314	msg = "ascii_reader: y and dy have different length"
	315	raise RuntimeError, msg
[8bd8ea4]	316	# If the data length is zero, consider this as
	317	# though we were not able to read the file.
[a7a5886]	318	if len(x) == 0:
[daa56d0]	319	raise RuntimeError, "ascii_reader: could not load file"
[de1da34]	320
[a7a5886]	321	#Let's re-order the data to make cal.
	322	# curve look better some cases
	323	ind = numpy.lexsort((ty, tx))
[de1da34]	324	for i in ind:
	325	x[i] = tx[ind[i]]
	326	y[i] = ty[ind[i]]
	327	if has_error_dy == True:
	328	dy[i] = tdy[ind[i]]
	329	if has_error_dx == True:
	330	dx[i] = tdx[ind[i]]
[7d6351e]	331	# Zeros in dx, dy
[da96629]	332	if has_error_dx:
[7d6351e]	333	dx[dx == 0] = _ZERO
[da96629]	334	if has_error_dy:
[7d6351e]	335	dy[dy == 0] = _ZERO
	336	#Data
	337	output.x = x[x != 0]
	338	output.y = y[x != 0]
	339	output.dy = dy[x != 0] if has_error_dy == True\
	340	else numpy.zeros(len(output.y))
	341	output.dx = dx[x != 0] if has_error_dx == True\
	342	else numpy.zeros(len(output.x))
[fca90f82]	343
[99d1af6]	344	if data_conv_q is not None:
	345	output.xaxis("\\rm{Q}", output.x_unit)
	346	else:
	347	output.xaxis("\\rm{Q}", 'A^{-1}')
	348	if data_conv_i is not None:
[0e2aa40]	349	output.yaxis("\\rm{Intensity}", output.y_unit)
[99d1af6]	350	else:
[7d6351e]	351	output.yaxis("\\rm{Intensity}", "cm^{-1}")
[fe78c7b]	352
	353	# Store loading process information
[7d6351e]	354	output.meta_data['loader'] = self.type_name
[83b81b8]	355	if len(output.x) < 1:
	356	raise RuntimeError, "%s is empty" % path
[8bd8ea4]	357	return output
[892f246]	358
[8bd8ea4]	359	else:
	360	raise RuntimeError, "%s is not a file" % path
	361	return None

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

SasView

source: sasview/sansdataloader/src/sans/dataloader/readers/ascii_reader.py @ a36a22a

Download in other formats: