Changeset 67cc2f7 in sasview for src/sas/sascalc/dataloader/readers/ascii_reader.py
- Timestamp:
- Sep 13, 2016 4:25:23 PM (8 years ago)
- Branches:
- master, ESS_GUI, ESS_GUI_Docs, ESS_GUI_batch_fitting, ESS_GUI_bumps_abstraction, ESS_GUI_iss1116, ESS_GUI_iss879, ESS_GUI_iss959, ESS_GUI_opencl, ESS_GUI_ordering, ESS_GUI_sync_sascalc, costrafo411, magnetic_scatt, release-4.1.1, release-4.1.2, release-4.2.2, release_4.0.1, ticket-1009, ticket-1094-headless, ticket-1242-2d-resolution, ticket-1243, ticket-1249, ticket885, unittest-saveload
- Children:
- 7d94915
- Parents:
- af00478
- git-author:
- Jeff Krzywon <krzywon@…> (09/13/16 16:25:23)
- git-committer:
- Jeff KRzywon <krzywon@…> (09/13/16 16:25:23)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
src/sas/sascalc/dataloader/readers/ascii_reader.py
rb699768 r67cc2f7 33 33 ## File type 34 34 type_name = "ASCII" 35 35 36 36 ## Wildcards 37 37 type = ["ASCII files (*.txt)|*.txt", … … 41 41 ## List of allowed extensions 42 42 ext = ['.txt', '.TXT', '.dat', '.DAT', '.abs', '.ABS', 'csv', 'CSV'] 43 43 44 44 ## Flag to bypass extension check 45 45 allow_all = True 46 46 47 47 def read(self, path): 48 48 """ 49 49 Load data file 50 50 51 51 :param path: file path 52 53 52 :return: Data1D object, or None 54 53 55 54 :raise RuntimeError: when the file can't be opened 56 55 :raise ValueError: when the length of the data vectors are inconsistent … … 62 61 try: 63 62 # Read in binary mode since GRASP frequently has no-ascii 64 # characters that br akes the open operation63 # characters that breaks the open operation 65 64 input_f = open(path,'rb') 66 65 except: … … 68 67 buff = input_f.read() 69 68 lines = buff.splitlines() 70 71 x = numpy.zeros(0) 72 y = numpy.zeros(0) 73 dy = numpy.zeros(0) 74 dx = numpy.zeros(0) 75 76 #temp. space to sort data 77 tx = numpy.zeros(0) 78 ty = numpy.zeros(0) 69 70 # Arrays for data storage 71 tx = numpy.zeros(0) 72 ty = numpy.zeros(0) 79 73 tdy = numpy.zeros(0) 80 74 tdx = numpy.zeros(0) 81 82 output = Data1D(x, y, dy=dy, dx=dx) 83 self.filename = output.filename = basename 84 85 data_conv_q = None 86 data_conv_i = None 87 88 if has_converter == True and output.x_unit != '1/A': 89 data_conv_q = Converter('1/A') 90 # Test it 91 data_conv_q(1.0, output.x_unit) 92 93 if has_converter == True and output.y_unit != '1/cm': 94 data_conv_i = Converter('1/cm') 95 # Test it 96 data_conv_i(1.0, output.y_unit) 97 75 98 76 # The first good line of data will define whether 99 77 # we have 2-column or 3-column ascii 100 78 has_error_dx = None 101 79 has_error_dy = None 102 80 103 81 #Initialize counters for data lines and header lines. 104 is_data = False # Has more than 5 lines82 is_data = False 105 83 # More than "5" lines of data is considered as actual 106 84 # data unless that is the only data 107 m um_data_lines = 585 min_data_pts = 5 108 86 # To count # of current data candidate lines 109 i = -187 candidate_lines = 0 110 88 # To count total # of previous data candidate lines 111 i1 = -1 112 # To count # of header lines 113 j = -1 114 # Helps to count # of header lines 115 j1 = -1 116 #minimum required number of columns of data; ( <= 4). 89 candidate_lines_previous = 0 90 #minimum required number of columns of data 117 91 lentoks = 2 118 92 for line in lines: 119 # Initial try for CSV (split on ,) 120 toks = line.split(',') 121 # Now try SCSV (split on ;) 122 if len(toks) < 2: 123 toks = line.split(';') 124 # Now go for whitespace 125 if len(toks) < 2: 126 toks = line.split() 93 toks = self.splitline(line) 94 # To remember the # of columns in the current line of data 95 new_lentoks = len(toks) 127 96 try: 97 if new_lentoks == 1 and not is_data: 98 ## If only one item in list, no longer data 99 raise ValueError 100 elif new_lentoks == 0: 101 ## If the line is blank, skip and continue on 102 ## In case of breaks within data sets. 103 continue 104 elif new_lentoks != lentoks and is_data: 105 ## If a footer is found, break the loop and save the data 106 break 107 elif new_lentoks != lentoks and not is_data: 108 ## If header lines are numerical 109 candidate_lines = 0 110 candidate_lines_previous = 0 111 128 112 #Make sure that all columns are numbers. 129 113 for colnum in range(len(toks)): 114 # Any non-floating point values throw ValueError 130 115 float(toks[colnum]) 131 116 117 candidate_lines += 1 132 118 _x = float(toks[0]) 133 119 _y = float(toks[1]) 134 135 #Reset the header line counters 136 if j == j1: 137 j = 0 138 j1 = 0 139 140 if i > 1: 120 _dx = None 121 _dy = None 122 123 #If 5 or more lines, this is considering the set data 124 if candidate_lines >= min_data_pts: 141 125 is_data = True 142 143 if data_conv_q is not None: 144 _x = data_conv_q(_x, units=output.x_unit) 145 146 if data_conv_i is not None: 147 _y = data_conv_i(_y, units=output.y_unit) 148 149 # If we have an extra token, check 150 # whether it can be interpreted as a 151 # third column. 152 _dy = None 153 if len(toks) > 2: 154 try: 155 _dy = float(toks[2]) 156 157 if data_conv_i is not None: 158 _dy = data_conv_i(_dy, units=output.y_unit) 159 160 except: 161 # The third column is not a float, skip it. 162 pass 163 164 # If we haven't set the 3rd column 165 # flag, set it now. 166 if has_error_dy == None: 167 has_error_dy = False if _dy == None else True 168 169 #Check for dx 170 _dx = None 171 if len(toks) > 3: 172 try: 173 _dx = float(toks[3]) 174 175 if data_conv_i is not None: 176 _dx = data_conv_i(_dx, units=output.x_unit) 177 178 except: 179 # The 4th column is not a float, skip it. 180 pass 181 182 # If we haven't set the 3rd column 183 # flag, set it now. 184 if has_error_dx == None: 185 has_error_dx = False if _dx == None else True 186 187 #After talked with PB, we decided to take care of only 188 # 4 columns of data for now. 189 #number of columns in the current line 190 #To remember the # of columns in the current 191 #line of data 192 new_lentoks = len(toks) 193 194 #If the previous columns not equal to the current, 195 #mark the previous as non-data and reset the dependents. 196 if lentoks != new_lentoks: 197 if is_data == True: 198 break 199 else: 200 i = -1 201 i1 = 0 202 j = -1 203 j1 = -1 204 205 #Delete the previously stored lines of data candidates 206 # if is not data. 207 if i < 0 and -1 < i1 < mum_data_lines and \ 208 is_data == False: 209 try: 210 x = numpy.zeros(0) 211 y = numpy.zeros(0) 212 except: 213 pass 214 215 x = numpy.append(x, _x) 216 y = numpy.append(y, _y) 217 218 if has_error_dy == True: 219 #Delete the previously stored lines of 220 # data candidates if is not data. 221 if i < 0 and -1 < i1 < mum_data_lines and \ 222 is_data == False: 223 try: 224 dy = numpy.zeros(0) 225 except: 226 pass 227 dy = numpy.append(dy, _dy) 228 229 if has_error_dx == True: 230 #Delete the previously stored lines of 231 # data candidates if is not data. 232 if i < 0 and -1 < i1 < mum_data_lines and \ 233 is_data == False: 234 try: 235 dx = numpy.zeros(0) 236 except: 237 pass 238 dx = numpy.append(dx, _dx) 239 240 #Same for temp. 241 #Delete the previously stored lines of data candidates 242 # if is not data. 243 if i < 0 and -1 < i1 < mum_data_lines and\ 126 127 # If a 3rd row is present, consider it dy 128 if new_lentoks > 2: 129 _dy = float(toks[2]) 130 has_error_dy = False if _dy == None else True 131 132 # If a 4th row is present, consider it dx 133 if new_lentoks > 3: 134 _dx = float(toks[3]) 135 has_error_dx = False if _dx == None else True 136 137 # Delete the previously stored lines of data candidates if 138 # the list is not data 139 if candidate_lines == 1 and -1 < candidate_lines_previous < min_data_pts and \ 244 140 is_data == False: 245 141 try: 246 142 tx = numpy.zeros(0) 247 143 ty = numpy.zeros(0) 144 tdy = numpy.zeros(0) 145 tdx = numpy.zeros(0) 248 146 except: 249 147 pass 250 148 149 if has_error_dy == True: 150 tdy = numpy.append(tdy, _dy) 151 if has_error_dx == True: 152 tdx = numpy.append(tdx, _dx) 251 153 tx = numpy.append(tx, _x) 252 154 ty = numpy.append(ty, _y) 253 254 if has_error_dy == True: 255 #Delete the previously stored lines of 256 # data candidates if is not data. 257 if i < 0 and -1 < i1 < mum_data_lines and \ 258 is_data == False: 259 try: 260 tdy = numpy.zeros(0) 261 except: 262 pass 263 tdy = numpy.append(tdy, _dy) 264 if has_error_dx == True: 265 #Delete the previously stored lines of 266 # data candidates if is not data. 267 if i < 0 and -1 < i1 < mum_data_lines and \ 268 is_data == False: 269 try: 270 tdx = numpy.zeros(0) 271 except: 272 pass 273 tdx = numpy.append(tdx, _dx) 274 275 #reset i1 and flag lentoks for the next 276 if lentoks < new_lentoks: 277 if is_data == False: 278 i1 = -1 155 279 156 #To remember the # of columns on the current line 280 157 # for the next line of data 281 lentoks = len(toks) 282 283 #Reset # of header lines and counts # 284 # of data candidate lines 285 if j == 0 and j1 == 0: 286 i1 = i + 1 287 i += 1 288 except: 158 lentoks = new_lentoks 159 candidate_lines_previous = candidate_lines 160 except ValueError: 289 161 # It is data and meet non - number, then stop reading 290 162 if is_data == True: 291 163 break 292 164 lentoks = 2 293 #Counting # of header lines 294 j += 1 295 if j == j1 + 1: 296 j1 = j 297 else: 298 j = -1 165 has_error_dx = None 166 has_error_dy = None 299 167 #Reset # of lines of data candidates 300 i = -1 301 302 # Couldn't parse this line, skip it 168 candidate_lines = 0 169 except: 303 170 pass 304 171 305 172 input_f.close() 306 173 # Sanity check 307 if has_error_dy == True and not len( y) == len(dy):174 if has_error_dy == True and not len(ty) == len(tdy): 308 175 msg = "ascii_reader: y and dy have different length" 309 176 raise RuntimeError, msg 310 if has_error_dx == True and not len( x) == len(dx):177 if has_error_dx == True and not len(tx) == len(tdx): 311 178 msg = "ascii_reader: y and dy have different length" 312 179 raise RuntimeError, msg 313 180 # If the data length is zero, consider this as 314 181 # though we were not able to read the file. 315 if len( x) == 0:182 if len(tx) == 0: 316 183 raise RuntimeError, "ascii_reader: could not load file" 317 184 318 185 #Let's re-order the data to make cal. 319 186 # curve look better some cases 320 187 ind = numpy.lexsort((ty, tx)) 188 x = numpy.zeros(len(tx)) 189 y = numpy.zeros(len(ty)) 190 dy = numpy.zeros(len(tdy)) 191 dx = numpy.zeros(len(tdx)) 192 output = Data1D(x, y, dy=dy, dx=dx) 193 self.filename = output.filename = basename 194 321 195 for i in ind: 322 196 x[i] = tx[ind[i]] … … 338 212 output.dx = dx[x != 0] if has_error_dx == True\ 339 213 else numpy.zeros(len(output.x)) 340 341 if data_conv_q is not None: 342 output.xaxis("\\rm{Q}", output.x_unit) 343 else: 344 output.xaxis("\\rm{Q}", 'A^{-1}') 345 if data_conv_i is not None: 346 output.yaxis("\\rm{Intensity}", output.y_unit) 347 else: 348 output.yaxis("\\rm{Intensity}", "cm^{-1}") 349 214 215 output.xaxis("\\rm{Q}", 'A^{-1}') 216 output.yaxis("\\rm{Intensity}", "cm^{-1}") 217 350 218 # Store loading process information 351 219 output.meta_data['loader'] = self.type_name … … 353 221 raise RuntimeError, "%s is empty" % path 354 222 return output 355 223 356 224 else: 357 225 raise RuntimeError, "%s is not a file" % path 358 226 return None 227 228 def splitline(self, line): 229 """ 230 Splits a line into pieces based on common delimeters 231 :param line: A single line of text 232 :return: list of values 233 """ 234 # Initial try for CSV (split on ,) 235 toks = line.split(',') 236 # Now try SCSV (split on ;) 237 if len(toks) < 2: 238 toks = line.split(';') 239 # Now go for whitespace 240 if len(toks) < 2: 241 toks = line.split() 242 return toks
Note: See TracChangeset
for help on using the changeset viewer.