[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 |
---|
[79492222] | 18 | from sas.dataloader.data_info import Data1D |
---|
[8bd8ea4] | 19 | |
---|
[daa56d0] | 20 | # Check whether we have a converter available |
---|
[99d1af6] | 21 | has_converter = True |
---|
| 22 | try: |
---|
[79492222] | 23 | from sas.data_util.nxsunit import Converter |
---|
[99d1af6] | 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() |
---|
[ef3445e2] | 69 | lines = buff.splitlines() |
---|
[470bf7e] | 70 | |
---|
[8bd8ea4] | 71 | x = numpy.zeros(0) |
---|
| 72 | y = numpy.zeros(0) |
---|
| 73 | dy = numpy.zeros(0) |
---|
[de1da34] | 74 | dx = numpy.zeros(0) |
---|
| 75 | |
---|
[ef3445e2] | 76 | #temp. space to sort data |
---|
[de1da34] | 77 | tx = numpy.zeros(0) |
---|
| 78 | ty = numpy.zeros(0) |
---|
| 79 | tdy = numpy.zeros(0) |
---|
| 80 | tdx = numpy.zeros(0) |
---|
| 81 | |
---|
| 82 | output = Data1D(x, y, dy=dy, dx=dx) |
---|
[8bd8ea4] | 83 | self.filename = output.filename = basename |
---|
[99d1af6] | 84 | |
---|
| 85 | data_conv_q = None |
---|
| 86 | data_conv_i = None |
---|
| 87 | |
---|
[ca10d8e] | 88 | if has_converter == True and output.x_unit != '1/A': |
---|
| 89 | data_conv_q = Converter('1/A') |
---|
[99d1af6] | 90 | # Test it |
---|
| 91 | data_conv_q(1.0, output.x_unit) |
---|
| 92 | |
---|
[ca10d8e] | 93 | if has_converter == True and output.y_unit != '1/cm': |
---|
| 94 | data_conv_i = Converter('1/cm') |
---|
[99d1af6] | 95 | # Test it |
---|
| 96 | data_conv_i(1.0, output.y_unit) |
---|
| 97 | |
---|
[8bd8ea4] | 98 | # The first good line of data will define whether |
---|
| 99 | # we have 2-column or 3-column ascii |
---|
[de1da34] | 100 | has_error_dx = None |
---|
| 101 | has_error_dy = None |
---|
[8bd8ea4] | 102 | |
---|
[892f246] | 103 | #Initialize counters for data lines and header lines. |
---|
[7d6351e] | 104 | is_data = False # Has more than 5 lines |
---|
[a7a5886] | 105 | # More than "5" lines of data is considered as actual |
---|
| 106 | # data unless that is the only data |
---|
[7d6351e] | 107 | mum_data_lines = 5 |
---|
[a7a5886] | 108 | # To count # of current data candidate lines |
---|
[7d6351e] | 109 | i = -1 |
---|
| 110 | # To count total # of previous data candidate lines |
---|
| 111 | i1 = -1 |
---|
| 112 | # To count # of header lines |
---|
[a7a5886] | 113 | j = -1 |
---|
[7d6351e] | 114 | # Helps to count # of header lines |
---|
[a7a5886] | 115 | j1 = -1 |
---|
[7d6351e] | 116 | #minimum required number of columns of data; ( <= 4). |
---|
| 117 | lentoks = 2 |
---|
[8bd8ea4] | 118 | for line in lines: |
---|
[86d13b5] | 119 | # Initial try for CSV (split on ,) |
---|
[ef9d209] | 120 | toks = line.split(',') |
---|
[86d13b5] | 121 | # Now try SCSV (split on ;) |
---|
| 122 | if len(toks) < 2: |
---|
| 123 | toks = line.split(';') |
---|
| 124 | # Now go for whitespace |
---|
[ef9d209] | 125 | if len(toks) < 2: |
---|
| 126 | toks = line.split() |
---|
[8bd8ea4] | 127 | try: |
---|
[5f2d3c78] | 128 | #Make sure that all columns are numbers. |
---|
| 129 | for colnum in range(len(toks)): |
---|
| 130 | float(toks[colnum]) |
---|
| 131 | |
---|
[8bd8ea4] | 132 | _x = float(toks[0]) |
---|
| 133 | _y = float(toks[1]) |
---|
| 134 | |
---|
[892f246] | 135 | #Reset the header line counters |
---|
| 136 | if j == j1: |
---|
| 137 | j = 0 |
---|
| 138 | j1 = 0 |
---|
| 139 | |
---|
| 140 | if i > 1: |
---|
| 141 | is_data = True |
---|
[d508be9] | 142 | |
---|
[99d1af6] | 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: |
---|
[7d6351e] | 147 | _y = data_conv_i(_y, units=output.y_unit) |
---|
[99d1af6] | 148 | |
---|
[8bd8ea4] | 149 | # If we have an extra token, check |
---|
| 150 | # whether it can be interpreted as a |
---|
| 151 | # third column. |
---|
| 152 | _dy = None |
---|
[a7a5886] | 153 | if len(toks) > 2: |
---|
[8bd8ea4] | 154 | try: |
---|
| 155 | _dy = float(toks[2]) |
---|
[99d1af6] | 156 | |
---|
| 157 | if data_conv_i is not None: |
---|
| 158 | _dy = data_conv_i(_dy, units=output.y_unit) |
---|
| 159 | |
---|
[8bd8ea4] | 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. |
---|
[de1da34] | 166 | if has_error_dy == None: |
---|
| 167 | has_error_dy = False if _dy == None else True |
---|
| 168 | |
---|
| 169 | #Check for dx |
---|
| 170 | _dx = None |
---|
[a7a5886] | 171 | if len(toks) > 3: |
---|
[de1da34] | 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 |
---|
[892f246] | 186 | |
---|
[7d6351e] | 187 | #After talked with PB, we decided to take care of only |
---|
[a7a5886] | 188 | # 4 columns of data for now. |
---|
[d508be9] | 189 | #number of columns in the current line |
---|
[7d6351e] | 190 | #To remember the # of columns in the current |
---|
[a7a5886] | 191 | #line of data |
---|
[0e5e586] | 192 | new_lentoks = len(toks) |
---|
[d508be9] | 193 | |
---|
[7d6351e] | 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: |
---|
[272b107] | 197 | if is_data == True: |
---|
| 198 | break |
---|
| 199 | else: |
---|
[d508be9] | 200 | i = -1 |
---|
| 201 | i1 = 0 |
---|
| 202 | j = -1 |
---|
| 203 | j1 = -1 |
---|
| 204 | |
---|
[a7a5886] | 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: |
---|
[892f246] | 209 | try: |
---|
[a7a5886] | 210 | x = numpy.zeros(0) |
---|
| 211 | y = numpy.zeros(0) |
---|
[892f246] | 212 | except: |
---|
| 213 | pass |
---|
| 214 | |
---|
[ef3445e2] | 215 | x = numpy.append(x, _x) |
---|
[7d6351e] | 216 | y = numpy.append(y, _y) |
---|
[892f246] | 217 | |
---|
[de1da34] | 218 | if has_error_dy == True: |
---|
[a7a5886] | 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: |
---|
[892f246] | 223 | try: |
---|
[7d6351e] | 224 | dy = numpy.zeros(0) |
---|
[892f246] | 225 | except: |
---|
[7d6351e] | 226 | pass |
---|
[8bd8ea4] | 227 | dy = numpy.append(dy, _dy) |
---|
[892f246] | 228 | |
---|
[de1da34] | 229 | if has_error_dx == True: |
---|
[a7a5886] | 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: |
---|
[892f246] | 234 | try: |
---|
[7d6351e] | 235 | dx = numpy.zeros(0) |
---|
[892f246] | 236 | except: |
---|
[7d6351e] | 237 | pass |
---|
[de1da34] | 238 | dx = numpy.append(dx, _dx) |
---|
| 239 | |
---|
| 240 | #Same for temp. |
---|
[a7a5886] | 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\ |
---|
| 244 | is_data == False: |
---|
[892f246] | 245 | try: |
---|
| 246 | tx = numpy.zeros(0) |
---|
| 247 | ty = numpy.zeros(0) |
---|
| 248 | except: |
---|
[7d6351e] | 249 | pass |
---|
[892f246] | 250 | |
---|
[7d6351e] | 251 | tx = numpy.append(tx, _x) |
---|
| 252 | ty = numpy.append(ty, _y) |
---|
[892f246] | 253 | |
---|
[de1da34] | 254 | if has_error_dy == True: |
---|
[a7a5886] | 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: |
---|
[892f246] | 259 | try: |
---|
| 260 | tdy = numpy.zeros(0) |
---|
| 261 | except: |
---|
[7d6351e] | 262 | pass |
---|
[de1da34] | 263 | tdy = numpy.append(tdy, _dy) |
---|
| 264 | if has_error_dx == True: |
---|
[a7a5886] | 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: |
---|
[892f246] | 269 | try: |
---|
| 270 | tdx = numpy.zeros(0) |
---|
| 271 | except: |
---|
[7d6351e] | 272 | pass |
---|
[de1da34] | 273 | tdx = numpy.append(tdx, _dx) |
---|
[d508be9] | 274 | |
---|
| 275 | #reset i1 and flag lentoks for the next |
---|
[a7a5886] | 276 | if lentoks < new_lentoks: |
---|
[d508be9] | 277 | if is_data == False: |
---|
[7d6351e] | 278 | i1 = -1 |
---|
[a7a5886] | 279 | #To remember the # of columns on the current line |
---|
| 280 | # for the next line of data |
---|
[0e5e586] | 281 | lentoks = len(toks) |
---|
[8bd8ea4] | 282 | |
---|
[7d6351e] | 283 | #Reset # of header lines and counts # |
---|
| 284 | # of data candidate lines |
---|
[a7a5886] | 285 | if j == 0 and j1 == 0: |
---|
[7d6351e] | 286 | i1 = i + 1 |
---|
[a7a5886] | 287 | i += 1 |
---|
[8bd8ea4] | 288 | except: |
---|
[892f246] | 289 | # It is data and meet non - number, then stop reading |
---|
| 290 | if is_data == True: |
---|
[7d6351e] | 291 | break |
---|
[d508be9] | 292 | lentoks = 2 |
---|
[7d6351e] | 293 | #Counting # of header lines |
---|
[a7a5886] | 294 | j += 1 |
---|
| 295 | if j == j1 + 1: |
---|
[7d6351e] | 296 | j1 = j |
---|
| 297 | else: |
---|
[892f246] | 298 | j = -1 |
---|
| 299 | #Reset # of lines of data candidates |
---|
| 300 | i = -1 |
---|
| 301 | |
---|
[7d6351e] | 302 | # Couldn't parse this line, skip it |
---|
[8bd8ea4] | 303 | pass |
---|
[892f246] | 304 | |
---|
[7d6351e] | 305 | input_f.close() |
---|
[8bd8ea4] | 306 | # Sanity check |
---|
[de1da34] | 307 | if has_error_dy == True and not len(y) == len(dy): |
---|
[a7a5886] | 308 | msg = "ascii_reader: y and dy have different length" |
---|
| 309 | raise RuntimeError, msg |
---|
[de1da34] | 310 | if has_error_dx == True and not len(x) == len(dx): |
---|
[a7a5886] | 311 | msg = "ascii_reader: y and dy have different length" |
---|
| 312 | raise RuntimeError, msg |
---|
[8bd8ea4] | 313 | # If the data length is zero, consider this as |
---|
| 314 | # though we were not able to read the file. |
---|
[a7a5886] | 315 | if len(x) == 0: |
---|
[daa56d0] | 316 | raise RuntimeError, "ascii_reader: could not load file" |
---|
[de1da34] | 317 | |
---|
[a7a5886] | 318 | #Let's re-order the data to make cal. |
---|
| 319 | # curve look better some cases |
---|
| 320 | ind = numpy.lexsort((ty, tx)) |
---|
[de1da34] | 321 | for i in ind: |
---|
| 322 | x[i] = tx[ind[i]] |
---|
| 323 | y[i] = ty[ind[i]] |
---|
| 324 | if has_error_dy == True: |
---|
| 325 | dy[i] = tdy[ind[i]] |
---|
| 326 | if has_error_dx == True: |
---|
| 327 | dx[i] = tdx[ind[i]] |
---|
[7d6351e] | 328 | # Zeros in dx, dy |
---|
[da96629] | 329 | if has_error_dx: |
---|
[7d6351e] | 330 | dx[dx == 0] = _ZERO |
---|
[da96629] | 331 | if has_error_dy: |
---|
[7d6351e] | 332 | dy[dy == 0] = _ZERO |
---|
| 333 | #Data |
---|
| 334 | output.x = x[x != 0] |
---|
| 335 | output.y = y[x != 0] |
---|
| 336 | output.dy = dy[x != 0] if has_error_dy == True\ |
---|
| 337 | else numpy.zeros(len(output.y)) |
---|
| 338 | output.dx = dx[x != 0] if has_error_dx == True\ |
---|
| 339 | else numpy.zeros(len(output.x)) |
---|
[fca90f82] | 340 | |
---|
[99d1af6] | 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: |
---|
[0e2aa40] | 346 | output.yaxis("\\rm{Intensity}", output.y_unit) |
---|
[99d1af6] | 347 | else: |
---|
[7d6351e] | 348 | output.yaxis("\\rm{Intensity}", "cm^{-1}") |
---|
[fe78c7b] | 349 | |
---|
| 350 | # Store loading process information |
---|
[7d6351e] | 351 | output.meta_data['loader'] = self.type_name |
---|
[83b81b8] | 352 | if len(output.x) < 1: |
---|
| 353 | raise RuntimeError, "%s is empty" % path |
---|
[8bd8ea4] | 354 | return output |
---|
[892f246] | 355 | |
---|
[8bd8ea4] | 356 | else: |
---|
| 357 | raise RuntimeError, "%s is not a file" % path |
---|
| 358 | return None |
---|