| 1 | #!/usr/bin/python -tt
|
|---|
| 2 | #
|
|---|
| 3 | # Werner Lustermann, Dominik Neise
|
|---|
| 4 | # ETH Zurich, TU Dortmund
|
|---|
| 5 | #
|
|---|
| 6 | from ctypes import *
|
|---|
| 7 | import numpy as np
|
|---|
| 8 | import pprint # for SlowData
|
|---|
| 9 | from scipy import signal
|
|---|
| 10 |
|
|---|
| 11 | # get the ROOT stuff + my shared libs
|
|---|
| 12 | from ROOT import gSystem
|
|---|
| 13 | # factfits_h.so is made from factfits.h and is used to access the data
|
|---|
| 14 | # make sure the location of factfits_h.so is in LD_LIBRARY_PATH.
|
|---|
| 15 | # having it in PYTHONPATH is *not* sufficient
|
|---|
| 16 | gSystem.Load('factfits_h.so')
|
|---|
| 17 | gSystem.Load('calfactfits_h.so')
|
|---|
| 18 | from ROOT import *
|
|---|
| 19 |
|
|---|
| 20 | class RawDataFeeder( object ):
|
|---|
| 21 | """ Wrapper class for RawData class
|
|---|
| 22 | capable of iterating over multiple RawData Files
|
|---|
| 23 | """
|
|---|
| 24 |
|
|---|
| 25 | def __init__(self, filelist):
|
|---|
| 26 | """ *filelist* list of files to iterate over
|
|---|
| 27 | the list should contain tuples, or sublists of two filenames
|
|---|
| 28 | the first should be a data file (\*.fits.gz)
|
|---|
| 29 | the second should be an amplitude calibration file(\*.drs.fits.gz)
|
|---|
| 30 | """
|
|---|
| 31 | # sanity check for input
|
|---|
| 32 | if type(filelist) != type(list()):
|
|---|
| 33 | raise TypeError('filelist should be a list')
|
|---|
| 34 | for entry in filelist:
|
|---|
| 35 | if len(entry) != 2:
|
|---|
| 36 | raise TypeError('the entries of filelist should have length == 2')
|
|---|
| 37 | for path in entry:
|
|---|
| 38 | if type(path) != type(str()):
|
|---|
| 39 | raise TypeError('the entries of filelist should be path, i.e. of type str()')
|
|---|
| 40 | #todo check if 'path' is a valid path
|
|---|
| 41 | # else: throw an Exception, or Warning?
|
|---|
| 42 |
|
|---|
| 43 | self.filelist = filelist
|
|---|
| 44 | self._current_RawData = RawData(filelist[0][0], filelist[0][1], return_dict=True)
|
|---|
| 45 | del filelist[0]
|
|---|
| 46 |
|
|---|
| 47 | def __iter__(self):
|
|---|
| 48 | return self
|
|---|
| 49 |
|
|---|
| 50 | def next():
|
|---|
| 51 | """ Method being called by the iterator.
|
|---|
| 52 | Since the RawData Objects are simply looped over, the event_id from the
|
|---|
| 53 | RawData object will not be unique.
|
|---|
| 54 | Each RawData obejct will start with event_id = 1 as usual.
|
|---|
| 55 | """
|
|---|
| 56 | try:
|
|---|
| 57 | return self._current_RawData.next()
|
|---|
| 58 | except StopIteration:
|
|---|
| 59 | # current_RawData was completely processed
|
|---|
| 60 | # delete it (I hope this calls the destructor of the fits file and/or closes it)
|
|---|
| 61 | del self._current_RawData
|
|---|
| 62 | # and remake it, if possible
|
|---|
| 63 | if len(self.filelist) > 0:
|
|---|
| 64 | self._current_RawData = RawData(filelist[0][0], filelist[0][1], return_dict=True)
|
|---|
| 65 | del filelist[0]
|
|---|
| 66 | else:
|
|---|
| 67 | raise
|
|---|
| 68 |
|
|---|
| 69 |
|
|---|
| 70 |
|
|---|
| 71 | class RawData( object ):
|
|---|
| 72 | """ raw data access and calibration
|
|---|
| 73 |
|
|---|
| 74 | - open raw data file and drs calibration file
|
|---|
| 75 | - performs amplitude calibration
|
|---|
| 76 | - performs baseline substraction if wanted
|
|---|
| 77 | - provides all data in an array:
|
|---|
| 78 | row = number of pixel
|
|---|
| 79 | col = length of region of interest
|
|---|
| 80 |
|
|---|
| 81 | """
|
|---|
| 82 |
|
|---|
| 83 |
|
|---|
| 84 | def __init__(self, data_file_name, calib_file_name,
|
|---|
| 85 | user_action_calib=lambda acal_data, data, blm, tom, gm, scells, nroi: None,
|
|---|
| 86 | baseline_file_name='',
|
|---|
| 87 | return_dict = True,
|
|---|
| 88 | do_calibration = True,
|
|---|
| 89 | use_CalFactFits = True):
|
|---|
| 90 | """ initialize object
|
|---|
| 91 |
|
|---|
| 92 | open data file and calibration data file
|
|---|
| 93 | get basic information about the data in data_file_name
|
|---|
| 94 | allocate buffers for data access
|
|---|
| 95 |
|
|---|
| 96 | data_file_name : fits or fits.gz file of the data including the path
|
|---|
| 97 | calib_file_name : fits or fits.gz file containing DRS calibration data
|
|---|
| 98 | baseline_file_name : npy file containing the baseline values
|
|---|
| 99 | """
|
|---|
| 100 | self.__module__='pyfact'
|
|---|
| 101 | # manual implementation of default value, but I need to find out
|
|---|
| 102 | # if the user of this class is aware of the new option
|
|---|
| 103 | if return_dict == False:
|
|---|
| 104 | print 'DEPRECATION WARNING:'
|
|---|
| 105 | print 'you are using RawData in a way, which is nor supported anymore.'
|
|---|
| 106 | print ' Please set: return_dict = True, in the __init__ call'
|
|---|
| 107 | self.return_dict = return_dict
|
|---|
| 108 | self.use_CalFactFits = use_CalFactFits
|
|---|
| 109 |
|
|---|
| 110 | self.do_calibration = do_calibration
|
|---|
| 111 |
|
|---|
| 112 | self.data_file_name = data_file_name
|
|---|
| 113 | self.calib_file_name = calib_file_name
|
|---|
| 114 | self.baseline_file_name = baseline_file_name
|
|---|
| 115 |
|
|---|
| 116 | self.user_action_calib = user_action_calib
|
|---|
| 117 |
|
|---|
| 118 | # baseline correction: True / False
|
|---|
| 119 | if len(baseline_file_name) == 0:
|
|---|
| 120 | self.correct_baseline = False
|
|---|
| 121 | else:
|
|---|
| 122 | self.correct_baseline = True
|
|---|
| 123 |
|
|---|
| 124 | # access data file
|
|---|
| 125 | if use_CalFactFits:
|
|---|
| 126 | try:
|
|---|
| 127 | data_file = CalFactFits(data_file_name, calib_file_name)
|
|---|
| 128 | except IOError:
|
|---|
| 129 | print 'problem accessing data file: ', data_file_name
|
|---|
| 130 | raise # stop ! no data
|
|---|
| 131 |
|
|---|
| 132 | self.data_file = data_file
|
|---|
| 133 | self.data = np.empty( data_file.npix * data_file.nroi, np.float64)
|
|---|
| 134 | data_file.SetNpcaldataPtr(self.data)
|
|---|
| 135 | self.data = self.data.reshape( data_file.npix, data_file.nroi )
|
|---|
| 136 | self.acal_data = self.data
|
|---|
| 137 |
|
|---|
| 138 | self.nroi = data_file.nroi
|
|---|
| 139 | self.npix = data_file.npix
|
|---|
| 140 | self.nevents = data_file.nevents
|
|---|
| 141 |
|
|---|
| 142 | # Data per event
|
|---|
| 143 | self.event_id = None
|
|---|
| 144 | self.trigger_type = None
|
|---|
| 145 | #self.start_cells = None
|
|---|
| 146 | #self.board_times = None
|
|---|
| 147 | self.start_cells = np.zeros( self.npix, np.int16 )
|
|---|
| 148 | self.board_times = np.zeros( 40, np.int32 )
|
|---|
| 149 |
|
|---|
| 150 | # data_file is a CalFactFits object
|
|---|
| 151 | # data_file.datafile is one of the two FactFits objects hold by a CalFactFits.
|
|---|
| 152 | # sorry for the strange naming ..
|
|---|
| 153 | data_file.datafile.SetPtrAddress('StartCellData', self.start_cells)
|
|---|
| 154 | data_file.datafile.SetPtrAddress('BoardTime', self.board_times)
|
|---|
| 155 |
|
|---|
| 156 | else:
|
|---|
| 157 | try:
|
|---|
| 158 | data_file = FactFits(self.data_file_name)
|
|---|
| 159 | except IOError:
|
|---|
| 160 | print 'problem accessing data file: ', data_file_name
|
|---|
| 161 | raise # stop ! no data
|
|---|
| 162 |
|
|---|
| 163 | self.data_file = data_file
|
|---|
| 164 |
|
|---|
| 165 | # get basic information about the data file
|
|---|
| 166 | #: region of interest (number of DRS slices read)
|
|---|
| 167 | self.nroi = data_file.GetUInt('NROI')
|
|---|
| 168 | #: number of pixels (should be 1440)
|
|---|
| 169 | self.npix = data_file.GetUInt('NPIX')
|
|---|
| 170 | #: number of events in the data run
|
|---|
| 171 | self.nevents = data_file.GetNumRows()
|
|---|
| 172 |
|
|---|
| 173 | # allocate the data memories
|
|---|
| 174 | self.event_id = c_ulong()
|
|---|
| 175 | self.trigger_type = c_ushort()
|
|---|
| 176 | #: 1D array with raw data
|
|---|
| 177 | self.data = np.zeros( self.npix * self.nroi, np.int16 ).reshape(self.npix ,self.nroi)
|
|---|
| 178 | #: slice where drs readout started
|
|---|
| 179 | self.start_cells = np.zeros( self.npix, np.int16 )
|
|---|
| 180 | #: time when the FAD was triggered, in some strange units...
|
|---|
| 181 | self.board_times = np.zeros( 40, np.int32 )
|
|---|
| 182 |
|
|---|
| 183 | # set the pointers to the data++
|
|---|
| 184 | data_file.SetPtrAddress('EventNum', self.event_id)
|
|---|
| 185 | data_file.SetPtrAddress('TriggerType', self.trigger_type)
|
|---|
| 186 | data_file.SetPtrAddress('StartCellData', self.start_cells)
|
|---|
| 187 | data_file.SetPtrAddress('Data', self.data)
|
|---|
| 188 | data_file.SetPtrAddress('BoardTime', self.board_times)
|
|---|
| 189 |
|
|---|
| 190 | # open the calibration file
|
|---|
| 191 | try:
|
|---|
| 192 | calib_file = FactFits(self.calib_file_name)
|
|---|
| 193 | except IOError:
|
|---|
| 194 | print 'problem accessing calibration file: ', calib_file_name
|
|---|
| 195 | raise
|
|---|
| 196 | #: drs calibration file
|
|---|
| 197 | self.calib_file = calib_file
|
|---|
| 198 |
|
|---|
| 199 | baseline_mean = calib_file.GetN('BaselineMean')
|
|---|
| 200 | gain_mean = calib_file.GetN('GainMean')
|
|---|
| 201 | trigger_offset_mean = calib_file.GetN('TriggerOffsetMean')
|
|---|
| 202 |
|
|---|
| 203 | self.Nblm = baseline_mean / self.npix
|
|---|
| 204 | self.Ngm = gain_mean / self.npix
|
|---|
| 205 | self.Ntom = trigger_offset_mean / self.npix
|
|---|
| 206 |
|
|---|
| 207 | self.blm = np.zeros(baseline_mean, np.float32).reshape(self.npix , self.Nblm)
|
|---|
| 208 | self.gm = np.zeros(gain_mean, np.float32).reshape(self.npix , self.Ngm)
|
|---|
| 209 | self.tom = np.zeros(trigger_offset_mean, np.float32).reshape(self.npix , self.Ntom)
|
|---|
| 210 |
|
|---|
| 211 | calib_file.SetPtrAddress('BaselineMean', self.blm)
|
|---|
| 212 | calib_file.SetPtrAddress('GainMean', self.gm)
|
|---|
| 213 | calib_file.SetPtrAddress('TriggerOffsetMean', self.tom)
|
|---|
| 214 | calib_file.GetRow(0)
|
|---|
| 215 |
|
|---|
| 216 | # make calibration constants double, so we never need to roll
|
|---|
| 217 | self.blm = np.hstack((self.blm, self.blm))
|
|---|
| 218 | self.gm = np.hstack((self.gm, self.gm))
|
|---|
| 219 | self.tom = np.hstack((self.tom, self.tom))
|
|---|
| 220 |
|
|---|
| 221 | self.v_bsl = np.zeros(self.npix) # array of baseline values (all ZERO)
|
|---|
| 222 |
|
|---|
| 223 | def __iter__(self):
|
|---|
| 224 | """ iterator """
|
|---|
| 225 | return self
|
|---|
| 226 |
|
|---|
| 227 | def next(self):
|
|---|
| 228 | """ used by __iter__ """
|
|---|
| 229 | if self.use_CalFactFits:
|
|---|
| 230 | if self.data_file.GetCalEvent() == False:
|
|---|
| 231 | raise StopIteration
|
|---|
| 232 | else:
|
|---|
| 233 | self.event_id = self.data_file.event_id
|
|---|
| 234 | self.trigger_type = self.data_file.event_triggertype
|
|---|
| 235 | #self.start_cells = self.data_file.event_offset
|
|---|
| 236 | #self.board_times = self.data_file.event_boardtimes
|
|---|
| 237 | #self.acal_data = self.data.copy().reshape(self.data_file.npix, self.data_file.nroi)
|
|---|
| 238 | else:
|
|---|
| 239 | if self.data_file.GetNextRow() == False:
|
|---|
| 240 | raise StopIteration
|
|---|
| 241 | else:
|
|---|
| 242 | if self.do_calibration == True:
|
|---|
| 243 | self.calibrate_drs_amplitude()
|
|---|
| 244 |
|
|---|
| 245 | #print 'nevents = ', self.nevents, 'event_id = ', self.event_id.value
|
|---|
| 246 | if self.return_dict:
|
|---|
| 247 | return self.__dict__
|
|---|
| 248 | else:
|
|---|
| 249 | return self.acal_data, self.start_cells, self.trigger_type.value
|
|---|
| 250 |
|
|---|
| 251 | def next_event(self):
|
|---|
| 252 | """ load the next event from disk and calibrate it
|
|---|
| 253 | """
|
|---|
| 254 | if self.use_CalFactFits:
|
|---|
| 255 | self.data_file.GetCalEvent()
|
|---|
| 256 | else:
|
|---|
| 257 | self.data_file.GetNextRow()
|
|---|
| 258 | self.calibrate_drs_amplitude()
|
|---|
| 259 |
|
|---|
| 260 | def calibrate_drs_amplitude(self):
|
|---|
| 261 | """ perform the drs amplitude calibration of the event data
|
|---|
| 262 |
|
|---|
| 263 | """
|
|---|
| 264 | # shortcuts
|
|---|
| 265 | blm = self.blm
|
|---|
| 266 | gm = self.gm
|
|---|
| 267 | tom = self.tom
|
|---|
| 268 |
|
|---|
| 269 | to_mV = 2000./4096.
|
|---|
| 270 | #: 2D array with amplitude calibrated dat in mV
|
|---|
| 271 | acal_data = self.data * to_mV # convert ADC counts to mV
|
|---|
| 272 |
|
|---|
| 273 |
|
|---|
| 274 | for pixel in range( self.npix ):
|
|---|
| 275 | #shortcuts
|
|---|
| 276 | sc = self.start_cells[pixel]
|
|---|
| 277 | roi = self.nroi
|
|---|
| 278 | # rotate the pixel baseline mean to the Data startCell
|
|---|
| 279 | acal_data[pixel,:] -= blm[pixel,sc:sc+roi]
|
|---|
| 280 | # the 'trigger offset mean' does not need to be rolled
|
|---|
| 281 | # on the contrary, it seems there is an offset in the DRS data,
|
|---|
| 282 | # which is related to its distance to the startCell, not to its
|
|---|
| 283 | # distance to the beginning of the physical pipeline in the DRS chip
|
|---|
| 284 | acal_data[pixel,:] -= tom[pixel,0:roi]
|
|---|
| 285 | # rotate the pixel gain mean to the Data startCell
|
|---|
| 286 | acal_data[pixel,:] /= gm[pixel,sc:sc+roi]
|
|---|
| 287 |
|
|---|
| 288 |
|
|---|
| 289 | self.acal_data = acal_data * 1907.35
|
|---|
| 290 |
|
|---|
| 291 | self.user_action_calib( self.acal_data,
|
|---|
| 292 | np.reshape(self.data, (self.npix, self.nroi) ), blm, tom, gm, self.start_cells, self.nroi)
|
|---|
| 293 |
|
|---|
| 294 |
|
|---|
| 295 | def baseline_read_values(self, file, bsl_hist='bsl_sum/hplt_mean'):
|
|---|
| 296 | """
|
|---|
| 297 |
|
|---|
| 298 | open ROOT file with baseline histogram and read baseline values
|
|---|
| 299 | file name of the root file
|
|---|
| 300 | bsl_hist path to the histogram containing the basline values
|
|---|
| 301 |
|
|---|
| 302 | """
|
|---|
| 303 |
|
|---|
| 304 | try:
|
|---|
| 305 | f = TFile(file)
|
|---|
| 306 | except:
|
|---|
| 307 | print 'Baseline data file could not be read: ', file
|
|---|
| 308 | return
|
|---|
| 309 |
|
|---|
| 310 | h = f.Get(bsl_hist)
|
|---|
| 311 |
|
|---|
| 312 | for i in range(self.npix):
|
|---|
| 313 | self.v_bsl[i] = h.GetBinContent(i+1)
|
|---|
| 314 |
|
|---|
| 315 | f.Close()
|
|---|
| 316 |
|
|---|
| 317 | def baseline_correct(self):
|
|---|
| 318 | """ subtract baseline from the data
|
|---|
| 319 |
|
|---|
| 320 | """
|
|---|
| 321 |
|
|---|
| 322 | for pixel in range(self.npix):
|
|---|
| 323 | self.acal_data[pixel,:] -= self.v_bsl[pixel]
|
|---|
| 324 |
|
|---|
| 325 | def info(self):
|
|---|
| 326 | """ print run information
|
|---|
| 327 |
|
|---|
| 328 | """
|
|---|
| 329 | print 'data file: ', self.data_file_name
|
|---|
| 330 | print 'calib file: ', self.calib_file_name
|
|---|
| 331 | print '... we need more information printed here ... '
|
|---|
| 332 |
|
|---|
| 333 | # -----------------------------------------------------------------------------
|
|---|
| 334 | class RawDataFake( object ):
|
|---|
| 335 | """ raw data FAKE access similar to real RawData access
|
|---|
| 336 | """
|
|---|
| 337 |
|
|---|
| 338 |
|
|---|
| 339 | def __init__(self, data_file_name, calib_file_name,
|
|---|
| 340 | user_action_calib=lambda acal_data, data, blm, tom, gm, scells, nroi: None,
|
|---|
| 341 | baseline_file_name=''):
|
|---|
| 342 | self.__module__='pyfact'
|
|---|
| 343 |
|
|---|
| 344 | self.nroi = 300
|
|---|
| 345 | self.npix = 9
|
|---|
| 346 | self.nevents = 1000
|
|---|
| 347 |
|
|---|
| 348 | self.simulator = None
|
|---|
| 349 |
|
|---|
| 350 | self.time = np.ones(1024) * 0.5
|
|---|
| 351 |
|
|---|
| 352 |
|
|---|
| 353 | self.event_id = c_ulong(0)
|
|---|
| 354 | self.trigger_type = c_ushort(4)
|
|---|
| 355 | self.data = np.zeros( self.npix * self.nroi, np.int16 ).reshape(self.npix ,self.nroi)
|
|---|
| 356 | self.start_cells = np.zeros( self.npix, np.int16 )
|
|---|
| 357 | self.board_times = np.zeros( 40, np.int32 )
|
|---|
| 358 | def __iter__(self):
|
|---|
| 359 | """ iterator """
|
|---|
| 360 | return self
|
|---|
| 361 |
|
|---|
| 362 | def next(self):
|
|---|
| 363 | """ used by __iter__ """
|
|---|
| 364 | self.event_id = c_ulong(self.event_id.value + 1)
|
|---|
| 365 | self.board_times = self.board_times + 42
|
|---|
| 366 |
|
|---|
| 367 | if self.event_id.value >= self.nevents:
|
|---|
| 368 | raise StopIteration
|
|---|
| 369 | else:
|
|---|
| 370 | self._make_event_data()
|
|---|
| 371 |
|
|---|
| 372 | return self.__dict__
|
|---|
| 373 |
|
|---|
| 374 | def _make_event_data(self):
|
|---|
| 375 | sample_times = self.time.cumsum() - time[0]
|
|---|
| 376 |
|
|---|
| 377 | # random start cell
|
|---|
| 378 | self.start_cells = np.ones( self.npix, np.int16 ) * np.random.randint(0,1024)
|
|---|
| 379 |
|
|---|
| 380 | starttime = self.start_cells[0]
|
|---|
| 381 |
|
|---|
| 382 | signal = self._std_sinus_simu(sample_times, starttime)
|
|---|
| 383 |
|
|---|
| 384 | data = np.vstack( (signal,signal) )
|
|---|
| 385 | for i in range(8):
|
|---|
| 386 | data = np.vstack( (data,signal) )
|
|---|
| 387 |
|
|---|
| 388 | self.data = data
|
|---|
| 389 |
|
|---|
| 390 | def _std_sinus_simu(self, times, starttime):
|
|---|
| 391 | period = 10 # in ns
|
|---|
| 392 |
|
|---|
| 393 | # give a jitter on starttime
|
|---|
| 394 | starttime = np.random.normal(startime, 0.05)
|
|---|
| 395 |
|
|---|
| 396 | phase = 0.0
|
|---|
| 397 | signal = 10 * np.sin(times * 2*np.pi/period + starttime + phase)
|
|---|
| 398 |
|
|---|
| 399 | # add some noise
|
|---|
| 400 | noise = np.random.normal(0.0, 0.5, signal.shape)
|
|---|
| 401 | signal += noise
|
|---|
| 402 | return signal
|
|---|
| 403 |
|
|---|
| 404 | def info(self):
|
|---|
| 405 | """ print run information
|
|---|
| 406 |
|
|---|
| 407 | """
|
|---|
| 408 |
|
|---|
| 409 | print 'data file: ', data_file_name
|
|---|
| 410 | print 'calib file: ', calib_file_name
|
|---|
| 411 | print 'calibration file'
|
|---|
| 412 | print 'N baseline_mean: ', self.Nblm
|
|---|
| 413 | print 'N gain mean: ', self.Ngm
|
|---|
| 414 | print 'N TriggeroffsetMean: ', self.Ntom
|
|---|
| 415 |
|
|---|
| 416 | # -----------------------------------------------------------------------------
|
|---|
| 417 |
|
|---|
| 418 | class SlowData( FactFits ):
|
|---|
| 419 | """ -Fact SlowData File-
|
|---|
| 420 | A Python wrapper for the fits-class implemented in pyfits.h
|
|---|
| 421 | provides easy access to the fits file meta data.
|
|---|
| 422 | * dictionary of file metadata - self.meta
|
|---|
| 423 | * dict of table metadata - self.columns
|
|---|
| 424 | * variable table column access, thus possibly increased speed while looping
|
|---|
| 425 | """
|
|---|
| 426 | def __init__(self, path):
|
|---|
| 427 | """ creates meta and columns dictionaries
|
|---|
| 428 | """
|
|---|
| 429 | self.path = path
|
|---|
| 430 | try:
|
|---|
| 431 | FactFits.__init__(self,path)
|
|---|
| 432 | except IOError:
|
|---|
| 433 | print 'problem accessing data file: ', data_file_name
|
|---|
| 434 | raise # stop ! no data
|
|---|
| 435 |
|
|---|
| 436 | self.meta = self._make_meta_dict()
|
|---|
| 437 | self.columns = self._make_columns_dict()
|
|---|
| 438 |
|
|---|
| 439 | self.treat_meta_dict()
|
|---|
| 440 |
|
|---|
| 441 |
|
|---|
| 442 | # list of columns, which are already registered
|
|---|
| 443 | # see method register()
|
|---|
| 444 | self._registered_cols = []
|
|---|
| 445 | # dict of column data, this is used, in order to be able to remove
|
|---|
| 446 | # the ctypes of
|
|---|
| 447 | self._table_cols = {}
|
|---|
| 448 |
|
|---|
| 449 | # I need to count the rows, since the normal loop mechanism seems not to work.
|
|---|
| 450 | self._current_row = 0
|
|---|
| 451 |
|
|---|
| 452 | self.stacked_cols = {}
|
|---|
| 453 |
|
|---|
| 454 | def _make_meta_dict(self):
|
|---|
| 455 | """ This method retrieves meta information about the fits file and
|
|---|
| 456 | stores this information in a dict
|
|---|
| 457 | return: dict
|
|---|
| 458 | key: string - all capital letters
|
|---|
| 459 | value: tuple( numerical value, string comment)
|
|---|
| 460 | """
|
|---|
| 461 | # intermediate variables for file metadata dict generation
|
|---|
| 462 | keys=self.GetPy_KeyKeys()
|
|---|
| 463 | values=self.GetPy_KeyValues()
|
|---|
| 464 | comments=self.GetPy_KeyComments()
|
|---|
| 465 | types=self.GetPy_KeyTypes()
|
|---|
| 466 |
|
|---|
| 467 | if len(keys) != len(values):
|
|---|
| 468 | raise TypeError('len(keys)',len(keys),' != len(values)', len(values))
|
|---|
| 469 | if len(keys) != len(types):
|
|---|
| 470 | raise TypeError('len(keys)',len(keys),' != len(types)', len(types))
|
|---|
| 471 | if len(keys) != len(comments):
|
|---|
| 472 | raise TypeError('len(keys)',len(keys),' != len(comments)', len(comments))
|
|---|
| 473 |
|
|---|
| 474 | meta_dict = {}
|
|---|
| 475 | for i in range(len(keys)):
|
|---|
| 476 | type = types[i]
|
|---|
| 477 | if type == 'I':
|
|---|
| 478 | value = int(values[i])
|
|---|
| 479 | elif type == 'F':
|
|---|
| 480 | value = float(values[i])
|
|---|
| 481 | elif type == 'B':
|
|---|
| 482 | if values[i] == 'T':
|
|---|
| 483 | value = True
|
|---|
| 484 | elif values[i] == 'F':
|
|---|
| 485 | value = False
|
|---|
| 486 | else:
|
|---|
| 487 | raise TypeError("meta-type is 'B', but meta-value is neither 'T' nor 'F'. meta-value:",values[i])
|
|---|
| 488 | elif type == 'T':
|
|---|
| 489 | value = values[i]
|
|---|
| 490 | else:
|
|---|
| 491 | raise TypeError("unknown meta-type: known meta types are: I,F,B and T. meta-type:",type)
|
|---|
| 492 | meta_dict[keys[i]]=(value, comments[i])
|
|---|
| 493 | return meta_dict
|
|---|
| 494 |
|
|---|
| 495 |
|
|---|
| 496 | def _make_columns_dict(self):
|
|---|
| 497 | """ This method retrieves information about the columns
|
|---|
| 498 | stored inside the fits files internal binary table.
|
|---|
| 499 | returns: dict
|
|---|
| 500 | key: string column name -- all capital letters
|
|---|
| 501 | values: tuple(
|
|---|
| 502 | number of elements in table field - integer
|
|---|
| 503 | size of element in bytes -- this is not really interesting for any user
|
|---|
| 504 | might be ommited in future versions
|
|---|
| 505 | type - a single character code -- should be translated into
|
|---|
| 506 | a comrehensible word
|
|---|
| 507 | unit - string like 'mV' or 'ADC count'
|
|---|
| 508 | """
|
|---|
| 509 | # intermediate variables for file table-metadata dict generation
|
|---|
| 510 | keys=self.GetPy_ColumnKeys()
|
|---|
| 511 | #offsets=self.GetPy_ColumnOffsets() #not needed on python level...
|
|---|
| 512 | nums=self.GetPy_ColumnNums()
|
|---|
| 513 | sizes=self.GetPy_ColumnSizes()
|
|---|
| 514 | types=self.GetPy_ColumnTypes()
|
|---|
| 515 | units=self.GetPy_ColumnUnits()
|
|---|
| 516 |
|
|---|
| 517 | # zip the values
|
|---|
| 518 | values = zip(nums,sizes,types,units)
|
|---|
| 519 | # create the columns dictionary
|
|---|
| 520 | columns = dict(zip(keys ,values))
|
|---|
| 521 | return columns
|
|---|
| 522 |
|
|---|
| 523 | def stack(self, on=True):
|
|---|
| 524 | self.next()
|
|---|
| 525 | for col in self._registered_cols:
|
|---|
| 526 | if isinstance( self.dict[col], type(np.array('')) ):
|
|---|
| 527 | self.stacked_cols[col] = self.dict[col]
|
|---|
| 528 | else:
|
|---|
| 529 | # elif isinstance(self.dict[col], ctypes._SimpleCData):
|
|---|
| 530 | self.stacked_cols[col] = np.array(self.dict[col])
|
|---|
| 531 | # else:
|
|---|
| 532 | # raise TypeError("I don't know how to stack "+col+". It is of type: "+str(type(self.dict[col])))
|
|---|
| 533 |
|
|---|
| 534 | def register(self, input_str):
|
|---|
| 535 | columns = self.columns
|
|---|
| 536 | if input_str.lower() == 'all':
|
|---|
| 537 | for col in columns:
|
|---|
| 538 | self._register(col)
|
|---|
| 539 | else:
|
|---|
| 540 | #check if colname is in columns:
|
|---|
| 541 | if input_str not in columns:
|
|---|
| 542 | error_msg = 'colname:'+ input_str +' is not a column in the binary table.\n'
|
|---|
| 543 | error_msg+= 'possible colnames are\n'
|
|---|
| 544 | for key in columns:
|
|---|
| 545 | error_msg += key+'\n'
|
|---|
| 546 | raise KeyError(error_msg)
|
|---|
| 547 | else:
|
|---|
| 548 | self._register(input_str)
|
|---|
| 549 |
|
|---|
| 550 | # 'private' method, do not use
|
|---|
| 551 | def _register( self, colname):
|
|---|
| 552 | columns = self.columns
|
|---|
| 553 | local = None
|
|---|
| 554 |
|
|---|
| 555 | number_of_elements = int(columns[colname][0])
|
|---|
| 556 | size_of_elements_in_bytes = int(columns[colname][1])
|
|---|
| 557 | ctypecode_of_elements = columns[colname][2]
|
|---|
| 558 | physical_unit_of_elements = columns[colname][3]
|
|---|
| 559 |
|
|---|
| 560 | # snippet from the C++ source code, or header file to be precise:
|
|---|
| 561 | #case 'L': gLog << "bool(8)"; break;
|
|---|
| 562 | #case 'B': gLog << "byte(8)"; break;
|
|---|
| 563 | #case 'I': gLog << "short(16)"; break;
|
|---|
| 564 | #case 'J': gLog << "int(32)"; break;
|
|---|
| 565 | #case 'K': gLog << "int(64)"; break;
|
|---|
| 566 | #case 'E': gLog << "float(32)"; break;
|
|---|
| 567 | #case 'D': gLog << "double(64)"; break;
|
|---|
| 568 |
|
|---|
| 569 |
|
|---|
| 570 |
|
|---|
| 571 | # the fields inside the columns can either contain single numbers,
|
|---|
| 572 | # or whole arrays of numbers as well.
|
|---|
| 573 | # we treat single elements differently...
|
|---|
| 574 | if number_of_elements == 1:
|
|---|
| 575 | # allocate some memory for a single number according to its type
|
|---|
| 576 | if ctypecode_of_elements == 'J': # J is for a 4byte int, i.e. an unsigned long
|
|---|
| 577 | local = ctypes.c_ulong()
|
|---|
| 578 | un_c_type = long
|
|---|
| 579 | elif ctypecode_of_elements == 'I': # I is for a 2byte int, i.e. an unsinged int
|
|---|
| 580 | local = ctypes.c_ushort()
|
|---|
| 581 | un_c_type = int
|
|---|
| 582 | elif ctypecode_of_elements == 'B': # B is for a byte
|
|---|
| 583 | local = ctypes.c_ubyte()
|
|---|
| 584 | un_c_type = int
|
|---|
| 585 | elif ctypecode_of_elements == 'D':
|
|---|
| 586 | local = ctypes.c_double()
|
|---|
| 587 | un_c_type = float
|
|---|
| 588 | elif ctypecode_of_elements == 'E':
|
|---|
| 589 | local = ctypes.c_float()
|
|---|
| 590 | un_c_type = float
|
|---|
| 591 | elif ctypecode_of_elements == 'A':
|
|---|
| 592 | local = ctypes.c_uchar()
|
|---|
| 593 | un_c_type = chr
|
|---|
| 594 | elif ctypecode_of_elements == 'K':
|
|---|
| 595 | local = ctypes.c_ulonglong()
|
|---|
| 596 | un_c_type = long
|
|---|
| 597 | else:
|
|---|
| 598 | raise TypeError('unknown ctypecode_of_elements:',ctypecode_of_elements)
|
|---|
| 599 | else:
|
|---|
| 600 | if ctypecode_of_elements == 'B': # B is for a byte
|
|---|
| 601 | nptype = np.int8
|
|---|
| 602 | elif ctypecode_of_elements == 'A': # A is for a char .. but I don't know how to handle it
|
|---|
| 603 | nptype = np.int8
|
|---|
| 604 | elif ctypecode_of_elements == 'I': # I is for a 2byte int
|
|---|
| 605 | nptype = np.int16
|
|---|
| 606 | elif ctypecode_of_elements == 'J': # J is for a 4byte int
|
|---|
| 607 | nptype = np.int32
|
|---|
| 608 | elif ctypecode_of_elements == 'K': # B is for a byte
|
|---|
| 609 | nptype = np.int64
|
|---|
| 610 | elif ctypecode_of_elements == 'E': # B is for a byte
|
|---|
| 611 | nptype = np.float32
|
|---|
| 612 | elif ctypecode_of_elements == 'D': # B is for a byte
|
|---|
| 613 | nptype = np.float64
|
|---|
| 614 | else:
|
|---|
| 615 | raise TypeError('unknown ctypecode_of_elements:',ctypecode_of_elements)
|
|---|
| 616 | local = np.zeros( number_of_elements, nptype)
|
|---|
| 617 |
|
|---|
| 618 | # Set the Pointer Address
|
|---|
| 619 | self.SetPtrAddress(colname, local)
|
|---|
| 620 | self._table_cols[colname] = local
|
|---|
| 621 | if number_of_elements > 1:
|
|---|
| 622 | self.__dict__[colname] = local
|
|---|
| 623 | self.dict[colname] = local
|
|---|
| 624 | else:
|
|---|
| 625 | # remove any traces of ctypes:
|
|---|
| 626 | self.__dict__[colname] = local.value
|
|---|
| 627 | self.dict[colname] = local.value
|
|---|
| 628 | self._registered_cols.append(colname)
|
|---|
| 629 |
|
|---|
| 630 |
|
|---|
| 631 | def treat_meta_dict(self):
|
|---|
| 632 | """make 'interesting' meta information available like normal members.
|
|---|
| 633 | non interesting are:
|
|---|
| 634 | TFORM, TUNIT, and TTYPE
|
|---|
| 635 | since these are available via the columns dict.
|
|---|
| 636 | """
|
|---|
| 637 |
|
|---|
| 638 | self.number_of_rows = self.meta['NAXIS2'][0]
|
|---|
| 639 | self.number_of_columns = self.meta['TFIELDS'][0]
|
|---|
| 640 |
|
|---|
| 641 | # there are some information in the meta dict, which are alsways there:
|
|---|
| 642 | # there are regarded as not interesting:
|
|---|
| 643 | uninteresting_meta = {}
|
|---|
| 644 | uninteresting_meta['arraylike'] = {}
|
|---|
| 645 | uninteresting = ['NAXIS', 'NAXIS1', 'NAXIS2',
|
|---|
| 646 | 'TFIELDS',
|
|---|
| 647 | 'XTENSION','EXTNAME','EXTREL',
|
|---|
| 648 | 'BITPIX', 'PCOUNT', 'GCOUNT',
|
|---|
| 649 | 'ORIGIN',
|
|---|
| 650 | 'PACKAGE', 'COMPILED', 'CREATOR',
|
|---|
| 651 | 'TELESCOP','TIMESYS','TIMEUNIT','VERSION']
|
|---|
| 652 | for key in uninteresting:
|
|---|
| 653 | if key in self.meta:
|
|---|
| 654 | uninteresting_meta[key]=self.meta[key]
|
|---|
| 655 | del self.meta[key]
|
|---|
| 656 |
|
|---|
| 657 | # the table meta data contains
|
|---|
| 658 |
|
|---|
| 659 |
|
|---|
| 660 | # shortcut to access the meta dict. But this needs to
|
|---|
| 661 | # be cleaned up quickly!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
|
|---|
| 662 | meta = self.meta
|
|---|
| 663 |
|
|---|
| 664 | # loop over keys:
|
|---|
| 665 | # * try to find array-like keys
|
|---|
| 666 | arraylike = {}
|
|---|
| 667 | singlelike = []
|
|---|
| 668 | for key in self.meta:
|
|---|
| 669 | stripped = key.rstrip('1234567890')
|
|---|
| 670 | if stripped == key:
|
|---|
| 671 | singlelike.append(key)
|
|---|
| 672 | else:
|
|---|
| 673 | if stripped not in arraylike:
|
|---|
| 674 | arraylike[stripped] = 0
|
|---|
| 675 | else:
|
|---|
| 676 | arraylike[stripped] += 1
|
|---|
| 677 | newmeta = {}
|
|---|
| 678 | for key in singlelike:
|
|---|
| 679 | newmeta[key.lower()] = meta[key]
|
|---|
| 680 | for key in arraylike:
|
|---|
| 681 | uninteresting_meta['arraylike'][key.lower()] = []
|
|---|
| 682 | for i in range(arraylike[key]+1):
|
|---|
| 683 | if key+str(i) in meta:
|
|---|
| 684 | uninteresting_meta['arraylike'][key.lower()].append(meta[key+str(i)])
|
|---|
| 685 | self.ui_meta = uninteresting_meta
|
|---|
| 686 | # make newmeta self
|
|---|
| 687 | for key in newmeta:
|
|---|
| 688 | self.__dict__[key]=newmeta[key]
|
|---|
| 689 |
|
|---|
| 690 | dict = self.__dict__.copy()
|
|---|
| 691 | del dict['meta']
|
|---|
| 692 | del dict['ui_meta']
|
|---|
| 693 | self.dict = dict
|
|---|
| 694 |
|
|---|
| 695 | def __iter__(self):
|
|---|
| 696 | """ iterator """
|
|---|
| 697 | return self
|
|---|
| 698 |
|
|---|
| 699 | def next(self):
|
|---|
| 700 | """ used by __iter__ """
|
|---|
| 701 | # Here one might check, if looping makes any sense, and if not
|
|---|
| 702 | # one could stop looping or so...
|
|---|
| 703 | # like this:
|
|---|
| 704 | #
|
|---|
| 705 | # if len(self._registered_cols) == 0:
|
|---|
| 706 | # print 'warning: looping without any registered columns'
|
|---|
| 707 | if self._current_row < self.number_of_rows:
|
|---|
| 708 | if self.GetNextRow() == False:
|
|---|
| 709 | raise StopIteration
|
|---|
| 710 | for col in self._registered_cols:
|
|---|
| 711 | if isinstance(self._table_cols[col], ctypes._SimpleCData):
|
|---|
| 712 | self.__dict__[col] = self._table_cols[col].value
|
|---|
| 713 | self.dict[col] = self._table_cols[col].value
|
|---|
| 714 |
|
|---|
| 715 | for col in self.stacked_cols:
|
|---|
| 716 | if isinstance(self.dict[col], type(np.array(''))):
|
|---|
| 717 | self.stacked_cols[col] = np.vstack( (self.stacked_cols[col],self.dict[col]) )
|
|---|
| 718 | else:
|
|---|
| 719 | self.stacked_cols[col] = np.vstack( (self.stacked_cols[col],np.array(self.dict[col])) )
|
|---|
| 720 | self._current_row += 1
|
|---|
| 721 | else:
|
|---|
| 722 | raise StopIteration
|
|---|
| 723 | return self
|
|---|
| 724 |
|
|---|
| 725 | def show(self):
|
|---|
| 726 | pprint.pprint(self.dict)
|
|---|
| 727 |
|
|---|
| 728 |
|
|---|
| 729 |
|
|---|
| 730 |
|
|---|
| 731 | class fnames( object ):
|
|---|
| 732 | """ organize file names of a FACT data run
|
|---|
| 733 |
|
|---|
| 734 | """
|
|---|
| 735 |
|
|---|
| 736 | def __init__(self, specifier = ['012', '023', '2011', '11', '24'],
|
|---|
| 737 | rpath = '/scratch_nfs/res/bsl/',
|
|---|
| 738 | zipped = True):
|
|---|
| 739 | """
|
|---|
| 740 | specifier : list of strings defined as:
|
|---|
| 741 | [ 'DRS calibration file', 'Data file', 'YYYY', 'MM', 'DD']
|
|---|
| 742 |
|
|---|
| 743 | rpath : directory path for the results; YYYYMMDD will be appended to rpath
|
|---|
| 744 | zipped : use zipped (True) or unzipped (Data)
|
|---|
| 745 |
|
|---|
| 746 | """
|
|---|
| 747 |
|
|---|
| 748 | self.specifier = specifier
|
|---|
| 749 | self.rpath = rpath
|
|---|
| 750 | self.zipped = zipped
|
|---|
| 751 |
|
|---|
| 752 | self.make( self.specifier, self.rpath, self.zipped )
|
|---|
| 753 |
|
|---|
| 754 |
|
|---|
| 755 | def make( self, specifier, rpath, zipped ):
|
|---|
| 756 | """ create (make) the filenames
|
|---|
| 757 |
|
|---|
| 758 | names : dictionary of filenames, tags { 'data', 'drscal', 'results' }
|
|---|
| 759 | data : name of the data file
|
|---|
| 760 | drscal : name of the drs calibration file
|
|---|
| 761 | results : radikal of file name(s) for results (to be completed by suffixes)
|
|---|
| 762 | """
|
|---|
| 763 |
|
|---|
| 764 | self.specifier = specifier
|
|---|
| 765 |
|
|---|
| 766 | if zipped:
|
|---|
| 767 | dpath = '/data00/fact-construction/raw/'
|
|---|
| 768 | ext = '.fits.gz'
|
|---|
| 769 | else:
|
|---|
| 770 | dpath = '/data03/fact-construction/raw/'
|
|---|
| 771 | ext = '.fits'
|
|---|
| 772 |
|
|---|
| 773 | year = specifier[2]
|
|---|
| 774 | month = specifier[3]
|
|---|
| 775 | day = specifier[4]
|
|---|
| 776 |
|
|---|
| 777 | yyyymmdd = year + month + day
|
|---|
| 778 | dfile = specifier[1]
|
|---|
| 779 | cfile = specifier[0]
|
|---|
| 780 |
|
|---|
| 781 | rpath = rpath + yyyymmdd + '/'
|
|---|
| 782 | self.rpath = rpath
|
|---|
| 783 | self.names = {}
|
|---|
| 784 |
|
|---|
| 785 | tmp = dpath + year + '/' + month + '/' + day + '/' + yyyymmdd + '_'
|
|---|
| 786 | self.names['data'] = tmp + dfile + ext
|
|---|
| 787 | self.names['drscal'] = tmp + cfile + '.drs' + ext
|
|---|
| 788 | self.names['results'] = rpath + yyyymmdd + '_' + dfile + '_' + cfile
|
|---|
| 789 |
|
|---|
| 790 | self.data = self.names['data']
|
|---|
| 791 | self.drscal = self.names['drscal']
|
|---|
| 792 | self.results = self.names['results']
|
|---|
| 793 |
|
|---|
| 794 | def info( self ):
|
|---|
| 795 | """ print complete filenames
|
|---|
| 796 |
|
|---|
| 797 | """
|
|---|
| 798 |
|
|---|
| 799 | print 'file names:'
|
|---|
| 800 | print 'data: ', self.names['data']
|
|---|
| 801 | print 'drs-cal: ', self.names['drscal']
|
|---|
| 802 | print 'results: ', self.names['results']
|
|---|
| 803 |
|
|---|
| 804 | # end of class definition: fnames( object )
|
|---|
| 805 |
|
|---|
| 806 | def _test_SlowData( filename ):
|
|---|
| 807 | print '-'*70
|
|---|
| 808 | print "opened :", filename, " as 'file'"
|
|---|
| 809 | print
|
|---|
| 810 | print '-'*70
|
|---|
| 811 | print 'type file.show() to look at its contents'
|
|---|
| 812 | print "type file.register( columnname ) or file.register('all') in order to register columns"
|
|---|
| 813 | print
|
|---|
| 814 | print " due column-registration you declare, that you would like to retrieve the contents of one of the columns"
|
|---|
| 815 | print " after column-registration, the 'file' has new member variables, they are named like the columns"
|
|---|
| 816 | print " PLEASE NOTE: immediatly after registration, the members exist, but they are empty."
|
|---|
| 817 | print " the values are assigned only, when you call file.next() or when you loop over the 'file'"
|
|---|
| 818 | print
|
|---|
| 819 | print "in order to loop over it, just go like this:"
|
|---|
| 820 | print "for row in file:"
|
|---|
| 821 | print " print row.columnname_one, row.columnname_two"
|
|---|
| 822 | print
|
|---|
| 823 | print ""
|
|---|
| 824 | print '-'*70
|
|---|
| 825 |
|
|---|
| 826 |
|
|---|
| 827 |
|
|---|
| 828 | def _test_iter( nevents ):
|
|---|
| 829 | """ test for function __iter__ """
|
|---|
| 830 |
|
|---|
| 831 | data_file_name = '/data00/fact-construction/raw/2011/11/24/20111124_117.fits.gz'
|
|---|
| 832 | calib_file_name = '/data00/fact-construction/raw/2011/11/24/20111124_114.drs.fits.gz'
|
|---|
| 833 | # data_file_name = '/home/luster/win7/FACT/data/raw/20120114/20120114_028.fits.gz'
|
|---|
| 834 | # calib_file_name = '/home/luster/win7/FACT/data/raw/20120114/20120114_022.drs.fits.gz'
|
|---|
| 835 | run = RawData( data_file_name, calib_file_name , return_dict=True)
|
|---|
| 836 |
|
|---|
| 837 | for event in run:
|
|---|
| 838 | print 'ev ', event['event_id'], 'data[0,0] = ', event['acal_data'][0,0], 'start_cell[0] = ', event['start_cells'][0], 'trigger type = ', event['trigger_type']
|
|---|
| 839 | if run.event_id == nevents:
|
|---|
| 840 | break
|
|---|
| 841 |
|
|---|
| 842 | if __name__ == '__main__':
|
|---|
| 843 | """ tests """
|
|---|
| 844 | import sys
|
|---|
| 845 | if len(sys.argv) == 1:
|
|---|
| 846 | print 'showing test of iterator of RawData class'
|
|---|
| 847 | print 'in order to test the SlowData classe please use:', sys.argv[0], 'fits-file-name'
|
|---|
| 848 | _test_iter(10)
|
|---|
| 849 |
|
|---|
| 850 |
|
|---|
| 851 | else:
|
|---|
| 852 | print 'showing test of SlowData class'
|
|---|
| 853 | print 'in case you wanted to test the RawData class, please give no commandline arguments'
|
|---|
| 854 | file = SlowData(sys.argv[1])
|
|---|
| 855 | _test_SlowData(sys.argv[1])
|
|---|
