schain Commit - r1400:005589ae0527 · Jicamarca Repository

hora y fecha en el primer bloque hdf5 >> consola

joabAM -

r1400:005589ae0527

parent child

Context file:

r1400:005589ae0527

Collapse all files

schainpy/model/io/jroIO_kamisr.py +3 -3

             ''''
             Created on Set 9, 2015
             @author: roj-idl71 Karim Kuyeng
             @update: 2021, Joab Apaza
             '''
             import os
             import sys
             import glob
             import fnmatch
             import datetime
             import time
             import re
             import h5py
             import numpy
             try:
                 from gevent import sleep
             except:
                 from time import sleep
             from schainpy.model.data.jroheaderIO import RadarControllerHeader, SystemHeader
             from schainpy.model.data.jrodata import Voltage
             from schainpy.model.proc.jroproc_base import ProcessingUnit, Operation, MPDecorator
             from numpy import imag
             class AMISRReader(ProcessingUnit):
                 '''
                 classdocs
                 '''
                 def __init__(self):
                     '''
                     Constructor
                     '''
                     ProcessingUnit.__init__(self)
                     self.set = None
                     self.subset = None
                     self.extension_file = '.h5'
                     self.dtc_str = 'dtc'
                     self.dtc_id = 0
                     self.status = True
                     self.isConfig = False
                     self.dirnameList = []
                     self.filenameList = []
                     self.fileIndex = None
                     self.flagNoMoreFiles = False
                     self.flagIsNewFile = 0
                     self.filename = ''
                     self.amisrFilePointer = None
                     self.realBeamCode = []
                     self.beamCodeMap = None
                     self.azimuthList = []
                     self.elevationList = []
                     self.dataShape = None
                     self.profileIndex = 0
                     self.beamCodeByFrame = None
                     self.radacTimeByFrame = None
                     self.dataset = None
                     self.__firstFile = True
                     self.buffer = None
                     self.timezone = 'ut'
                     self.__waitForNewFile = 20
                     self.__filename_online = None
                     #Is really necessary create the output object in the initializer
                     self.dataOut = Voltage()
                     self.dataOut.error=False
                 def setup(self,path=None,
                                 startDate=None,
                                 endDate=None,
                                 startTime=None,
                                 endTime=None,
                                 walk=True,
                                 timezone='ut',
                                 all=0,
                                 code = None,
                                 nCode = 0,
                                 nBaud = 0,
                                 online=False):
                     self.timezone = timezone
                     self.all = all
                     self.online = online
                     self.code = code
                     self.nCode = int(nCode)
                     self.nBaud = int(nBaud)
                     #self.findFiles()
                     if not(online):
                         #Busqueda de archivos offline
                         self.searchFilesOffLine(path, startDate, endDate, startTime, endTime, walk)
                     else:
                         self.searchFilesOnLine(path, startDate, endDate, startTime,endTime,walk)
                     if not(self.filenameList):
-                        print("There is no files into the folder: %s"%(path))
+                        raise schainpy.admin.SchainWarning("There is no files into the folder: %s"%(path))
                         sys.exit()
                     self.fileIndex = 0
                     self.readNextFile(online)
                     '''
                     Add code
                     '''
                     self.isConfig = True
                     # print("Setup Done")
                     pass
                 def readAMISRHeader(self,fp):
                     if self.isConfig and (not self.flagNoMoreFiles):
                         newShape = fp.get('Raw11/Data/Samples/Data').shape[1:]
                         if self.dataShape != newShape and newShape != None:
-                            print("\nNEW FILE HAS A DIFFERENT SHAPE")
+                            raise schainpy.admin.SchainError("NEW FILE HAS A DIFFERENT SHAPE: ")
                             print(self.dataShape,newShape,"\n")
                             return 0
                     else:
                         self.dataShape = fp.get('Raw11/Data/Samples/Data').shape[1:]
                     header = 'Raw11/Data/RadacHeader'
                     self.beamCodeByPulse = fp.get(header+'/BeamCode') # LIST OF BEAMS PER PROFILE, TO BE USED ON REARRANGE
                     if (self.startDate> datetime.date(2021, 7, 15)): #Se cambió la forma de extracción de Apuntes el 17
                         self.beamcodeFile = fp['Setup/Beamcodefile'][()].decode()
                         self.trueBeams = self.beamcodeFile.split("\n")
                         self.trueBeams.pop()#remove last
                         [self.realBeamCode.append(x) for x in self.trueBeams if x not in self.realBeamCode]
                         self.beamCode = [int(x, 16) for x in self.realBeamCode]
                     else:
                         _beamCode= fp.get('Raw11/Data/Beamcodes') #se usa la manera previa al cambio de apuntes
                         self.beamCode = _beamCode[0,:]
                     if self.beamCodeMap == None:
                         self.beamCodeMap = fp['Setup/BeamcodeMap']
                         for beam in self.beamCode:
                             beamAziElev = numpy.where(self.beamCodeMap[:,0]==beam)
                             beamAziElev = beamAziElev[0].squeeze()
                             self.azimuthList.append(self.beamCodeMap[beamAziElev,1])
                             self.elevationList.append(self.beamCodeMap[beamAziElev,2])
                             #print("Beamssss: ",self.beamCodeMap[beamAziElev,1],self.beamCodeMap[beamAziElev,2])
                     #print(self.beamCode)
                     #self.code = fp.get(header+'/Code') # NOT USE FOR THIS
                     self.frameCount = fp.get(header+'/FrameCount')# NOT USE FOR THIS
                     self.modeGroup = fp.get(header+'/ModeGroup')# NOT USE FOR THIS
                     self.nsamplesPulse = fp.get(header+'/NSamplesPulse')# TO GET NSA OR USING DATA FOR THAT
                     self.pulseCount = fp.get(header+'/PulseCount')# NOT USE FOR THIS
                     self.radacTime = fp.get(header+'/RadacTime')# 1st TIME ON FILE ANDE CALCULATE THE REST WITH IPP*nindexprofile
                     self.timeCount = fp.get(header+'/TimeCount')# NOT USE FOR THIS
                     self.timeStatus = fp.get(header+'/TimeStatus')# NOT USE FOR THIS
                     self.rangeFromFile = fp.get('Raw11/Data/Samples/Range')
                     self.frequency =  fp.get('Rx/Frequency')
                     txAus = fp.get('Raw11/Data/Pulsewidth')
                     self.nblocks = self.pulseCount.shape[0] #nblocks
                     self.nprofiles = self.pulseCount.shape[1] #nprofile
                     self.nsa = self.nsamplesPulse[0,0] #ngates
                     self.nchannels = len(self.beamCode)
                     self.ippSeconds = (self.radacTime[0][1] -self.radacTime[0][0]) #Ipp in seconds
                     #self.__waitForNewFile = self.nblocks  # wait depending on the number of blocks since each block is 1 sec
                     self.__waitForNewFile = self.nblocks * self.nprofiles * self.ippSeconds # wait until new file is created
                     #filling radar controller header parameters
                     self.__ippKm = self.ippSeconds *.15*1e6 # in km
                     self.__txA = (txAus.value)*.15 #(ipp[us]*.15km/1us) in km
                     self.__txB = 0
                     nWindows=1
                     self.__nSamples = self.nsa
                     self.__firstHeight = self.rangeFromFile[0][0]/1000 #in km
                     self.__deltaHeight = (self.rangeFromFile[0][1] - self.rangeFromFile[0][0])/1000
                     #for now until understand why the code saved is different (code included even though code not in tuf file)
                     #self.__codeType = 0
                    # self.__nCode = None
                    # self.__nBaud = None
                     self.__code = self.code
                     self.__codeType = 0
                     if self.code != None:
                         self.__codeType = 1
                     self.__nCode = self.nCode
                     self.__nBaud = self.nBaud
                     #self.__code = 0
                     #filling system header parameters
                     self.__nSamples = self.nsa
                     self.newProfiles = self.nprofiles/self.nchannels
                     self.__channelList = list(range(self.nchannels))
                     self.__frequency = self.frequency[0][0]
                     return 1
                 def createBuffers(self):
                     pass
                 def __setParameters(self,path='', startDate='',endDate='',startTime='', endTime='', walk=''):
                     self.path = path
                     self.startDate = startDate
                     self.endDate = endDate
                     self.startTime = startTime
                     self.endTime = endTime
                     self.walk = walk
                 def __checkPath(self):
                     if os.path.exists(self.path):
                         self.status = 1
                     else:
                         self.status = 0
                         print('Path:%s does not exists'%self.path)
                     return
                 def __selDates(self, amisr_dirname_format):
                     try:
                         year = int(amisr_dirname_format[0:4])
                         month = int(amisr_dirname_format[4:6])
                         dom = int(amisr_dirname_format[6:8])
                         thisDate = datetime.date(year,month,dom)
                         #margen de un día extra, igual luego se filtra for fecha y hora
                         if (thisDate>=(self.startDate - datetime.timedelta(days=1)) and thisDate <= (self.endDate)+ datetime.timedelta(days=1)):
                             return amisr_dirname_format
                     except:
                         return None
                 def __findDataForDates(self,online=False):
                     if not(self.status):
                         return None
                     pat = '\d+.\d+'
                     dirnameList = [re.search(pat,x) for x in os.listdir(self.path)]
                     dirnameList = [x for x in dirnameList if x!=None]
                     dirnameList = [x.string for x in dirnameList]
                     if not(online):
                         dirnameList = [self.__selDates(x) for x in dirnameList]
                         dirnameList = [x for x in dirnameList if x!=None]
                     if len(dirnameList)>0:
                         self.status = 1
                         self.dirnameList = dirnameList
                         self.dirnameList.sort()
                     else:
                         self.status = 0
                         return None
                 def __getTimeFromData(self):
                     startDateTime_Reader = datetime.datetime.combine(self.startDate,self.startTime)
                     endDateTime_Reader = datetime.datetime.combine(self.endDate,self.endTime)
                     print('Filtering Files from %s to %s'%(startDateTime_Reader, endDateTime_Reader))
                     print('........................................')
                     filter_filenameList = []
                     self.filenameList.sort()
                     #for i in range(len(self.filenameList)-1):
                     for i in range(len(self.filenameList)):
                         filename = self.filenameList[i]
                         fp = h5py.File(filename,'r')
                         time_str = fp.get('Time/RadacTimeString')
                         startDateTimeStr_File = time_str[0][0].decode('UTF-8').split('.')[0]
                         #startDateTimeStr_File = "2019-12-16 09:21:11"
                         junk = time.strptime(startDateTimeStr_File, '%Y-%m-%d %H:%M:%S')
                         startDateTime_File = datetime.datetime(junk.tm_year,junk.tm_mon,junk.tm_mday,junk.tm_hour, junk.tm_min, junk.tm_sec)
                         #endDateTimeStr_File = "2019-12-16 11:10:11"
                         endDateTimeStr_File = time_str[-1][-1].decode('UTF-8').split('.')[0]
                         junk = time.strptime(endDateTimeStr_File, '%Y-%m-%d %H:%M:%S')
                         endDateTime_File = datetime.datetime(junk.tm_year,junk.tm_mon,junk.tm_mday,junk.tm_hour, junk.tm_min, junk.tm_sec)
                         fp.close()
                         #print("check time", startDateTime_File)
                         if self.timezone == 'lt':
                             startDateTime_File = startDateTime_File - datetime.timedelta(minutes = 300)
                             endDateTime_File = endDateTime_File - datetime.timedelta(minutes = 300)
                         if (startDateTime_File >=startDateTime_Reader and endDateTime_File<=endDateTime_Reader):
                             filter_filenameList.append(filename)
                         if (startDateTime_File>endDateTime_Reader):
                             break
                     filter_filenameList.sort()
                     self.filenameList = filter_filenameList
                     return 1
                 def __filterByGlob1(self, dirName):
                     filter_files = glob.glob1(dirName, '*.*%s'%self.extension_file)
                     filter_files.sort()
                     filterDict = {}
                     filterDict.setdefault(dirName)
                     filterDict[dirName] = filter_files
                     return filterDict
                 def __getFilenameList(self, fileListInKeys, dirList):
                     for value in fileListInKeys:
                         dirName = list(value.keys())[0]
                         for file in value[dirName]:
                             filename = os.path.join(dirName, file)
                             self.filenameList.append(filename)
                 def __selectDataForTimes(self, online=False):
                     #aun no esta implementado el filtro for tiempo
                     if not(self.status):
                         return None
                     dirList = [os.path.join(self.path,x) for x in self.dirnameList]
                     fileListInKeys = [self.__filterByGlob1(x) for x in dirList]
                     self.__getFilenameList(fileListInKeys, dirList)
                     if not(online):
                         #filtro por tiempo
                         if not(self.all):
                             self.__getTimeFromData()
                         if len(self.filenameList)>0:
                             self.status = 1
                             self.filenameList.sort()
                         else:
                             self.status = 0
                             return None
                     else:
                         #get the last file - 1
                         self.filenameList = [self.filenameList[-2]]
                     new_dirnameList = []
                     for dirname in self.dirnameList:
                         junk = numpy.array([dirname in x for x in self.filenameList])
                         junk_sum = junk.sum()
                         if junk_sum > 0:
                             new_dirnameList.append(dirname)
                     self.dirnameList = new_dirnameList
                     return 1
                 def searchFilesOnLine(self, path, startDate, endDate, startTime=datetime.time(0,0,0),
                                         endTime=datetime.time(23,59,59),walk=True):
                     if endDate ==None:
                      startDate = datetime.datetime.utcnow().date()
                      endDate = datetime.datetime.utcnow().date()
                     self.__setParameters(path=path, startDate=startDate, endDate=endDate,startTime = startTime,endTime=endTime, walk=walk)
                     self.__checkPath()
                     self.__findDataForDates(online=True)
                     self.dirnameList = [self.dirnameList[-1]]
                     self.__selectDataForTimes(online=True)
                     return
                 def searchFilesOffLine(self,
                                         path,
                                         startDate,
                                         endDate,
                                         startTime=datetime.time(0,0,0),
                                         endTime=datetime.time(23,59,59),
                                         walk=True):
                     self.__setParameters(path, startDate, endDate, startTime, endTime, walk)
                     self.__checkPath()
                     self.__findDataForDates()
                     self.__selectDataForTimes()
                     for i in range(len(self.filenameList)):
                         print("%s" %(self.filenameList[i]))
                     return
                 def __setNextFileOffline(self):
                     try:
                         self.filename = self.filenameList[self.fileIndex]
                         self.amisrFilePointer = h5py.File(self.filename,'r')
                         self.fileIndex += 1
                     except:
                         self.flagNoMoreFiles = 1
-                        print("No more Files")
+                        raise schainpy.admin.SchainError('No more files to read')
                         return 0
                     self.flagIsNewFile = 1
                     print("Setting the file: %s"%self.filename)
                     return 1
                 def __setNextFileOnline(self):
                     filename = self.filenameList[0]
                     if self.__filename_online != None:
                         self.__selectDataForTimes(online=True)
                         filename = self.filenameList[0]
                         wait = 0
                         self.__waitForNewFile=300 ## DEBUG:
                         while self.__filename_online == filename:
                             print('waiting %d seconds to get a new file...'%(self.__waitForNewFile))
                             if wait == 5:
                                 self.flagNoMoreFiles = 1
                                 return 0
                             sleep(self.__waitForNewFile)
                             self.__selectDataForTimes(online=True)
                             filename = self.filenameList[0]
                             wait += 1
                     self.__filename_online = filename
                     self.amisrFilePointer = h5py.File(filename,'r')
                     self.flagIsNewFile = 1
                     self.filename = filename
                     print("Setting the file: %s"%self.filename)
                     return 1
                 def readData(self):
                     buffer = self.amisrFilePointer.get('Raw11/Data/Samples/Data')
                     re = buffer[:,:,:,0]
                     im = buffer[:,:,:,1]
                     dataset = re + im*1j
                     self.radacTime = self.amisrFilePointer.get('Raw11/Data/RadacHeader/RadacTime')
                     timeset = self.radacTime[:,0]
                     return dataset,timeset
                 def reshapeData(self):
                 #self.beamCodeByPulse, self.beamCode, self.nblocks, self.nprofiles, self.nsa,
                     channels = self.beamCodeByPulse[0,:]
                     nchan = self.nchannels
                     #self.newProfiles = self.nprofiles/nchan #must be defined on filljroheader
                     nblocks = self.nblocks
                     nsamples = self.nsa
                     #Dimensions : nChannels, nProfiles, nSamples
                     new_block = numpy.empty((nblocks, nchan, numpy.int_(self.newProfiles), nsamples), dtype="complex64")
                     ############################################
                     for thisChannel in range(nchan):
                         new_block[:,thisChannel,:,:] = self.dataset[:,numpy.where(channels==self.beamCode[thisChannel])[0],:]
                     new_block = numpy.transpose(new_block, (1,0,2,3))
                     new_block = numpy.reshape(new_block, (nchan,-1, nsamples))
                     return new_block
                 def updateIndexes(self):
                     pass
                 def fillJROHeader(self):
                     #fill radar controller header
                     self.dataOut.radarControllerHeaderObj = RadarControllerHeader(ipp=self.__ippKm,
                                                                                   txA=self.__txA,
                                                                                   txB=0,
                                                                                   nWindows=1,
                                                                                   nHeights=self.__nSamples,
                                                                                   firstHeight=self.__firstHeight,
                                                                                   deltaHeight=self.__deltaHeight,
                                                                                   codeType=self.__codeType,
                                                                                   nCode=self.__nCode, nBaud=self.__nBaud,
                                                                                   code = self.__code,
                                                                                   fClock=1)
                     #fill system header
                     self.dataOut.systemHeaderObj = SystemHeader(nSamples=self.__nSamples,
                                                                 nProfiles=self.newProfiles,
                                                                 nChannels=len(self.__channelList),
                                                                 adcResolution=14,
                                                                 pciDioBusWidth=32)
                     self.dataOut.type = "Voltage"
                     self.dataOut.data = None
                     self.dataOut.dtype = numpy.dtype([('real','<i8'),('imag','<i8')])
             #        self.dataOut.nChannels = 0
             #        self.dataOut.nHeights = 0
                     self.dataOut.nProfiles = self.newProfiles*self.nblocks
                     #self.dataOut.heightList = self.__firstHeigth + numpy.arange(self.__nSamples, dtype = numpy.float)*self.__deltaHeigth
                     ranges = numpy.reshape(self.rangeFromFile.value,(-1))
                     self.dataOut.heightList =  ranges/1000.0 #km
                     self.dataOut.channelList = self.__channelList
                     self.dataOut.blocksize = self.dataOut.nChannels * self.dataOut.nHeights
             #        self.dataOut.channelIndexList = None
                     self.dataOut.azimuthList = numpy.array(self.azimuthList)
                     self.dataOut.elevationList = numpy.array(self.elevationList)
                     self.dataOut.codeList = numpy.array(self.beamCode)
                     #print(self.dataOut.elevationList)
                     self.dataOut.flagNoData = True
                     #Set to TRUE if the data is discontinuous
                     self.dataOut.flagDiscontinuousBlock = False
                     self.dataOut.utctime = None
                     #self.dataOut.timeZone = -5 #self.__timezone/60  #timezone like jroheader, difference in minutes between UTC and localtime
                     if self.timezone == 'lt':
                         self.dataOut.timeZone = time.timezone / 60. #get the timezone in minutes
                     else:
                         self.dataOut.timeZone = 0 #by default time is UTC
                     self.dataOut.dstFlag = 0
                     self.dataOut.errorCount = 0
                     self.dataOut.nCohInt = 1
                     self.dataOut.flagDecodeData = False #asumo que la data esta decodificada
                     self.dataOut.flagDeflipData = False #asumo que la data esta sin flip
                     self.dataOut.flagShiftFFT = False
                     self.dataOut.ippSeconds = self.ippSeconds
                     #Time interval between profiles
                     #self.dataOut.timeInterval = self.dataOut.ippSeconds * self.dataOut.nCohInt
                     self.dataOut.frequency = self.__frequency
                     self.dataOut.realtime = self.online
                     pass
                 def readNextFile(self,online=False):
                     if not(online):
                         newFile = self.__setNextFileOffline()
                     else:
                         newFile = self.__setNextFileOnline()
                     if not(newFile):
                         self.dataOut.error = True
                         return 0
                     if not self.readAMISRHeader(self.amisrFilePointer):
                         self.dataOut.error = True
                         return 0
                     self.createBuffers()
                     self.fillJROHeader()
                     #self.__firstFile = False
                     self.dataset,self.timeset = self.readData()
                     if self.endDate!=None:
                      endDateTime_Reader = datetime.datetime.combine(self.endDate,self.endTime)
                      time_str = self.amisrFilePointer.get('Time/RadacTimeString')
                      startDateTimeStr_File = time_str[0][0].decode('UTF-8').split('.')[0]
                      junk = time.strptime(startDateTimeStr_File, '%Y-%m-%d %H:%M:%S')
                      startDateTime_File = datetime.datetime(junk.tm_year,junk.tm_mon,junk.tm_mday,junk.tm_hour, junk.tm_min, junk.tm_sec)
                      if self.timezone == 'lt':
                       startDateTime_File = startDateTime_File - datetime.timedelta(minutes = 300)
                      if (startDateTime_File>endDateTime_Reader):
                          return 0
                     self.jrodataset = self.reshapeData()
                     #----self.updateIndexes()
                     self.profileIndex = 0
                     return 1
                 def __hasNotDataInBuffer(self):
                     if self.profileIndex >= (self.newProfiles*self.nblocks):
                         return 1
                     return 0
                 def getData(self):
                     if self.flagNoMoreFiles:
                         self.dataOut.flagNoData = True
                         return 0
                     if self.__hasNotDataInBuffer():
                         if not (self.readNextFile(self.online)):
                             return 0
                     if self.dataset is None: # setear esta condicion cuando no hayan datos por leer
                         self.dataOut.flagNoData = True
                         return 0
                     #self.dataOut.data = numpy.reshape(self.jrodataset[self.profileIndex,:],(1,-1))
                     self.dataOut.data = self.jrodataset[:,self.profileIndex,:]
                     #print("R_t",self.timeset)
                     #self.dataOut.utctime = self.jrotimeset[self.profileIndex]
                     #verificar basic header de jro data y ver si es compatible con este valor
                     #self.dataOut.utctime = self.timeset + (self.profileIndex * self.ippSeconds * self.nchannels)
                     indexprof = numpy.mod(self.profileIndex, self.newProfiles)
                     indexblock = self.profileIndex/self.newProfiles
                     #print (indexblock, indexprof)
                     diffUTC = 0
                     t_comp = (indexprof * self.ippSeconds * self.nchannels) + diffUTC #
                     #print("utc :",indexblock," __ ",t_comp)
                     #print(numpy.shape(self.timeset))
                     self.dataOut.utctime = self.timeset[numpy.int_(indexblock)] + t_comp
                     #self.dataOut.utctime = self.timeset[self.profileIndex] + t_comp
                     self.dataOut.profileIndex = self.profileIndex
                     #print("N profile:",self.profileIndex,self.newProfiles,self.nblocks,self.dataOut.utctime)
                     self.dataOut.flagNoData = False
                     # if indexprof == 0:
                     #     print("kamisr: ",self.dataOut.utctime)
                     self.profileIndex += 1
                     return self.dataOut.data
                 def run(self, **kwargs):
                     '''
                     This method will be called many times so here you should put all your code
                     '''
                     #print("running kamisr")
                     if not self.isConfig:
                         self.setup(**kwargs)
                         self.isConfig = True
                     self.getData()

schainpy/model/io/jroIO_param.py +9 -1

             import os
             import time
             import datetime
             import numpy
             import h5py
             import schainpy.admin
             from schainpy.model.data.jrodata import *
             from schainpy.model.proc.jroproc_base import ProcessingUnit, Operation, MPDecorator
             from schainpy.model.io.jroIO_base import *
             from schainpy.utils import log
             class HDFReader(Reader, ProcessingUnit):
                 """Processing unit to read HDF5 format files
                 This unit reads HDF5 files created with `HDFWriter` operation contains
                 by default two groups Data and Metadata all variables would be saved as `dataOut`
                 attributes.
                 It is possible to read any HDF5 file by given the structure in the `description`
                 parameter, also you can add extra values to metadata with the parameter `extras`.
                 Parameters:
                 -----------
                 path : str
                     Path where files are located.
                 startDate : date
                     Start date of the files
                 endDate : list
                     End date of the files
                 startTime : time
                     Start time of the files
                 endTime : time
                     End time of the files
                 description : dict, optional
                     Dictionary with the description of the HDF5 file
                 extras : dict, optional
                     Dictionary with extra metadata to be be added to `dataOut`
                 Examples
                 --------
                 desc = {
                     'Data': {
                         'data_output': ['u', 'v', 'w'],
                         'utctime': 'timestamps',
                     }  ,
                     'Metadata': {
                         'heightList': 'heights'
                     }
                 }
                 desc = {
                     'Data': {
                         'data_output': 'winds',
                         'utctime': 'timestamps'
                     },
                     'Metadata': {
                         'heightList': 'heights'
                     }
                 }
                 extras = {
                     'timeZone': 300
                 }
                 reader = project.addReadUnit(
                     name='HDFReader',
                     path='/path/to/files',
                     startDate='2019/01/01',
                     endDate='2019/01/31',
                     startTime='00:00:00',
                     endTime='23:59:59',
                     # description=json.dumps(desc),
                     # extras=json.dumps(extras),
                     )
                 """
                 __attrs__ = ['path', 'startDate', 'endDate', 'startTime', 'endTime', 'description', 'extras']
                 def __init__(self):
                     ProcessingUnit.__init__(self)
                     self.dataOut = Parameters()
                     self.ext = ".hdf5"
                     self.optchar = "D"
                     self.meta = {}
                     self.data = {}
                     self.open_file = h5py.File
                     self.open_mode = 'r'
                     self.description = {}
                     self.extras = {}
                     self.filefmt = "*%Y%j***"
                     self.folderfmt = "*%Y%j"
                     self.utcoffset = 0
                 def setup(self, **kwargs):
                     self.set_kwargs(**kwargs)
                     if not self.ext.startswith('.'):
                         self.ext = '.{}'.format(self.ext)
                     if self.online:
                         log.log("Searching files in online mode...", self.name)
                         for nTries in range(self.nTries):
                             fullpath = self.searchFilesOnLine(self.path, self.startDate,
                                 self.endDate, self.expLabel, self.ext, self.walk,
                                 self.filefmt, self.folderfmt)
                             pathname, filename = os.path.split(fullpath)
                             #print(pathname,filename)
                             try:
                                 fullpath = next(fullpath)
                             except:
                                 fullpath = None
                             if fullpath:
                                 break
                             log.warning(
                                 'Waiting {} sec for a valid file in {}: try {} ...'.format(
                                     self.delay, self.path, nTries + 1),
                                 self.name)
                             time.sleep(self.delay)
                         if not(fullpath):
                             raise schainpy.admin.SchainError(
                                 'There isn\'t any valid file in {}'.format(self.path))
                         pathname, filename = os.path.split(fullpath)
                         self.year = int(filename[1:5])
                         self.doy = int(filename[5:8])
                         self.set = int(filename[8:11]) - 1
                     else:
                         log.log("Searching files in {}".format(self.path), self.name)
                         self.filenameList = self.searchFilesOffLine(self.path, self.startDate,
                             self.endDate, self.expLabel, self.ext, self.walk, self.filefmt, self.folderfmt)
                     self.setNextFile()
                     return
                 def readFirstHeader(self):
                     '''Read metadata and data'''
                     self.__readMetadata()
                     self.__readData()
                     self.__setBlockList()
                     if 'type' in self.meta:
                         self.dataOut = eval(self.meta['type'])()
                     for attr in self.meta:
                         #print("attr: ", attr)
                         setattr(self.dataOut, attr, self.meta[attr])
                     self.blockIndex = 0
                     return
                 def __setBlockList(self):
                     '''
                     Selects the data within the times defined
                     self.fp
                     self.startTime
                     self.endTime
                     self.blockList
                     self.blocksPerFile
                     '''
                     startTime = self.startTime
                     endTime = self.endTime
                     thisUtcTime = self.data['utctime'] + self.utcoffset
                     self.interval = numpy.min(thisUtcTime[1:] - thisUtcTime[:-1])
                     thisDatetime = datetime.datetime.utcfromtimestamp(thisUtcTime[0])
                     self.startFileDatetime = thisDatetime
                     thisDate = thisDatetime.date()
                     thisTime = thisDatetime.time()
                     startUtcTime = (datetime.datetime.combine(thisDate, startTime) - datetime.datetime(1970, 1, 1)).total_seconds()
                     endUtcTime = (datetime.datetime.combine(thisDate, endTime) - datetime.datetime(1970, 1, 1)).total_seconds()
                     ind = numpy.where(numpy.logical_and(thisUtcTime >= startUtcTime, thisUtcTime < endUtcTime))[0]
                     self.blockList = ind
                     self.blocksPerFile = len(ind)
                     self.blocksPerFile = len(thisUtcTime)
                     return
                 def __readMetadata(self):
                     '''
                     Reads Metadata
                     '''
                     meta = {}
                     if self.description:
                         for key, value in self.description['Metadata'].items():
                             meta[key] = self.fp[value][()]
                     else:
                         grp = self.fp['Metadata']
                         for name in grp:
                             meta[name] = grp[name][()]
                     if self.extras:
                         for key, value in self.extras.items():
                             meta[key] = value
                     self.meta = meta
                     return
                 def checkForRealPath(self, nextFile, nextDay):
                     # print("check FRP")
                     # dt = self.startFileDatetime + datetime.timedelta(1)
                     # filename = '{}.{}{}'.format(self.path, dt.strftime('%Y%m%d'), self.ext)
                     # fullfilename = os.path.join(self.path, filename)
                     # print("check Path ",fullfilename,filename)
                     # if os.path.exists(fullfilename):
                     #     return fullfilename, filename
                     # return None, filename
                     return None,None
                 def __readData(self):
                     data = {}
                     if self.description:
                         for key, value in self.description['Data'].items():
                             if isinstance(value, str):
                                 if isinstance(self.fp[value], h5py.Dataset):
                                     data[key] = self.fp[value][()]
                                 elif isinstance(self.fp[value], h5py.Group):
                                     array = []
                                     for ch in self.fp[value]:
                                         array.append(self.fp[value][ch][()])
                                     data[key] = numpy.array(array)
                             elif isinstance(value, list):
                                 array = []
                                 for ch in value:
                                     array.append(self.fp[ch][()])
                                 data[key] = numpy.array(array)
                     else:
                         grp = self.fp['Data']
                         for name in grp:
                             if isinstance(grp[name], h5py.Dataset):
                                 array = grp[name][()]
                             elif isinstance(grp[name], h5py.Group):
                                 array = []
                                 for ch in grp[name]:
                                     array.append(grp[name][ch][()])
                                 array = numpy.array(array)
                             else:
                                 log.warning('Unknown type: {}'.format(name))
                             if name in self.description:
                                 key = self.description[name]
                             else:
                                 key = name
                             data[key] = array
                     self.data = data
                     return
                 def getData(self):
                     if not self.isDateTimeInRange(self.startFileDatetime, self.startDate, self.endDate, self.startTime, self.endTime):
                         self.dataOut.flagNoData = True
                         self.blockIndex = self.blocksPerFile
                         #self.dataOut.error = True      TERMINA EL PROGRAMA, removido
                         return
                     for attr in self.data:
                         if self.data[attr].ndim == 1:
                             setattr(self.dataOut, attr, self.data[attr][self.blockIndex])
                         else:
                             setattr(self.dataOut, attr, self.data[attr][:, self.blockIndex])
                     self.dataOut.flagNoData = False
                     self.blockIndex += 1
+                    if self.blockIndex == 1:
                         log.log("Block No. {}/{} -> {}".format(
                             self.blockIndex,
                             self.blocksPerFile,
                             self.dataOut.datatime.ctime()), self.name)
+                    else:
+                        log.log("Block No. {}/{} ".format(
+                            self.blockIndex,
+                            self.blocksPerFile),self.name)
                     return
                 def run(self, **kwargs):
                     if not(self.isConfig):
                         self.setup(**kwargs)
                         self.isConfig = True
                     if self.blockIndex == self.blocksPerFile:
                         self.setNextFile()
                     self.getData()
                     return
             @MPDecorator
             class HDFWriter(Operation):
                 """Operation to write HDF5 files.
                 The HDF5 file contains by default two groups Data and Metadata where
                 you can save any `dataOut` attribute specified by `dataList` and `metadataList`
                 parameters, data attributes are normaly time dependent where the metadata
                 are not.
                 It is possible to customize the structure of the HDF5 file with the
                 optional description parameter see the examples.
                 Parameters:
                 -----------
                 path : str
                     Path where files will be saved.
                 blocksPerFile : int
                     Number of blocks per file
                 metadataList : list
                     List of the dataOut attributes that will be saved as metadata
                 dataList : int
                     List of the dataOut attributes that will be saved as data
                 setType : bool
                     If True the name of the files corresponds to the timestamp of the data
                 description : dict, optional
                     Dictionary with the desired description of the HDF5 file
                 Examples
                 --------
                 desc = {
                     'data_output': {'winds': ['z', 'w', 'v']},
                     'utctime': 'timestamps',
                     'heightList': 'heights'
                 }
                 desc = {
                     'data_output': ['z', 'w', 'v'],
                     'utctime': 'timestamps',
                     'heightList': 'heights'
                 }
                 desc = {
                     'Data': {
                         'data_output': 'winds',
                         'utctime': 'timestamps'
                     },
                     'Metadata': {
                         'heightList': 'heights'
                     }
                 }
                 writer = proc_unit.addOperation(name='HDFWriter')
                 writer.addParameter(name='path', value='/path/to/file')
                 writer.addParameter(name='blocksPerFile', value='32')
                 writer.addParameter(name='metadataList', value='heightList,timeZone')
                 writer.addParameter(name='dataList',value='data_output,utctime')
                 # writer.addParameter(name='description',value=json.dumps(desc))
                 """
                 ext = ".hdf5"
                 optchar = "D"
                 filename = None
                 path = None
                 setFile = None
                 fp = None
                 firsttime = True
                 #Configurations
                 blocksPerFile = None
                 blockIndex = None
                 dataOut = None
                 #Data Arrays
                 dataList = None
                 metadataList = None
                 currentDay = None
                 lastTime = None
                 def __init__(self):
                     Operation.__init__(self)
                     return
                 def setup(self, path=None, blocksPerFile=10, metadataList=None, dataList=None, setType=None, description=None):
                     self.path = path
                     self.blocksPerFile = blocksPerFile
                     self.metadataList = metadataList
                     self.dataList = [s.strip() for s in dataList]
                     self.setType = setType
                     self.description = description
                     if self.metadataList is None:
                         self.metadataList = self.dataOut.metadata_list
                     tableList = []
                     dsList = []
                     for i in range(len(self.dataList)):
                         dsDict = {}
                         if hasattr(self.dataOut, self.dataList[i]):
                             dataAux = getattr(self.dataOut, self.dataList[i])
                             dsDict['variable'] = self.dataList[i]
                         else:
                             log.warning('Attribute {} not found in dataOut', self.name)
                             continue
                         if dataAux is None:
                             continue
                         elif isinstance(dataAux, (int, float, numpy.integer, numpy.float)):
                             dsDict['nDim'] = 0
                         else:
                             dsDict['nDim'] = len(dataAux.shape)
                             dsDict['shape'] = dataAux.shape
                             dsDict['dsNumber'] = dataAux.shape[0]
                             dsDict['dtype'] = dataAux.dtype
                         dsList.append(dsDict)
                     self.dsList = dsList
                     self.currentDay = self.dataOut.datatime.date()
                 def timeFlag(self):
                     currentTime = self.dataOut.utctime
                     timeTuple = time.localtime(currentTime)
                     dataDay = timeTuple.tm_yday
                     #print("time UTC: ",currentTime, self.dataOut.datatime)
                     if self.lastTime is None:
                         self.lastTime = currentTime
                         self.currentDay = dataDay
                         return False
                     timeDiff = currentTime - self.lastTime
                     #Si el dia es diferente o si la diferencia entre un dato y otro supera la hora
                     if dataDay != self.currentDay:
                         self.currentDay = dataDay
                         return True
                     elif timeDiff > 3*60*60:
                         self.lastTime = currentTime
                         return True
                     else:
                         self.lastTime = currentTime
                         return False
                 def run(self, dataOut, path, blocksPerFile=10, metadataList=None,
                         dataList=[], setType=None, description={}):
                     self.dataOut = dataOut
                     if not(self.isConfig):
                         self.setup(path=path, blocksPerFile=blocksPerFile,
                                    metadataList=metadataList, dataList=dataList,
                                    setType=setType, description=description)
                         self.isConfig = True
                         self.setNextFile()
                     self.putData()
                     return
                 def setNextFile(self):
                     ext = self.ext
                     path = self.path
                     setFile = self.setFile
                     timeTuple = time.gmtime(self.dataOut.utctime)
                     #print("path: ",timeTuple)
                     subfolder = 'd%4.4d%3.3d' % (timeTuple.tm_year,timeTuple.tm_yday)
                     fullpath = os.path.join(path, subfolder)
                     if os.path.exists(fullpath):
                         filesList = os.listdir(fullpath)
                         filesList = [k for k in filesList if k.startswith(self.optchar)]
                         if len( filesList ) > 0:
                             filesList = sorted(filesList, key=str.lower)
                             filen = filesList[-1]
                             # el filename debera tener el siguiente formato
                             # 0 1234 567 89A BCDE (hex)
                             # x YYYY DDD SSS .ext
                             if isNumber(filen[8:11]):
                                 setFile = int(filen[8:11]) #inicializo mi contador de seteo al seteo del ultimo file
                             else:
                                 setFile = -1
                         else:
                             setFile = -1 #inicializo mi contador de seteo
                     else:
                         os.makedirs(fullpath)
                         setFile = -1 #inicializo mi contador de seteo
                     if self.setType is None:
                         setFile += 1
                         file = '%s%4.4d%3.3d%03d%s' % (self.optchar,
                                                        timeTuple.tm_year,
                                                        timeTuple.tm_yday,
                                                        setFile,
                                                        ext )
                     else:
                         setFile = timeTuple.tm_hour*60+timeTuple.tm_min
                         file = '%s%4.4d%3.3d%04d%s' % (self.optchar,
                                                        timeTuple.tm_year,
                                                        timeTuple.tm_yday,
                                                        setFile,
                                                        ext )
                     self.filename = os.path.join( path, subfolder, file )
                     #Setting HDF5 File
                     self.fp = h5py.File(self.filename, 'w')
                     #write metadata
                     self.writeMetadata(self.fp)
                     #Write data
                     self.writeData(self.fp)
                 def getLabel(self, name, x=None):
                     if x is None:
                         if 'Data' in self.description:
                             data = self.description['Data']
                             if 'Metadata' in self.description:
                                 data.update(self.description['Metadata'])
                         else:
                             data = self.description
                         if name in data:
                             if isinstance(data[name], str):
                                 return data[name]
                             elif isinstance(data[name], list):
                                 return None
                             elif isinstance(data[name], dict):
                                 for key, value in data[name].items():
                                     return key
                         return name
                     else:
                         if 'Metadata' in self.description:
                             meta = self.description['Metadata']
                         else:
                             meta = self.description
                         if name in meta:
                             if isinstance(meta[name], list):
                                 return meta[name][x]
                             elif isinstance(meta[name], dict):
                                 for key, value in meta[name].items():
                                     return value[x]
                         if 'cspc' in name:
                             return 'pair{:02d}'.format(x)
                         else:
                             return 'channel{:02d}'.format(x)
                 def writeMetadata(self, fp):
                     if self.description:
                         if 'Metadata' in self.description:
                             grp = fp.create_group('Metadata')
                         else:
                             grp = fp
                     else:
                         grp = fp.create_group('Metadata')
                     for i in range(len(self.metadataList)):
                         if not hasattr(self.dataOut, self.metadataList[i]):
                             log.warning('Metadata: `{}` not found'.format(self.metadataList[i]), self.name)
                             continue
                         value = getattr(self.dataOut, self.metadataList[i])
                         if isinstance(value, bool):
                             if value is True:
                                 value = 1
                             else:
                                 value = 0
                         grp.create_dataset(self.getLabel(self.metadataList[i]), data=value)
                     return
                 def writeData(self, fp):
                     if self.description:
                         if 'Data' in self.description:
                             grp = fp.create_group('Data')
                         else:
                             grp = fp
                     else:
                         grp = fp.create_group('Data')
                     dtsets = []
                     data = []
                     for dsInfo in self.dsList:
                         if dsInfo['nDim'] == 0:
                             ds = grp.create_dataset(
                                 self.getLabel(dsInfo['variable']),
                                 (self.blocksPerFile, ),
                                 chunks=True,
                                 dtype=numpy.float64)
                             dtsets.append(ds)
                             data.append((dsInfo['variable'], -1))
                         else:
                             label = self.getLabel(dsInfo['variable'])
                             if label is not None:
                                 sgrp = grp.create_group(label)
                             else:
                                 sgrp = grp
                             for i in range(dsInfo['dsNumber']):
                                 ds = sgrp.create_dataset(
                                     self.getLabel(dsInfo['variable'], i),
                                     (self.blocksPerFile, ) + dsInfo['shape'][1:],
                                     chunks=True,
                                     dtype=dsInfo['dtype'])
                                 dtsets.append(ds)
                                 data.append((dsInfo['variable'], i))
                     fp.flush()
                     log.log('Creating file: {}'.format(fp.filename), self.name)
                     self.ds = dtsets
                     self.data = data
                     self.firsttime = True
                     self.blockIndex = 0
                     return
                 def putData(self):
                     if (self.blockIndex == self.blocksPerFile) or self.timeFlag():
                         self.closeFile()
                         self.setNextFile()
                     for i, ds in enumerate(self.ds):
                         attr, ch = self.data[i]
                         if ch == -1:
                             ds[self.blockIndex] = getattr(self.dataOut, attr)
                         else:
                             ds[self.blockIndex] = getattr(self.dataOut, attr)[ch]
                     self.fp.flush()
                     self.blockIndex += 1
+                    if self.blockIndex == 1:
+                        log.log('Block No. {}/{} --> {}'.format(self.blockIndex, self.blocksPerFile,self.dataOut.datatime.ctime()), self.name)
+                    else:
                         log.log('Block No. {}/{}'.format(self.blockIndex, self.blocksPerFile), self.name)
                     return
                 def closeFile(self):
                     if self.blockIndex != self.blocksPerFile:
                         for ds in self.ds:
                             ds.resize(self.blockIndex, axis=0)
                     if self.fp:
                         self.fp.flush()
                         self.fp.close()
                 def close(self):
                     self.closeFile()

General Comments 0

Write
Preview

You need to be logged in to leave comments. Login now

No TODOs yet

	Site-wide shortcuts
/	Use quick search box
g h	Goto home page
g g	Goto my private gists page
g G	Goto my public gists page
g 0-9	Goto bookmarked items from 0-9
n r	New repository page
n g	New gist page

	Repositories
g s	Goto summary page
g c	Goto changelog page
g f	Goto files page
g F	Goto files page with file search activated
g p	Goto pull requests page
g o	Goto repository settings
g O	Goto repository access permissions settings
t s	Toggle sidebar on some pages