| 1 | # -*- coding: iso-8859-15 -*- |
|---|
| 2 | # Python reader of ISD files |
|---|
| 3 | ## e.g. # read_ISD.py -c '#' -f 722317-13970-2005 -d description.dat -g true -t 19491201000000,seconds -s 'Baton Rouge METRO R,30.537,-91.147,23.2' |
|---|
| 4 | import numpy as np |
|---|
| 5 | from optparse import OptionParser |
|---|
| 6 | import os |
|---|
| 7 | from netCDF4 import Dataset as NetCDFFile |
|---|
| 8 | |
|---|
| 9 | main = 'read_ISD.py' |
|---|
| 10 | errormsg = 'ERROR -- error -- ERROR -- error' |
|---|
| 11 | warnmsg = 'WARNING -- warning -- WARNING -- warning' |
|---|
| 12 | |
|---|
| 13 | # version |
|---|
| 14 | version=1.1 |
|---|
| 15 | |
|---|
| 16 | # Filling values for floats, integer and string |
|---|
| 17 | fillValueF = 1.e20 |
|---|
| 18 | fillValueI = -99999 |
|---|
| 19 | fillValueS = '---' |
|---|
| 20 | |
|---|
| 21 | # Length of the string variables |
|---|
| 22 | StringLength = 200 |
|---|
| 23 | |
|---|
| 24 | ####### Different fixed tabuled values |
|---|
| 25 | |
|---|
| 26 | quality_vals = ['Passed gross limits check', 'Passed all quality control checks', \ |
|---|
| 27 | 'Suspect', 'Erroneous', \ |
|---|
| 28 | 'Passed gross limits check , data originate from an NCDC data source', \ |
|---|
| 29 | 'Passed all quality control checks, data originate from an NCDC data source', \ |
|---|
| 30 | 'Suspect, data originate from an NCDC data source', \ |
|---|
| 31 | 'Erroneous, data originate from an NCDC data source', \ |
|---|
| 32 | 'Passed gross limits check if element is present' ] |
|---|
| 33 | |
|---|
| 34 | def searchInlist(listname, nameFind): |
|---|
| 35 | """ Function to search a value within a list |
|---|
| 36 | listname = list |
|---|
| 37 | nameFind = value to find |
|---|
| 38 | >>> searInlist(['1', '2', '3', '5'], '5') |
|---|
| 39 | True |
|---|
| 40 | """ |
|---|
| 41 | for x in listname: |
|---|
| 42 | if x == nameFind: |
|---|
| 43 | return True |
|---|
| 44 | return False |
|---|
| 45 | |
|---|
| 46 | def set_attribute(ncvar, attrname, attrvalue): |
|---|
| 47 | """ Sets a value of an attribute of a netCDF variable. Removes previous attribute value if exists |
|---|
| 48 | ncvar = object netcdf variable |
|---|
| 49 | attrname = name of the attribute |
|---|
| 50 | attrvalue = value of the attribute |
|---|
| 51 | """ |
|---|
| 52 | import numpy as np |
|---|
| 53 | from netCDF4 import Dataset as NetCDFFile |
|---|
| 54 | |
|---|
| 55 | attvar = ncvar.ncattrs() |
|---|
| 56 | if searchInlist(attvar, attrname): |
|---|
| 57 | attr = ncvar.delncattr(attrname) |
|---|
| 58 | |
|---|
| 59 | attr = ncvar.setncattr(attrname, attrvalue) |
|---|
| 60 | |
|---|
| 61 | return ncvar |
|---|
| 62 | |
|---|
| 63 | def basicvardef(varobj, vstname, vlname, vunits): |
|---|
| 64 | """ Function to give the basic attributes to a variable |
|---|
| 65 | varobj= netCDF variable object |
|---|
| 66 | vstname= standard name of the variable |
|---|
| 67 | vlname= long name of the variable |
|---|
| 68 | vunits= units of the variable |
|---|
| 69 | """ |
|---|
| 70 | attr = varobj.setncattr('standard_name', vstname) |
|---|
| 71 | attr = varobj.setncattr('long_name', vlname) |
|---|
| 72 | attr = varobj.setncattr('units', vunits) |
|---|
| 73 | |
|---|
| 74 | return |
|---|
| 75 | |
|---|
| 76 | def remove_NONascii(string): |
|---|
| 77 | """ Function to remove that characters which are not in the standard 127 ASCII |
|---|
| 78 | string= string to transform |
|---|
| 79 | >>> remove_NONascii('LluÃs') |
|---|
| 80 | Lluis |
|---|
| 81 | """ |
|---|
| 82 | fname = 'remove_NONascii' |
|---|
| 83 | |
|---|
| 84 | newstring = string |
|---|
| 85 | |
|---|
| 86 | RTFchar= ['á', 'é', 'Ã', 'ó', 'ú', 'à ', 'Ú', 'ì', 'ò', 'ù', 'â', 'ê', 'î', 'ÃŽ', \ |
|---|
| 87 | 'û', 'À', 'ë', 'ï', 'ö', 'ÃŒ', 'ç', 'ñ','Ê', 'Å', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', \ |
|---|
| 88 | 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã', 'Ã',\ |
|---|
| 89 | 'Ã', 'Å', '\n', '\t'] |
|---|
| 90 | ASCchar= ['a', 'e', 'i', 'o', 'u', 'a', 'e', 'i', 'o', 'u', 'a', 'e', 'i', 'o', \ |
|---|
| 91 | 'u', 'a', 'e', 'i', 'o', 'u', 'c', 'n','ae', 'oe', 'A', 'E', 'I', 'O', 'U', 'A', \ |
|---|
| 92 | 'E', 'I', 'O', 'U', 'A', 'E', 'I', 'O', 'U', 'A', 'E', 'I', 'O', 'U', 'C', 'N',\ |
|---|
| 93 | 'AE', 'OE', '', ' '] |
|---|
| 94 | |
|---|
| 95 | Nchars = len(RTFchar) |
|---|
| 96 | for ichar in range(Nchars): |
|---|
| 97 | foundchar = string.find(RTFchar[ichar]) |
|---|
| 98 | if foundchar != -1: |
|---|
| 99 | newstring = newstring.replace(RTFchar[ichar], ASCchar[ichar]) |
|---|
| 100 | |
|---|
| 101 | return newstring |
|---|
| 102 | |
|---|
| 103 | def read_description(fdobs, dbg): |
|---|
| 104 | """ reads the description file of the observational data-set |
|---|
| 105 | fdobs= descriptive observational data-set |
|---|
| 106 | dbg= boolean argument for debugging |
|---|
| 107 | * Each station should have a 'description.dat' file with: |
|---|
| 108 | institution=Institution who creates the data |
|---|
| 109 | department=Department within the institution |
|---|
| 110 | scientists=names of the data producers |
|---|
| 111 | contact=contact of the data producers |
|---|
| 112 | description=description of the observations |
|---|
| 113 | acknowledgement=sentence of acknowlegement |
|---|
| 114 | comment=comment for the measurements |
|---|
| 115 | |
|---|
| 116 | MissingValue='|' list of ASCII values for missing values within the data |
|---|
| 117 | (as they appear!) |
|---|
| 118 | comment=comments |
|---|
| 119 | |
|---|
| 120 | varN='|' list of variable names |
|---|
| 121 | varLN='|' list of long variable names |
|---|
| 122 | varU='|' list units of the variables |
|---|
| 123 | varBUFR='|' list BUFR code of the variables |
|---|
| 124 | varTYPE='|' list of variable types ('D', 'F', 'I', 'I64', 'S') |
|---|
| 125 | varOPER='|' list of operations to do to the variables to meet their units ([oper],[val]) |
|---|
| 126 | [oper]=-,sumc,subc,mulc,divc (nothing, add, rest, multiply and divide) |
|---|
| 127 | varL='|' list of length of the sections for each variable within a given line (Fortran-like) |
|---|
| 128 | |
|---|
| 129 | NAMElon=name of the variable with the longitude (x position) |
|---|
| 130 | NAMElat=name of the variable with the latitude (y position) |
|---|
| 131 | NAMEheight=ind_alt |
|---|
| 132 | NAMEtime=name of the varibale with the time |
|---|
| 133 | FMTtime=format of the time (as in 'C', 'CFtime' for already CF-like time) |
|---|
| 134 | |
|---|
| 135 | """ |
|---|
| 136 | fname = 'read_description' |
|---|
| 137 | |
|---|
| 138 | descobj = open(fdobs, 'r') |
|---|
| 139 | desc = {} |
|---|
| 140 | |
|---|
| 141 | namevalues = [] |
|---|
| 142 | |
|---|
| 143 | for line in descobj: |
|---|
| 144 | if line[0:1] != '#' and len(line) > 1 : |
|---|
| 145 | descn = remove_NONascii(line.split('=')[0]) |
|---|
| 146 | descv = remove_NONascii(line.split('=')[1]) |
|---|
| 147 | namevalues.append(descn) |
|---|
| 148 | if descn[0:3] != 'var': |
|---|
| 149 | if descn != 'MissingValue': |
|---|
| 150 | desc[descn] = descv |
|---|
| 151 | else: |
|---|
| 152 | desc[descn] = [] |
|---|
| 153 | for dn in descv.split('|'): desc[descn].append(dn) |
|---|
| 154 | print ' ' + fname + ': missing values found:',desc[descn] |
|---|
| 155 | elif descn[0:3] == 'var': |
|---|
| 156 | desc[descn] = descv.split('|') |
|---|
| 157 | elif descn[0:4] == 'NAME': |
|---|
| 158 | desc[descn] = descv |
|---|
| 159 | elif descn[0:3] == 'FMT': |
|---|
| 160 | desc[descn] = descv |
|---|
| 161 | |
|---|
| 162 | if dbg: |
|---|
| 163 | Nvars = len(desc['varN']) |
|---|
| 164 | print ' ' + fname + ": description content of '" + fdobs + "'______________" |
|---|
| 165 | for varn in namevalues: |
|---|
| 166 | if varn[0:3] != 'var': |
|---|
| 167 | print ' ' + varn + ':',desc[varn] |
|---|
| 168 | elif varn == 'varN': |
|---|
| 169 | print ' * Variables:' |
|---|
| 170 | for ivar in range(Nvars): |
|---|
| 171 | varname = desc['varN'][ivar] |
|---|
| 172 | varLname = desc['varLN'][ivar] |
|---|
| 173 | varunits = desc['varU'][ivar] |
|---|
| 174 | varsec = int(desc['varL'][ivar]) |
|---|
| 175 | if desc.has_key('varBUFR'): |
|---|
| 176 | varbufr = desc['varBUFR'][ivar] |
|---|
| 177 | else: |
|---|
| 178 | varbufr = None |
|---|
| 179 | if desc['varOPER'] is not None: |
|---|
| 180 | opv = desc['varOPER'][ivar] |
|---|
| 181 | else: |
|---|
| 182 | opv = None |
|---|
| 183 | print ' ', ivar, varname+':',varLname,'[',varunits, \ |
|---|
| 184 | ']','bufr code:',varbufr,'oper:',opv |
|---|
| 185 | |
|---|
| 186 | descobj.close() |
|---|
| 187 | |
|---|
| 188 | return desc |
|---|
| 189 | |
|---|
| 190 | def value_fmt(val, miss, op, fmt): |
|---|
| 191 | """ Function to transform an ASCII value to a given format |
|---|
| 192 | val= value to transform |
|---|
| 193 | miss= list of possible missing values |
|---|
| 194 | op= operation to perform to the value |
|---|
| 195 | fmt= format to take: |
|---|
| 196 | 'D': float double precission |
|---|
| 197 | 'F': float |
|---|
| 198 | 'I': integer |
|---|
| 199 | 'I64': 64-bits integer |
|---|
| 200 | 'S': string |
|---|
| 201 | >>> value_fmt('9876.12', '-999', 'F') |
|---|
| 202 | 9876.12 |
|---|
| 203 | """ |
|---|
| 204 | |
|---|
| 205 | fname = 'value_fmt' |
|---|
| 206 | |
|---|
| 207 | aopers = ['sumc','subc','mulc','divc'] |
|---|
| 208 | |
|---|
| 209 | fmts = ['D', 'F', 'I', 'I64', 'S'] |
|---|
| 210 | Nfmts = len(fmts) |
|---|
| 211 | |
|---|
| 212 | if not searchInlist(miss,val): |
|---|
| 213 | if searchInlist(miss,'empty') and len(val) == 0: |
|---|
| 214 | newval = None |
|---|
| 215 | else: |
|---|
| 216 | if op != '-': |
|---|
| 217 | opern = op.split(',')[0] |
|---|
| 218 | operv = np.float(op.split(',')[1]) |
|---|
| 219 | |
|---|
| 220 | if not searchInlist(aopers,opern): |
|---|
| 221 | print errormsg |
|---|
| 222 | print ' ' + fname + ": operation '" + opern + "' not ready!!" |
|---|
| 223 | print ' availables:',aopers |
|---|
| 224 | quit(-1) |
|---|
| 225 | else: |
|---|
| 226 | opern = 'sumc' |
|---|
| 227 | operv = 0. |
|---|
| 228 | |
|---|
| 229 | if not searchInlist(fmts, fmt): |
|---|
| 230 | print errormsg |
|---|
| 231 | print ' ' + fname + ": format '" + fmt + "' not ready !!" |
|---|
| 232 | quit(-1) |
|---|
| 233 | else: |
|---|
| 234 | if fmt == 'D': |
|---|
| 235 | opv = np.float32(operv) |
|---|
| 236 | if opern == 'sumc': newval = np.float32(val) + opv |
|---|
| 237 | elif opern == 'subc': newval = np.float32(val) - opv |
|---|
| 238 | elif opern == 'mulc': newval = np.float32(val) * opv |
|---|
| 239 | elif opern == 'divc': newval = np.float32(val) / opv |
|---|
| 240 | elif fmt == 'F': |
|---|
| 241 | opv = np.float(operv) |
|---|
| 242 | if opern == 'sumc': newval = np.float(val) + opv |
|---|
| 243 | elif opern == 'subc': newval = np.float(val) - opv |
|---|
| 244 | elif opern == 'mulc': newval = np.float(val) * opv |
|---|
| 245 | elif opern == 'divc': newval = np.float(val) / opv |
|---|
| 246 | elif fmt == 'I': |
|---|
| 247 | opv = int(operv) |
|---|
| 248 | if opern == 'sumc': newval = int(val) + opv |
|---|
| 249 | elif opern == 'subc': newval = int(val) - opv |
|---|
| 250 | elif opern == 'mulc': newval = int(val) * opv |
|---|
| 251 | elif opern == 'divc': newval = int(val) / opv |
|---|
| 252 | elif fmt == 'I64': |
|---|
| 253 | opv = np.int64(operv) |
|---|
| 254 | if opern == 'sumc': newval = np.int64(val) + opv |
|---|
| 255 | elif opern == 'subc': newval = np.int64(val) - opv |
|---|
| 256 | elif opern == 'mulc': newval = np.int64(val) * opv |
|---|
| 257 | elif opern == 'divc': newval = np.int64(val) / opv |
|---|
| 258 | elif fmt == 'S': |
|---|
| 259 | newval = val |
|---|
| 260 | else: |
|---|
| 261 | newval = None |
|---|
| 262 | |
|---|
| 263 | return newval |
|---|
| 264 | |
|---|
| 265 | def writing_str_nc(varo, values, Lchar): |
|---|
| 266 | """ Function to write string values in a netCDF variable as a chain of 1char values |
|---|
| 267 | varo= netCDF variable object |
|---|
| 268 | values = list of values to introduce |
|---|
| 269 | Lchar = length of the string in the netCDF file |
|---|
| 270 | """ |
|---|
| 271 | |
|---|
| 272 | Nvals = len(values) |
|---|
| 273 | for iv in range(Nvals): |
|---|
| 274 | if values[iv] is None: |
|---|
| 275 | stringv = 'None' |
|---|
| 276 | else: |
|---|
| 277 | stringv = values[iv] |
|---|
| 278 | charvals = np.chararray(Lchar) |
|---|
| 279 | Lstr = len(stringv) |
|---|
| 280 | charvals[Lstr:Lchar] = '' |
|---|
| 281 | |
|---|
| 282 | for ich in range(Lstr): |
|---|
| 283 | charvals[ich] = stringv[ich:ich+1] |
|---|
| 284 | |
|---|
| 285 | varo[iv,:] = charvals |
|---|
| 286 | |
|---|
| 287 | return |
|---|
| 288 | |
|---|
| 289 | def Stringtimes_CF(tvals, fmt, Srefdate, tunits, dbg): |
|---|
| 290 | """ Function to transform a given data in String formats to a CF date |
|---|
| 291 | tvals= string temporal values |
|---|
| 292 | fmt= format of the the time values |
|---|
| 293 | Srefdate= reference date in [YYYY][MM][DD][HH][MI][SS] format |
|---|
| 294 | tunits= units to use ('weeks', 'days', 'hours', 'minutes', 'seconds') |
|---|
| 295 | dbg= debug |
|---|
| 296 | >>> Stringtimes_CF(['19760217082712','20150213101837'], '%Y%m%d%H%M%S', |
|---|
| 297 | '19491201000000', 'hours', False) |
|---|
| 298 | [229784.45333333 571570.31027778] |
|---|
| 299 | """ |
|---|
| 300 | import datetime as dt |
|---|
| 301 | |
|---|
| 302 | fname = 'Stringtimes' |
|---|
| 303 | |
|---|
| 304 | dimt = len(tvals) |
|---|
| 305 | |
|---|
| 306 | yrref = int(Srefdate[0:4]) |
|---|
| 307 | monref = int(Srefdate[4:6]) |
|---|
| 308 | dayref = int(Srefdate[6:8]) |
|---|
| 309 | horref = int(Srefdate[8:10]) |
|---|
| 310 | minref = int(Srefdate[10:12]) |
|---|
| 311 | secref = int(Srefdate[12:14]) |
|---|
| 312 | refdate = dt.datetime( yrref, monref, dayref, horref, minref, secref) |
|---|
| 313 | |
|---|
| 314 | cftimes = np.zeros((dimt), dtype=np.float) |
|---|
| 315 | |
|---|
| 316 | Nfmt=len(fmt.split('%')) |
|---|
| 317 | |
|---|
| 318 | if dbg: print ' ' + fname + ': fmt=',fmt,'refdate:',Srefdate,'uits:',tunits, \ |
|---|
| 319 | 'date dt_days dt_time deltaseconds CFtime _______' |
|---|
| 320 | for it in range(dimt): |
|---|
| 321 | |
|---|
| 322 | # Removing excess of mili-seconds (up to 6 decimals) |
|---|
| 323 | if fmt.split('%')[Nfmt-1] == 'f': |
|---|
| 324 | tpoints = tvals[it].split('.') |
|---|
| 325 | if len(tpoints[len(tpoints)-1]) > 6: |
|---|
| 326 | milisec = '{0:.6f}'.format(np.float('0.'+tpoints[len(tpoints)-1]))[0:7] |
|---|
| 327 | newtval = '' |
|---|
| 328 | for ipt in range(len(tpoints)-1): |
|---|
| 329 | newtval = newtval + tpoints[ipt] + '.' |
|---|
| 330 | newtval = newtval + str(milisec)[2:len(milisec)+1] |
|---|
| 331 | else: |
|---|
| 332 | newtval = tvals[it] |
|---|
| 333 | tval = dt.datetime.strptime(newtval, fmt) |
|---|
| 334 | else: |
|---|
| 335 | tval = dt.datetime.strptime(tvals[it], fmt) |
|---|
| 336 | |
|---|
| 337 | deltatime = tval - refdate |
|---|
| 338 | deltaseconds = deltatime.days*24.*3600. + deltatime.seconds + \ |
|---|
| 339 | deltatime.microseconds/100000. |
|---|
| 340 | if tunits == 'weeks': |
|---|
| 341 | deltat = 7.*24.*3600. |
|---|
| 342 | elif tunits == 'days': |
|---|
| 343 | deltat = 24.*3600. |
|---|
| 344 | elif tunits == 'hours': |
|---|
| 345 | deltat = 3600. |
|---|
| 346 | elif tunits == 'minutes': |
|---|
| 347 | deltat = 60. |
|---|
| 348 | elif tunits == 'seconds': |
|---|
| 349 | deltat = 1. |
|---|
| 350 | else: |
|---|
| 351 | print errormsg |
|---|
| 352 | print ' ' + fname + ": time units '" + tunits + "' not ready !!" |
|---|
| 353 | quit(-1) |
|---|
| 354 | |
|---|
| 355 | cftimes[it] = deltaseconds / deltat |
|---|
| 356 | if dbg: |
|---|
| 357 | print ' ' + tvals[it], deltatime, deltaseconds, cftimes[it] |
|---|
| 358 | |
|---|
| 359 | return cftimes |
|---|
| 360 | |
|---|
| 361 | def additional_vars(string): |
|---|
| 362 | """ Function for the additional variables |
|---|
| 363 | string: sectino of the file line which has to be evaluated |
|---|
| 364 | """ |
|---|
| 365 | fname = 'additional_vars' |
|---|
| 366 | |
|---|
| 367 | Lstring=len(string) |
|---|
| 368 | |
|---|
| 369 | # Generic list of additional variables '-' for range of values |
|---|
| 370 | addvars0 = ['AA1-4', 'AB1', 'AC1', 'AD1', 'AE1', 'AG1', 'AH1-6', 'AI1-6', 'AJ1', \ |
|---|
| 371 | 'AK1', 'AL1-4', 'AM1', 'AN1', 'AO1-4', 'AP1-4', 'HPD', 'AU1-9', 'AW1-4', \ |
|---|
| 372 | 'AX1-6', 'AY1-2', 'AZ1-2', 'CB1-2', 'CF1-3', 'CG1-3', 'CH1-2', 'CI1', 'CN1-4', \ |
|---|
| 373 | 'CO1-9', 'CR1', 'CT1-3', 'CU1-3', 'CV1-3', 'CW1-3', 'CX1-3', 'ED1', 'GA1-6', \ |
|---|
| 374 | 'GE1', 'GF1', 'GG1-6', 'GH1', 'GJ1', 'GK1', 'GM1', 'GN1', 'GO1', 'GP1', 'GQ1', \ |
|---|
| 375 | 'GR1', 'HL1', 'IA1-2', 'IB1-2', 'IC1', 'KA1-4', 'KB1-3', 'KC1-2', 'KD1-2', \ |
|---|
| 376 | 'KE1', 'KF1', 'KG1-2', 'MA1', 'MD1', 'ME1', 'MF1', 'MG1', 'MH1', 'MK1', \ |
|---|
| 377 | 'MV1-7', 'MW1-7', 'OA1-3', 'OB1-2', 'OC1', 'OD1-3', 'OE1-3', 'RH1-3', 'SA1', \ |
|---|
| 378 | 'ST1', 'UA1', 'UG1-2', 'WA1', 'WD1', 'WG1', 'WJ1', 'REM', 'EQD', 'N01-99', \ |
|---|
| 379 | 'QNN'] |
|---|
| 380 | |
|---|
| 381 | addvars = [] |
|---|
| 382 | for avar in addvars0: |
|---|
| 383 | if avar.find('-') == -1: |
|---|
| 384 | addvars.append(avar) |
|---|
| 385 | else: |
|---|
| 386 | totrange=avar.split('-') |
|---|
| 387 | erange = totrange[1] |
|---|
| 388 | |
|---|
| 389 | if len(erange) > 1: |
|---|
| 390 | irange = totrange[0][1:3] |
|---|
| 391 | else: |
|---|
| 392 | irange = totrange[0][2:3] |
|---|
| 393 | |
|---|
| 394 | for ir in range(int(irange), int(erange)+1): |
|---|
| 395 | if len(erange) > 1: |
|---|
| 396 | addvars.append(avar[0:1] + str(ir).zfill(2)) |
|---|
| 397 | else: |
|---|
| 398 | addvars.append(avar[0:2] + str(ir)) |
|---|
| 399 | |
|---|
| 400 | # print addvars |
|---|
| 401 | # Looking for the variables |
|---|
| 402 | ichar=0 |
|---|
| 403 | |
|---|
| 404 | additionals = [] |
|---|
| 405 | while ichar <= Lstring: |
|---|
| 406 | sec3 = string[ichar:ichar+3] |
|---|
| 407 | if searchInlist(addvars, sec3): |
|---|
| 408 | additionals.append(sec3) |
|---|
| 409 | # print ' ' , sec3, 'Additional!!' |
|---|
| 410 | ichar = ichar + 2 |
|---|
| 411 | |
|---|
| 412 | ichar = ichar + 1 |
|---|
| 413 | |
|---|
| 414 | return additionals |
|---|
| 415 | |
|---|
| 416 | def read_datavalues_conline(dataf, comchar, fmt, oper, Lv, miss, varns, dbg): |
|---|
| 417 | """ Function to read from an ASCII file values in ISH format (continuos line) |
|---|
| 418 | dataf= data file |
|---|
| 419 | comchar= list of the characters indicating comment in the file |
|---|
| 420 | dbg= debug mode or not |
|---|
| 421 | fmt= list of kind of values to be found |
|---|
| 422 | oper= list of operations to perform |
|---|
| 423 | Lv= length of the value section |
|---|
| 424 | miss= missing value |
|---|
| 425 | varns= list of name of the variables to find |
|---|
| 426 | """ |
|---|
| 427 | fname = 'read_datavalues_conline' |
|---|
| 428 | |
|---|
| 429 | ofile = open(dataf, 'r') |
|---|
| 430 | Nvals = len(fmt) |
|---|
| 431 | |
|---|
| 432 | if oper is None: |
|---|
| 433 | opers = [] |
|---|
| 434 | for ioper in range(Nvals): |
|---|
| 435 | opers.append('-') |
|---|
| 436 | else: |
|---|
| 437 | opers = oper |
|---|
| 438 | |
|---|
| 439 | finalvalues = {} |
|---|
| 440 | |
|---|
| 441 | iline = 0 |
|---|
| 442 | for line in ofile: |
|---|
| 443 | line = line.replace('\n','').replace(chr(13),'') |
|---|
| 444 | Lline = len(line) |
|---|
| 445 | if not searchInlist(comchar,line[0:1]) and len(line) > 1: |
|---|
| 446 | # Removing no-value columns |
|---|
| 447 | values = [] |
|---|
| 448 | |
|---|
| 449 | for ivar in range(Nvals): |
|---|
| 450 | if ivar == 0: |
|---|
| 451 | isec = 0 |
|---|
| 452 | else: |
|---|
| 453 | isec = int(Lv[ivar-1]) |
|---|
| 454 | |
|---|
| 455 | esec = int(Lv[ivar]) |
|---|
| 456 | val = line[isec:esec] |
|---|
| 457 | |
|---|
| 458 | values.append(val) |
|---|
| 459 | if dbg: |
|---|
| 460 | print iline, varns[ivar],'value:',values[ivar],miss,opers[ivar], \ |
|---|
| 461 | fmt[ivar] |
|---|
| 462 | |
|---|
| 463 | if iline == 0: |
|---|
| 464 | listvals = [] |
|---|
| 465 | listvals.append(value_fmt(values[ivar], miss, opers[ivar], \ |
|---|
| 466 | fmt[ivar])) |
|---|
| 467 | finalvalues[varns[ivar]] = listvals |
|---|
| 468 | else: |
|---|
| 469 | listvals = finalvalues[varns[ivar]] |
|---|
| 470 | listvals.append(value_fmt(values[ivar], miss, opers[ivar], \ |
|---|
| 471 | fmt[ivar])) |
|---|
| 472 | finalvalues[varns[ivar]] = listvals |
|---|
| 473 | # Additional variables |
|---|
| 474 | if Lline > esec and line[esec:esec+3] == 'ADD': |
|---|
| 475 | if iline == 0: print ' ' + fname + ': Additional variables!!' |
|---|
| 476 | ichar=esec+3 |
|---|
| 477 | print ' ', additional_vars(line[esec+4:Lline+1]) |
|---|
| 478 | |
|---|
| 479 | else: |
|---|
| 480 | # First line without values |
|---|
| 481 | if iline == 0: iline = -1 |
|---|
| 482 | |
|---|
| 483 | iline = iline + 1 |
|---|
| 484 | |
|---|
| 485 | ofile.close() |
|---|
| 486 | |
|---|
| 487 | return finalvalues |
|---|
| 488 | |
|---|
| 489 | def adding_station_desc(onc,stdesc): |
|---|
| 490 | """ Function to add a station description in a netCDF file |
|---|
| 491 | onc= netCDF object |
|---|
| 492 | stdesc= station description lon, lat, height |
|---|
| 493 | """ |
|---|
| 494 | fname = 'adding_station_desc' |
|---|
| 495 | |
|---|
| 496 | newvar = onc.createVariable( 'station', 'c', ('StrLength')) |
|---|
| 497 | newvar[0:len(stdesc[0])] = stdesc[0].replace('!', ' ') |
|---|
| 498 | |
|---|
| 499 | newdim = onc.createDimension('nst',1) |
|---|
| 500 | |
|---|
| 501 | newvar = objfile.createVariable( 'lonstGDM', 'c', ('nst','StrLength')) |
|---|
| 502 | Gv = int(stdesc[1]) |
|---|
| 503 | Dv = int((stdesc[1] - Gv)*60.) |
|---|
| 504 | Mv = int((stdesc[1] - Gv - Dv/60.)*3600.) |
|---|
| 505 | writing_str_nc(newvar, [str(Gv)+"d" + str(Dv)+"m" + str(Mv)+'s'], StringLength) |
|---|
| 506 | |
|---|
| 507 | if onc.variables.has_key('lon'): |
|---|
| 508 | print warnmsg |
|---|
| 509 | print ' ' + fname + ": variable 'lon' already exist !!" |
|---|
| 510 | print " renaming it as 'lonst'" |
|---|
| 511 | lonname = 'lonst' |
|---|
| 512 | else: |
|---|
| 513 | lonname = 'lon' |
|---|
| 514 | |
|---|
| 515 | newvar = onc.createVariable( lonname, 'f4', ('nst')) |
|---|
| 516 | basicvardef(newvar, lonname, 'longitude', 'degrees_West' ) |
|---|
| 517 | newvar[:] = stdesc[1] |
|---|
| 518 | |
|---|
| 519 | newvar = objfile.createVariable( 'latstGDM', 'c', ('nst','StrLength')) |
|---|
| 520 | Gv = int(stdesc[2]) |
|---|
| 521 | Dv = int((stdesc[2] - Gv)*60.) |
|---|
| 522 | Mv = int((stdesc[2] - Gv - Dv/60.)*3600.) |
|---|
| 523 | writing_str_nc(newvar, [str(Gv)+"d" + str(Dv)+"m" + str(Mv)+'s'], StringLength) |
|---|
| 524 | |
|---|
| 525 | if onc.variables.has_key('lat'): |
|---|
| 526 | print warnmsg |
|---|
| 527 | print ' ' + fname + ": variable 'lat' already exist !!" |
|---|
| 528 | print " renaming it as 'latst'" |
|---|
| 529 | latname = 'latst' |
|---|
| 530 | else: |
|---|
| 531 | latname = 'lat' |
|---|
| 532 | |
|---|
| 533 | newvar = onc.createVariable( latname, 'f4', ('nst')) |
|---|
| 534 | basicvardef(newvar, lonname, 'latitude', 'degrees_North' ) |
|---|
| 535 | newvar[:] = stdesc[2] |
|---|
| 536 | |
|---|
| 537 | if onc.variables.has_key('height'): |
|---|
| 538 | print warnmsg |
|---|
| 539 | print ' ' + fname + ": variable 'height' already exist !!" |
|---|
| 540 | print " renaming it as 'heightst'" |
|---|
| 541 | heightname = 'heightst' |
|---|
| 542 | else: |
|---|
| 543 | heightname = 'height' |
|---|
| 544 | |
|---|
| 545 | newvar = onc.createVariable( heightname, 'f4', ('nst')) |
|---|
| 546 | basicvardef(newvar, heightname, 'height above sea level', 'm' ) |
|---|
| 547 | newvar[:] = stdesc[3] |
|---|
| 548 | |
|---|
| 549 | return |
|---|
| 550 | |
|---|
| 551 | ####### ###### ##### #### ### ## # |
|---|
| 552 | |
|---|
| 553 | strCFt="Refdate,tunits (CF reference date [YYYY][MM][DD][HH][MI][SS] format and " + \ |
|---|
| 554 | " and time units: 'weeks', 'days', 'hours', 'miuntes', 'seconds')" |
|---|
| 555 | |
|---|
| 556 | parser = OptionParser() |
|---|
| 557 | parser.add_option("-c", "--comments", dest="charcom", |
|---|
| 558 | help="':', list of characters used for comments", metavar="VALUES") |
|---|
| 559 | parser.add_option("-d", "--descriptionfile", dest="fdesc", |
|---|
| 560 | help="description file to use", metavar="FILE") |
|---|
| 561 | parser.add_option("-f", "--file", dest="obsfile", |
|---|
| 562 | help="observational file to use", metavar="FILE") |
|---|
| 563 | parser.add_option("-g", "--debug", dest="debug", |
|---|
| 564 | help="whther debug is required ('false', 'true')", metavar="VALUE") |
|---|
| 565 | parser.add_option("-s", "--stationLocation", dest="stloc", |
|---|
| 566 | help="',' list with Name, longitude, latitude and height of the station", metavar="VALUES") |
|---|
| 567 | parser.add_option("-t", "--CFtime", dest="CFtime", help=strCFt, metavar="VALUE") |
|---|
| 568 | (opts, args) = parser.parse_args() |
|---|
| 569 | |
|---|
| 570 | ####### ####### |
|---|
| 571 | ## MAIN |
|---|
| 572 | ####### |
|---|
| 573 | |
|---|
| 574 | ofile='ISD.nc' |
|---|
| 575 | |
|---|
| 576 | if opts.charcom is None: |
|---|
| 577 | print warnmsg |
|---|
| 578 | print ' ' + main + ': No list of comment characters provided!!' |
|---|
| 579 | print ' assuming no need!' |
|---|
| 580 | charcomments = [] |
|---|
| 581 | else: |
|---|
| 582 | charcomments = opts.charcom.split(':') |
|---|
| 583 | |
|---|
| 584 | if opts.fdesc is None: |
|---|
| 585 | print errormsg |
|---|
| 586 | print ' ' + main + ': No description file for the observtional data provided!!' |
|---|
| 587 | quit(-1) |
|---|
| 588 | |
|---|
| 589 | if opts.obsfile is None: |
|---|
| 590 | print errormsg |
|---|
| 591 | print ' ' + main + ': ISD file not provided !!' |
|---|
| 592 | quit(-1) |
|---|
| 593 | |
|---|
| 594 | if opts.debug is None: |
|---|
| 595 | print warnmsg |
|---|
| 596 | print ' ' + main + ': No debug provided!!' |
|---|
| 597 | print " assuming 'False'" |
|---|
| 598 | debug = False |
|---|
| 599 | else: |
|---|
| 600 | if opts.debug == 'true': |
|---|
| 601 | debug = True |
|---|
| 602 | else: |
|---|
| 603 | debug = False |
|---|
| 604 | |
|---|
| 605 | if not os.path.isfile(opts.fdesc): |
|---|
| 606 | print errormsg |
|---|
| 607 | print ' ' + main + ": description file '" + opts.fdesc + "' does not exist !!" |
|---|
| 608 | quit(-1) |
|---|
| 609 | |
|---|
| 610 | if not os.path.isfile(opts.obsfile): |
|---|
| 611 | print errormsg |
|---|
| 612 | print ' ' + main + ": observational file '" + opts.obsfile + "' does not exist !!" |
|---|
| 613 | quit(-1) |
|---|
| 614 | |
|---|
| 615 | if opts.CFtime is None: |
|---|
| 616 | print warnmsg |
|---|
| 617 | print ' ' + main + ': No CFtime criteria are provided !!' |
|---|
| 618 | print " either time is already in CF-format ('timeFMT=CFtime') in '" + \ |
|---|
| 619 | opts.fdesc + "'" |
|---|
| 620 | print " or assuming refdate: '19491201000000' and time units: 'hours'" |
|---|
| 621 | referencedate = '19491201000000' |
|---|
| 622 | timeunits = 'hours' |
|---|
| 623 | else: |
|---|
| 624 | referencedate = opts.CFtime.split(',')[0] |
|---|
| 625 | timeunits = opts.CFtime.split(',')[1] |
|---|
| 626 | |
|---|
| 627 | if opts.stloc is None: |
|---|
| 628 | print errornmsg |
|---|
| 629 | print ' ' + main + ': No station location provided !!' |
|---|
| 630 | quit(-1) |
|---|
| 631 | else: |
|---|
| 632 | print opts.stloc.split(',') |
|---|
| 633 | stationdesc = [opts.stloc.split(',')[0], np.float(opts.stloc.split(',')[1]), \ |
|---|
| 634 | np.float(opts.stloc.split(',')[2]), np.float(opts.stloc.split(',')[3])] |
|---|
| 635 | |
|---|
| 636 | # Reading description file |
|---|
| 637 | ## |
|---|
| 638 | description = read_description(opts.fdesc, debug) |
|---|
| 639 | |
|---|
| 640 | Nvariables=len(description['varN']) |
|---|
| 641 | formats = description['varTYPE'] |
|---|
| 642 | |
|---|
| 643 | if len(formats) != Nvariables: |
|---|
| 644 | print errormsg |
|---|
| 645 | print ' ' + main + ': number of formats:',len(formats),' and number of ' + \ |
|---|
| 646 | 'variables', Nvariables,' does not coincide!!' |
|---|
| 647 | print ' * what is found is _______' |
|---|
| 648 | if Nvariables > len(formats): |
|---|
| 649 | Nshow = len(formats) |
|---|
| 650 | for ivar in range(Nshow): |
|---|
| 651 | print ' ',description['varN'][ivar],':', description['varLN'][ivar],\ |
|---|
| 652 | '[', description['varU'][ivar], '] fmt:', formats[ivar] |
|---|
| 653 | print ' missing values for:', description['varN'][Nshow:Nvariables+1] |
|---|
| 654 | else: |
|---|
| 655 | Nshow = Nvariables |
|---|
| 656 | for ivar in range(Nshow): |
|---|
| 657 | print ' ',description['varN'][ivar],':', description['varLN'][ivar],\ |
|---|
| 658 | '[', description['varU'][ivar], '] fmt:', formats[ivar] |
|---|
| 659 | print ' excess of formats for:', formats[Nshow:len(formats)+1] |
|---|
| 660 | |
|---|
| 661 | quit(-1) |
|---|
| 662 | |
|---|
| 663 | # Reading values |
|---|
| 664 | ## |
|---|
| 665 | datavalues = read_datavalues_conline(opts.obsfile, charcomments, formats, \ |
|---|
| 666 | description['varOPER'], description['varL'], description['MissingValue'], \ |
|---|
| 667 | description['varN'], debug) |
|---|
| 668 | |
|---|
| 669 | # Total number of values |
|---|
| 670 | Ntvalues = len(datavalues[description['varN'][0]]) |
|---|
| 671 | print main + ': total temporal values found:',Ntvalues |
|---|
| 672 | |
|---|
| 673 | objfile = NetCDFFile(ofile, 'w') |
|---|
| 674 | |
|---|
| 675 | # Creation of dimensions |
|---|
| 676 | ## |
|---|
| 677 | objfile.createDimension('time',None) |
|---|
| 678 | objfile.createDimension('StrLength',StringLength) |
|---|
| 679 | |
|---|
| 680 | # Creation of variables |
|---|
| 681 | ## |
|---|
| 682 | for ivar in range(Nvariables): |
|---|
| 683 | varn = description['varN'][ivar] |
|---|
| 684 | print " including: '" + varn + "' ... .. ." |
|---|
| 685 | |
|---|
| 686 | if formats[ivar] == 'D': |
|---|
| 687 | newvar = objfile.createVariable(varn, 'f32', ('time'), fill_value=fillValueF) |
|---|
| 688 | basicvardef(newvar, varn, description['varLN'][ivar], \ |
|---|
| 689 | description['varU'][ivar]) |
|---|
| 690 | vals = np.array(datavalues[varn]) |
|---|
| 691 | for iv in range(Ntvalues): |
|---|
| 692 | if vals[iv] is None: vals[iv] = fillValueF |
|---|
| 693 | newvar[:] = vals |
|---|
| 694 | # newvar[:] = np.where(datavalues[varn] is None, fillValueF, datavalues[varn]) |
|---|
| 695 | elif formats[ivar] == 'F': |
|---|
| 696 | newvar = objfile.createVariable(varn, 'f', ('time'), fill_value=fillValueF) |
|---|
| 697 | basicvardef(newvar, varn, description['varLN'][ivar], \ |
|---|
| 698 | description['varU'][ivar]) |
|---|
| 699 | vals = np.array(datavalues[varn]) |
|---|
| 700 | for iv in range(Ntvalues): |
|---|
| 701 | if vals[iv] is None: vals[iv] = fillValueF |
|---|
| 702 | newvar[:] = vals |
|---|
| 703 | # newvar[:] = np.where(datavalues[varn] is None, fillValueF, datavalues[varn]) |
|---|
| 704 | elif formats[ivar] == 'I': |
|---|
| 705 | newvar = objfile.createVariable(varn, 'i', ('time'), fill_value=fillValueI) |
|---|
| 706 | basicvardef(newvar, varn, description['varLN'][ivar], \ |
|---|
| 707 | description['varU'][ivar]) |
|---|
| 708 | # Why is not wotking with integers? |
|---|
| 709 | vals = np.array(datavalues[varn]) |
|---|
| 710 | for iv in range(Ntvalues): |
|---|
| 711 | if vals[iv] is None: vals[iv] = fillValueI |
|---|
| 712 | newvar[:] = vals |
|---|
| 713 | elif formats[ivar] == 'S': |
|---|
| 714 | newvar = objfile.createVariable(varn, 'c', ('time','StrLength')) |
|---|
| 715 | basicvardef(newvar, varn, description['varLN'][ivar], \ |
|---|
| 716 | description['varU'][ivar]) |
|---|
| 717 | writing_str_nc(newvar, datavalues[varn], StringLength) |
|---|
| 718 | |
|---|
| 719 | # Time variable in CF format |
|---|
| 720 | ## |
|---|
| 721 | if description['FMTtime'] == 'CFtime': |
|---|
| 722 | timevals = datavalues[description['NAMEtime']] |
|---|
| 723 | iv = 0 |
|---|
| 724 | for ivn in description['varN']: |
|---|
| 725 | if ivn == description['NAMEtime']: |
|---|
| 726 | tunits = description['varU'][iv] |
|---|
| 727 | break |
|---|
| 728 | iv = iv + 1 |
|---|
| 729 | else: |
|---|
| 730 | # Time as a composition of different columns |
|---|
| 731 | tcomposite = description['NAMEtime'].find('@') |
|---|
| 732 | if tcomposite != -1: |
|---|
| 733 | timevars = description['NAMEtime'].split('@') |
|---|
| 734 | |
|---|
| 735 | print warnmsg |
|---|
| 736 | print ' ' + main + ': time values as combination of different columns!' |
|---|
| 737 | print ' combining:',timevars,' with a final format: ',description['FMTtime'] |
|---|
| 738 | |
|---|
| 739 | timeSvals = [] |
|---|
| 740 | if debug: print ' ' + main + ': creating times _______' |
|---|
| 741 | for it in range(Ntvalues): |
|---|
| 742 | tSvals = '' |
|---|
| 743 | for tvar in timevars: |
|---|
| 744 | tSvals = tSvals + datavalues[tvar][it] + ' ' |
|---|
| 745 | |
|---|
| 746 | timeSvals.append(tSvals[0:len(tSvals)-1]) |
|---|
| 747 | if debug: print it, '*' + timeSvals[it] + '*' |
|---|
| 748 | |
|---|
| 749 | timevals = Stringtimes_CF(timeSvals, description['FMTtime'], referencedate, \ |
|---|
| 750 | timeunits, debug) |
|---|
| 751 | else: |
|---|
| 752 | timevals = Stringtimes_CF(datavalues[description['NAMEtime']], \ |
|---|
| 753 | description['FMTtime'], referencedate, timeunits, debug) |
|---|
| 754 | |
|---|
| 755 | CFtimeRef = referencedate[0:4] +'-'+ referencedate[4:6] +'-'+ referencedate[6:8] + \ |
|---|
| 756 | ' ' + referencedate[8:10] +':'+ referencedate[10:12] +':'+ referencedate[12:14] |
|---|
| 757 | tunits = timeunits + ' since ' + CFtimeRef |
|---|
| 758 | |
|---|
| 759 | if objfile.variables.has_key('time'): |
|---|
| 760 | print warnmsg |
|---|
| 761 | print ' ' + main + ": variable 'time' already exist !!" |
|---|
| 762 | print " renaming it as 'CFtime'" |
|---|
| 763 | timeCFname = 'CFtime' |
|---|
| 764 | newdim = objfile.renameDimension('time','CFtime') |
|---|
| 765 | newvar = objfile.createVariable( timeCFname, 'f8', ('CFtime')) |
|---|
| 766 | basicvardef(newvar, timeCFname, 'time', tunits ) |
|---|
| 767 | else: |
|---|
| 768 | timeCFname = 'time' |
|---|
| 769 | newvar = objfile.createVariable( timeCFname, 'f8', ('time')) |
|---|
| 770 | basicvardef(newvar, timeCFname, 'time', tunits ) |
|---|
| 771 | |
|---|
| 772 | set_attribute(newvar, 'calendar', 'standard') |
|---|
| 773 | newvar[:] = timevals |
|---|
| 774 | |
|---|
| 775 | # Global attributes |
|---|
| 776 | ## |
|---|
| 777 | for descn in description.keys(): |
|---|
| 778 | if descn[0:3] != 'var' and descn[0:4] != 'NAME' and descn[0:3] != 'FMT': |
|---|
| 779 | string='' |
|---|
| 780 | for sval in description[descn]: |
|---|
| 781 | string = string + str(sval) + ', ' |
|---|
| 782 | set_attribute(objfile, descn, string) |
|---|
| 783 | |
|---|
| 784 | set_attribute(objfile,'author_nc','Lluis Fita') |
|---|
| 785 | set_attribute(objfile,'institution_nc','Laboratoire de Meteorology Dynamique, ' + \ |
|---|
| 786 | 'LMD-Jussieu, UPMC, Paris') |
|---|
| 787 | set_attribute(objfile,'country_nc','France') |
|---|
| 788 | set_attribute(objfile,'script_nc',main) |
|---|
| 789 | set_attribute(objfile,'version_script',version) |
|---|
| 790 | |
|---|
| 791 | # Adding three variables with the station name, location, longitude, latitude and height |
|---|
| 792 | adding_station_desc(objfile,stationdesc) |
|---|
| 793 | |
|---|
| 794 | objfile.sync() |
|---|
| 795 | objfile.close() |
|---|
| 796 | |
|---|
| 797 | print main + ": Successfull generation of netcdf observational file '" + ofile + "' !!" |
|---|