def main():
# Create the domain
domain = Domain()
# Parse command line arguments
parser = argparse.ArgumentParser(description=__doc__)
utils.add_standard_command_options(parser)
utils.add_edb_command_options(parser)
parser.add_argument(
'-F',
action="store_true", dest='force',
help="Force - overwrite target edb"
)
parser.add_argument(
'--delimiter', default='\t',
action="store", dest='delimiter',
help='Delimiter used in enviman csv-tables'
)
parser.add_argument(
'-i',
action="store", dest='indir',
help=("Directory containing enviman tables in .csv format," +
" named the same as Enviman MS Excel sheets, " +
"but with space replaced by '_'")
)
parser.add_argument(
'--substancemap',
action=MappingAction, dest='substancemap',
help=("File mapping substance indices from enviman to Airviro")
)
parser.add_argument(
'--source_srs', metavar="EPSG", type=int,
action='store', dest='source_srs',
help='Coordinate system of enviman data'
)
parser.add_argument(
'--target_srs', metavar="EPSG", type=int,
action='store', dest='target_srs',
help='Coordinate system of Airviro data'
)
args = parser.parse_args()
edb = Edb(domain, args.user, args.edb)
PAG_Source_path = path.join(args.indir, 'PAG_Source.csv')
AnEm_Table_path = path.join(args.indir, 'AnEm_Table.csv')
Mon_Table_path = path.join(args.indir, 'Mon_Table.csv')
Hour_Table_path = path.join(args.indir, 'Hour_Table.csv')
Comb_Table_path = path.join(args.indir, 'Comb_Table.csv')
Comb_Def_path = path.join(args.indir, 'Comb_Def.csv')
subgrp_ind = get_max_subgrp_index(edb) + 1
timevar_ind = get_max_timevar_index(edb) + 1
if path.exists(PAG_Source_path):
sources = read_sources(
PAG_Source_path,
args.delimiter
)
else:
sources = None
log.warning('PAG_Source.csv not found')
if args.source_srs is not None and args.target_srs is not None:
log.debug("Preparing spatial reference transformation")
target_srs = osr.SpatialReference()
source_srs = osr.SpatialReference()
target_srs.ImportFromEPSG(args.target_srs)
source_srs.ImportFromEPSG(args.source_srs)
srs_transform = osr.CoordinateTransformation(source_srs, target_srs)
for sourceid, source in sources.iteritems():
transform(source, srs_transform)
else:
log.debug("No spatial reference transformation will be performed")
if path.exists(AnEm_Table_path) and sources is not None:
read_emissions(
sources,
AnEm_Table_path,
args.delimiter,
args.substancemap
)
else:
log.warning('AnEmi.csv not found')
if path.exists(Comb_Table_path):
subgrps = read_subgrps(
Comb_Table_path,
Comb_Def_path,
args.delimiter,
args.substancemap
)
else:
log.warning('Comb_Table.csv not found')
if path.exists(Hour_Table_path) and path.exists(Mon_Table_path):
timevars = read_timevars(
sources,
Hour_Table_path,
Mon_Table_path,
args.delimiter
)
else:
timevars = None
for key, subgrp in subgrps.iteritems():
subgrp.INDEX = subgrp_ind
subgrp_ind += 1
for key, tvar in timevars.iteritems():
tvar.INDEX = subgrp_ind
timevar_ind += 1
for src in sources.values():
if timevars is not None:
timevarid = '_'.join([src.ALOB['MonthlyID'], src.ALOB['HourlyID']])
timevarind = timevars[timevarid].INDEX
else:
timevarind = 1
unwritten_subgroups = False
if src.ALOB['Consumption'] != '0':
if subgrps is not None:
subgrp = subgrps[int(src.ALOB['ProcessID'])]
src.add_subgrp(
SUBGRP=subgrp.INDEX,
ACTIVITY=float(src.ALOB['Consumption']),
TIMEVAR=timevarind,
UNIT='ton/year'
)
elif not unwritten_subgroups:
log.warning(
"Could not add substance group emissions " +
"due to missing Comb Table"
)
unwritten_subgroups = True
for emis in src.EMISSION:
emis.TIMEVAR = timevarind
if timevars is not None:
with TimevarStream(edb, mode='w', sourcetype='point') as outstream:
log.info('Writing timevars...')
writer = ModelWriter(outstream)
for tvar in timevars.values():
log.debug('Writing timevar %i' % tvar.INDEX)
writer.write(tvar)
if subgrps is not None:
with codecs.open(
'/usr/airviro/data/SCAC/prod/industry/enviman_subgrps.txt',
mode='w',
encoding='HP Roman8') as outstream:
# with SubgrpStream(edb, mode='w') as outstream:
log.info('Writing subgrps...')
writer = ModelWriter(outstream)
for subgrp in subgrps.values():
log.debug('Writing subgrps %i' % subgrp.INDEX)
writer.write(subgrp)
with codecs.open(
'/usr/airviro/data/SCAC/prod/industry/enviman_sources.txt',
mode='w',
encoding='HP Roman8') as outstream:
# with SourceStream(edb, mode='w') as outstream:
log.info('Writing sources...')
writer = ModelWriter(outstream)
for src in sources.values():
log.debug('Writing source %s' % src.NAME)
writer.write(src)
def main():
actionList = [
'addSubstance',
'removeSubstance',
'mapSubstances',
'addCodeTree',
'removeCodeTree',
'moveCodeTree',
'extractByGeocode',
'uniteGrids',
'scale'
]
# setting up parser
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.ArgumentDefaultsHelpFormatter
)
utils.add_standard_command_options(parser)
parser.add_argument("-d", "--doc",
action="store_true", dest="doc",
help="Prints more detailed documentation and exit")
parser.add_argument("-u", "--user",
action="store", dest="user", default=None,
help="Specify user manually")
parser.add_argument("-e", "--edb",
action="store", dest="edb", default=None,
help="Name of target edb")
parser.add_argument("-s", "--substance", action="store",
dest="substance",
help="Name of substance to add or remove")
parser.add_argument("--substanceFactorFile", action="store",
dest="substanceFactorFile",
help="File with substance factors")
parser.add_argument("--substanceMapping", action="store",
dest="substanceMapping",
help="File with mapping for substance indices")
parser.add_argument("--codeMapping", action="store",
dest="codeMapping",
help="File with mapping from old codes to new")
parser.add_argument("--edbMapping", action="store",
dest="edbMapping",
help="File with mapping from gc to edb")
parser.add_argument("-f", "--force",
action="store_true", dest="force", default=False,
help="Start the process without confirming the domain")
parser.add_argument("--factor",
action="store", type=float, dest="factor",
help="Factor used to scale emissions")
parser.add_argument(
"-n", "--names", action="store",
dest="sourceNames", default=['.*'],
nargs='*',
help="Space separated list of regular expressions"
)
parser.add_argument(
"--sourceTypes", action="store",
dest="sourceTypes",
nargs='*',
default=['griddb', 'sources'],
choices=["griddb", "sources", "subgrpdb", "svehdb", "vehdb"],
help="Source types to process"
)
parser.add_argument(
"--codeType", action="store",
dest="codeType", default='ac',
help="Code type to process"
)
parser.add_argument(
"--codeIndex", action="store",
dest="codeIndex", default=1, type=int,
help="Index of code tree to use in processing"
)
parser.add_argument(
"--newIndex", action="store",
dest="newIndex", default=None,
help="New index when moving an already defined code tree"
)
parser.add_argument(
"--codeLevel", action="store",
dest="codeLevel", default=1, type=int,
help="Level of code tree to use in processing")
parser.add_argument(
"--actions", action="store",
dest="actions",
nargs='*',
choices=actionList,
help="Action list"
)
args = parser.parse_args()
if args.doc:
print doc
sys.exit()
# if len(args) > 0:
# log.error("Incorrect number of arguments")
# sys.exit(1)
if args.edb is None:
log.error("Need to specify edb")
sys.exit(1)
if args.user is None:
log.error("Need to specify user")
sys.exit(1)
dmn = Domain()
edb = Edb(dmn, args.user, args.edb)
subdb = Subdb(edb)
subdb.read()
rsrc = Rsrc(edb.rsrcPath())
# Check if edb exist
if not edb.exists():
log.error("Edb " + args.edb + " does not exist for user " +
args.user + " in domain " + dmn.name)
sys.exit(1)
if args.codeType.lower() == "ac":
codeType = "ACTIVITYCODE"
elif args.codeType.lower() == "gc":
codeType = "GEOCODE"
else:
parser.error("codeType should be either 'gc' or 'ac'")
if args.codeIndex < 1:
raise ValueError("Minimum code index is 1")
if args.codeIndex > 1 and codeType == "gc":
raise OSError("Multiple geocode trees not implemented")
for a in args.actions:
if a not in actionList:
parser.error("Unknown action %s" % a)
log.info("User: "******"Edb: " + args.edb)
loadGridData = False
# Set values in argsDict
argDict = {
"edb": edb,
"user": args.user,
"domain": dmn,
"grids": [],
"filters": args.sourceNames,
"codeType": codeType,
"codeIndex": args.codeIndex,
"codeLevel": args.codeLevel,
"sources": None,
"griddb": None,
"subgrpdb": None,
"roaddb": None,
"vehdb": None,
"svehdb": None,
"rsrc": rsrc
}
# ----- Reading sources from edb -------------------------------
if 'griddb' in args.sourceTypes:
grids = []
gridNames = edb.listGrids()
for name in gridNames:
grd = egrid.Egrid(edb, name)
grd.readAsc()
grids.append(grd)
argDict["grids"] = grids
if 'sources' in args.sourceTypes:
argDict["sources"] = list(ModelReader(SourceStream(edb, mode='r')))
if 'roaddb' in args.sourceTypes:
log.error("Not implemented with new roaddb-structure")
sys.exit(1)
if 'subgrpdb' in args.sourceTypes:
subgrpdb = Subgrpdb(edb)
subgrpdb.read()
argDict["subgrpdb"] = subgrpdb
# emfacdbdb = Emfacdb(edb)
# emfacdb.read()
# argDict["subgrpdb"] = subgrpdb
if 'svehdb' in args.sourceTypes:
svehdb = Svehdb(edb)
svehdb.read()
argDict["svehdb"] = svehdb
# TODO: add option for vehdb
# parse additional args
# -----addSubstance--------------------------------------
if "addSubstance" in args.actions:
if args.substance is None:
parser.error("Action addSubstance needs" +
"--substance to be specified")
if args.substanceFactorFile is None:
parser.error("Option addSubstance also needs" +
"--substanceFactorFile to be specified")
if args.substance not in subdb.substIndices:
raise ValueError(
"Substance %s not in substance list" % args.substance)
substanceNameFactorDict = parseMapping(
path.abspath(args.substanceFactorFile),
valueType=float)
substanceFactorDict = {}
# Converts mapping between substance name and factor to
# mapping between substance index and factor
for name, factor in substanceNameFactorDict.iteritems():
if name not in subdb.substIndices:
raise KeyError(
"Substance : %s not found in substance list" % name
)
ind = subdb.substIndices[name]
substanceFactorDict[ind] = factor
argDict["substanceFactorDict"] = substanceFactorDict
argDict["substance"] = subdb.substIndices[args.substance]
# ----removeSubstance------------------------------------
if "removeSubstance" in args.actions:
if args.substance is None:
parser.error("Action removeSubstance needs" +
"--substance to be specified")
if args.substance not in subdb.substIndices:
log.error("Substance %s not in substance list" % args.substance)
sys.exit(1)
argDict["substance"] = subdb.substIndices[args.substance]
# ----mapSubstances--------------------------------------
if "mapSubstances" in args.actions:
if args.substanceMapping is None:
parser.error("Action mapSubstances needs" +
"--substanceMapping to be specified")
substanceMappingFile = path.abspath(args.substanceMapping)
substanceMapping = parseMapping(
substanceMappingFile,
keyType=int,
valueType=int
)
argDict["substanceMapping"] = substanceMapping
# ----addCodeTree-----------------------------------------
if "addCodeTree" in args.actions:
if args.codeMapping is None:
parser.error("Action addCodeTree needs" +
"--codeMapping to be specified")
if args.newIndex is None:
parser.error("Action addCodeTree needs" +
"--newIndex to be specified")
codeMappingFile = path.abspath(args.codeMapping)
codeMapping = parseMapping(codeMappingFile)
argDict["codeMapping"] = codeMapping
argDict["newIndex"] = int(args.newIndex)
# ----removeCodeTree-------------------------------------
# Only arguments with default values needed, no validation is needed
# ----moveCodeTree---------------------------------------
if "moveCodeTree" in args.actions:
if args.newIndex is None:
parser.error("Action moveCodeTree needs" +
"--newIndex to be specified")
try:
argDict["newIndex"] = int(args.newIndex)
except ValueError:
log.error("newIndex must be an integer value")
sys.exit(1)
# ----extractByGeocode-----------------------------------
if "extractByGeocode" in args.actions:
if args.edbMapping is None:
parser.error("Action extractByGeocode needs" +
"--edbMapping to be specified")
log.info("Slicing edb by geocode")
edbMappingFile = path.abspath(args.edbMapping)
# Geocode has to be int
edbMapping = parseMapping(edbMappingFile, keyType=int)
argDict["edbMapping"] = edbMapping
if 'scale' in args.actions:
argDict['substance'] = subdb.substIndices[args.substance]
argDict['factor'] = args.factor
if args.factor is None:
log.error("Must specify --factor")
sys.exit(1)
if args.substance is None:
log.error("Must specify --substance")
sys.exit(1)
# Processing actions
for action in args.actions:
log.info("Running action: %s" % action)
argDict = eval(action)(argDict)
# Writes each processed grid to edb
if loadGridData:
for grd in argDict["grids"]:
if not match(grd["NAME"], argDict["filters"]) or grd.hasSubgrp():
continue
else:
log.debug(
"Wrote grid %s to edb: %s, user: %s" % (
grd.par["NAME"].val, grd.edb.name, grd.user)
)
grd.load()
else:
for grd in argDict["grids"]:
if not match(grd["NAME"], argDict["filters"]) or grd.hasSubgrp():
continue
else:
log.debug(
"Wrote .asc-file for grid %s to edb: %s, user: %s" % (
grd.par["NAME"].val, grd.edb.name, grd.user)
)
grd.writeAsc(grd.getAscPath())
# Writes each processed database to edb
if argDict["sources"] is not None:
with SourceStream(edb, mode="w") as stream:
model_writer = ModelWriter(stream)
for src in argDict["sources"]:
log.debug('Wrote source %s' % src.NAME)
model_writer.write(src)
for dbName in ["roaddb", "subgrpdb", "svehdb", "vehdb"]:
if argDict[dbName] is not None:
argDict[dbName].write()
log.info(
"Wrote %s to edb: %s, user: %s" % (
dbName, edb.name, edb.user)
)
def main():
#-----------Setting up and unsing option parser-----------------------
parser=OptionParser(usage= usage, version=version)
parser.add_option("-u",'--user',
action="store",dest="user",
help="Name of target edb user")
parser.add_option("-e","--edb",
action="store",dest="edb",
help="Name of target edb")
parser.add_option("-i","--infile",
action="store",dest="infile",
help="Input csv file")
# parser.add_option("-y","--year",
# action="store",dest="year",
# help="Only store sources for given year")
parser.add_option("-v",dest='loglevel',
action="store_const",default=get_loglevel(),
help="produce verbose output")
parser.add_option("-t", "--template",
action="store",dest="cf",default=None,
help="Generate default controlfile")
parser.add_option("-o", "--outfile",
action="store",dest="outfile",default=None,
help="Name of outfiles (without extension)")
parser.add_option("-d","--delimiter",
action="store",dest="delimiter",default="\t",
help="Delimiter used in csv-file")
parser.add_option("-c","--filterCol",
action="store",dest="filterCol",
help="Header of column to use as filter")
parser.add_option("-f","--filterVal",
action="store",dest="filterVal",
help="Value to use in filter")
# parser.add_option("-g", "--geocodeRasterDir",
# action="store",dest="geocodeRasterDir",default=None,
# help="Directory with geocode rasters")
(options, args) = parser.parse_args()
#--------------------Init logger-----------------------
# rootLogger = logger.RootLogger(level=options.loglevel)
logging.basicConfig(
format='%(levelname)s:%(name)s: %(message)s',
level=options.loglevel,
)
global log
# log = rootLogger.getLogger(sys.argv[0])
log = logging.getLogger(parser.prog)
#-----------------Validating options-------------------
if options.cf is not None:
generateCf(path.abspath(options.cf),controlFileTemplate)
log.info("Wrote default controlfile")
return 1
if options.user is None:
log.error("Need to specify -u <user>")
return 1
if options.edb is None:
log.error("Need to specify -e <edb>")
return 1
# if options.year is None:
# log.error("Need to specify -y <year>")
# return 1
# if len(options.year)!=4:
# log.error("Year should be given with four digits")
# return 1
if len(args)!=1:
log.error("Controlfile should be given as argument")
return 1
dmn=Domain()
edb=Edb(dmn,options.user,options.edb)
if not edb.exists():
log.error("Edb %s does not exist" %options.edb)
return 1
log.info("Parsing controlfile")
cf=ControlFile(args[0])
cdbPars=re.compile("companydb\.par\.(\w*?):").findall(cf.content)
fdbPars=re.compile("facilitydb\.par\.(\w*?):").findall(cf.content)
sdbPars=re.compile("sourcedb\.par\.(\w*?):").findall(cf.content)
substEmisNr=re.compile("sourcedb\.subst_emis\.([0-9]*)\.emis").findall(cf.content)
subgrpEmisNr=re.compile("sourcedb\.subgrp_emis\.([0-9]*)\.emis").findall(cf.content)
cdbCols={}
cdbDefaults={}
for par in cdbPars:
cdbCols[par]=cf.findString("companydb.par.%s:" %par)
cdbDefaults[par]=cf.findString("companydb.par.%s.default:" %par,
optional=True,default=None)
fdbCols={}
fdbDefaults={}
for par in fdbPars:
fdbCols[par]=cf.findString("facilitydb.par.%s:" %par)
fdbDefaults[par]=cf.findString("facilitydb.par.%s.default:" %par,
optional=True,default=None)
sdbCols={}
sdbDefaults={}
for par in sdbPars:
sdbCols[par]=cf.findString("sourcedb.par.%s:" %par)
sdbDefaults[par]=cf.findString("sourcedb.par.%s.default:" %par,
optional=True,default=None)
substEmisCols={}
substEmisDefaults={}
if substEmisNr is not None:
for emisNr in substEmisNr:
cols={}
defaults={}
emisPars=re.compile("sourcedb\.subst_emis\.%s\.(\w*?):" %(emisNr)).findall(cf.content)
emisDefaultPars=re.compile(
"sourcedb\.subst_emis\.%s\.(\w*?)\.default:" %(emisNr)).findall(cf.content)
if emisPars is not None:
for par in emisPars:
cols[par]=cf.findString("sourcedb.subst_emis.%s.%s:" %(emisNr,par))
if emisDefaultPars is not None:
for par in emisDefaultPars:
defaults[par]=cf.findString("sourcedb.subst_emis.%s.%s.default:" %(emisNr,par),
optional=True,default=None)
substEmisCols[emisNr]=cols
substEmisDefaults[emisNr]=defaults
subgrpEmisCols={}
subgrpEmisDefaults={}
if subgrpEmisNr is not None:
for emisNr in subgrpEmisNr:
cols={}
defaults={}
emisPars=re.compile("sourcedb\.subgrp_emis\.%s\.(\w*?):" %(emisNr)).findall(cf.content)
emisDefaultPars=re.compile(
"sourcedb\.subgrp_emis\.%s\.(\w*?)\.default:" %(emisNr)).findall(cf.content)
if emisPars is not None:
for par in emisPars:
cols[par]=cf.findString("sourcedb.subgrp_emis.%s.%s:" %(emisNr,par))
if emisDefaultPars is not None:
for par in emisDefaultPars:
defaults[par]=cf.findString("sourcedb.subgrp_emis.%s.%s.default:" %(emisNr,par),
optional=True,default=None)
subgrpEmisCols[emisNr]=cols
subgrpEmisDefaults[emisNr]=defaults
log.info("Reading subdb...")
subdb=Subdb(edb)
subdb.read()
log.info("Reading companydb...")
companydb=Companydb(edb)
companydb.read()
log.info("Reading sourcedb...")
# source_stream = SourceStream(edb, 'w')
source_stream = open(options.outfile, 'w')
source_writer = ModelWriter(source_stream,encoding="HP Roman8")
log.info("Reading facilitydb...")
facilitydb=Facilitydb(edb)
facilitydb.read()
log.info("Reading subgrpdb")
subgrpdb=Subgrpdb(edb)
subgrpdb.read()
log.info("Reading edb.rsrc")
rsrc=Rsrc(edb.rsrcPath())
acCodeTables=[]
for i in range(rsrc.numberOfCodeTrees("ac")):
acCodeTables.append(CodeTable(rsrc.path,codeType="ac",codeIndex=i+1))
gcCodeTables=[]
for i in range(rsrc.numberOfCodeTrees("gc")):
gcCodeTables.append(CodeTable(rsrc.path,codeType="gc",codeIndex=i+1))
geocodeRasters=[]
rast1=Raster()
rast1.read("/usr/airviro/data/geo/topdown/dynamicRasters/dynamic__GEOCODE__1.txt")
rast2=Raster()
rast2.read("/usr/airviro/data/geo/topdown/dynamicRasters/dynamic__GEOCODE__2.txt")
geocodeRasters.append(rast1)
geocodeRasters.append(rast2)
log.info("Reading csv-file")
table=DataTable()
table.read(options.infile,delimiter=options.delimiter,encoding="ISO-8859-15")
if options.filterCol is not None:
if options.filterCol not in table.colIndex:
log.error("Filter column header not found in table")
sys.exit(1)
invalid=False
nFiltered=0
nRows=0
log.info("Processing rows")
for rowInd,row in enumerate(table.data):
nRows+=1
if options.filterCol is not None:
filterVal=row[table.colIndex[options.filterCol]]
if options.filterVal!=str(filterVal):
nFiltered+=1
continue
comp = Company()
for par in comp.parOrder:
val=cdbDefaults.get(par,None)
if par in cdbCols:
colId=cdbCols[par]
try:
tableVal=row[table.colIndex[colId]]
except KeyError:
log.error(
"No column with header %s, columns: %s" %(
colId,str(table.listIds())))
if tableVal is not None:
val = tableVal
if val is not None:
#Too long names are truncated
if par=="NAME" and len(val)>45:
val=val[:45]
comp[par]=val
fac = Facility()
for par in fac.parOrder:
val=fdbDefaults.get(par,None)
if par in fdbCols:
colId=fdbCols[par]
tableVal=row[table.colIndex[colId]]
if tableVal is not None:
val = tableVal
if val is not None:
#Too long names are truncated
if par=="NAME" and len(val)>45:
val=val[:45]
fac[par]=val
src = Source()
for par in ["X1", "Y1","X2","Y2",
"PX","PY","NAME","INFO","INFO2","DATE","CHANGED",
"CHIMNEY HEIGHT","GASTEMPERATURE","GAS FLOW",
"SEARCHKEY1","SEARCHKEY2","SEARCHKEY3",
"SEARCHKEY4","SEARCHKEY5","CHIMNEY OUT","CHIMNEY IN",
"HOUSE WIDTH","HOUSE HEIGHT","NOSEGMENTS","BUILD_WIDTHS",
"BUILD_HEIGHTS","BUILD_LENGTHS","BUILD_DISTFARWALL",
"BUILD_CENTER","GEOCODE","FORMULAMACRO","ALOB"]:
val=sdbDefaults.get(par,None)
if par in sdbCols:
colId=sdbCols[par]
tableVal=row[table.colIndex[colId]]
if tableVal is not None:
val = tableVal
if val is not None:
#validate code
if par=="GEOCODE" and val is not None:
gcList=val.split()
for codeIndex,code in enumerate(gcList):
if not gcCodeTables[codeIndex].hasCode(code):
log.error("Invalid geo code %s on row %i" %(code,rowInd))
invalid=True
#Too long names are truncated
if par=="NAME" and len(val)>45:
val=val[:45]
#Store in src object and convert to correct type
src._fieldvalues[par] = lazy_parse(
src, par, val)
gc1=geocodeRasters[0].getVal(src.get_coord()[0],src.get_coord()[1])
gc2=geocodeRasters[1].getVal(src.get_coord()[0],src.get_coord()[1])
src.GEOCODE = [str(int(gc1)) + "." + str(int(gc2))]
for emisNr,emis in substEmisCols.items():
substEmis={"unit":None,"ac":None,"substance":None,"emis":None}
for par in substEmis.keys():
if par in emis:
substEmis[par]=row[table.colIndex[emis[par]]]
else:
try:
substEmis[par]=substEmisDefaults[emisNr][par]
except KeyError:
log.error(
"Need to specify column or default value for subgrp emis %i" %emisNr)
substInd=subdb.substIndex(substEmis["substance"])
if substInd is None:
log.error("Invalid substance name %s on row %i" %(
substEmis["substance"],rowInd))
sys.exit(1)
try:
unit=rsrc.sub[substEmis["unit"]]
except KeyError:
log.error("Invalid unit name %s on row %i" %(emis["unit"],rowInd))
sys.exit(1)
acList=substEmis["ac"].split('\\')[0].split()
for codeIndex,code in enumerate(acList):
# if code == "2.A.4.2":
# import pdb; pdb.set_trace()
refCode = acCodeTables[codeIndex].checkCode(code)
if refCode == "-":
log.error("Invalid activity code %s on row %i" %(code,rowInd))
sys.exit(1)
if refCode != code:
acList[codeIndex] = refCode
substEmis["ac"] = acList
if substEmis["emis"] is not None and substEmis["emis"]!="0":
try:
emis = src.add_emission()
emis.UNIT = substEmis["unit"]
emis.ACTCODE = substEmis["ac"] # needs re-formatting
emis.EMISSION = float(substEmis["emis"])
emis.SUBSTANCE = substInd
emis.auto_adjust_unit(edb)
except:
# print substEmis
# log.error("Invalid substance emission on row %i" %rowInd)
invalid=True
src.EMISSION=src.EMISSION[:-1]
for emis in subgrpEmisCols.values():
subgrpEmis={"unit":None,"ac":None,"name":None,"emis":None}
for par in subgrpEmis.keys():
if par in emis:
subgrpEmis[par]=row[table.colIndex[emis[par]]]
else:
try:
subgrpEmis[par]=subgrpEmisDefaults[emisNr][par]
except KeyError:
log.error(
"Need to specify column or default value for subgrp emis %i" %emisNr)
#validating subgrp name
try:
subgrp=subgrpdb.getByName(subgrpEmis["name"])
except KeyError:
log.error("Invalid subgrp name %s on row %i" %(subgrpEmis["name"],rowInd))
invalid=True
#validating subgrp emis unit
try:
unitFactor=rsrc.subGrpEm[subgrpEmis["unit"]]
except KeyError:
log.error("Invalid unit %s for subgrp emission on row %i" %(
subgrpEmis["unit"],rowInd))
invalid=True
#validating subgrp activity code
acList=subgrpEmis["ac"].split()
for codeIndex,code in enumerate(acList):
refCode = acCodeTables[codeIndex].checkCode(code)
if refCode == "-":
log.error("Invalid activity code %s on row %i" %(code,rowInd))
invalid=True
break
if refCode != code:
acList[codeIndex] = refCode
substEmis["ac"] = acList
try:
src.addSubgrpEmis(subgrp.index,emis=subgrpEmis["emis"],unit=subgrpEmis["unit"],
ac=subgrpEmis["ac"])
except:
log.error("Invalid subgrp emission on row %i" %rowInd)
invalid=True
companydb.append(comp,force=True)
facilitydb.append(fac,force=True)
source_writer.write(src)
# sourcedb.append(src)
if invalid:
log.info("No output written due to validation errors")
sys.exit(0)
if len(companydb.items)>0:
if options.outfile is None:
log.info("Writing companydb")
else:
log.info("Writing company db to file")
companydb.write(filename=options.outfile+".companydb")
if len(facilitydb.items)>0:
if options.outfile is None:
log.info("Writing facilitydb")
else:
log.info("Writing facilitydb to file")
facilitydb.write(filename=options.outfile+".facilitydb")
# if len(sourcedb.sources)>0:
# if options.outfile is None:
# log.info("Writing sourcedb")
# else:
# log.info("Writing sourcedb to file")
# sourcedb.write(filename=options.outfile+".sourcedb")
if options.filterCol is not None:
log.info("Filtered out %i out of %i" %(nFiltered,nRows))
def main():
# Parse command line arguments
parser = argparse.ArgumentParser(description=__doc__)
utils.add_standard_command_options(parser)
parser.add_argument("controlfile", metavar='CONTROLFILE',
action="store",
help="Controlfile for topdown processing")
parser.add_argument("-t", "--template", metavar='TEMPLATEFILE',
action="store",dest="cf",default=None,
help="Generate default controlfile")
args = parser.parse_args()
if args.cf is not None:
generateCf(args.cf)
log.info("Wrote default controlfile")
sys.exit(0)
log.info("Starting topdown processing")
# Opening controlfile
cf = ControlFile(args.controlfile)
dmn = Domain()
log.info("Reading topdown table")
tdTableName = cf.findExistingPath("topDownTable:")
tdTable = DataTable()
tdTable.keys.append("Code")
tdTable.read(tdTableName,delimiter=";")
log.info("Reading national totals table")
natTotalTableName = cf.findExistingPath("nationalTotalTable:")
natTable = DataTable(desc=[{"id": "Code", "type":unicode},
{"id": "description", "type":unicode}])
natTable.keys.append("Code")
natTable.read(natTotalTableName, units=True, defaultType=str)
notationKeys = ["NE", "NO", "NA", "IE"]
log.debug("Remove notation keys from national totals table")
for row in natTable.data:
for i in range(len(row)):
if row[i] in notationKeys:
row[i] = None
log.debug("Convert all emission columns in national totals to float")
for colId in natTable.listIds():
if colId not in ["Code","description"]:
natTable.convertCol(colId,float)
log.debug("Store units from national totals for each substance in dict")
natUnits={}
for col in natTable.desc:
if col.get("units",None)!=None:
natUnits[col["id"]]=col["units"]
log.debug("Read remaining data from control file")
bottomupEdbName = cf.findString("bottomUpEdb:")
topDownEdbName = cf.findString("topDownEdb:")
emissionsEdbName = cf.findString("emissionsEdb:")
userName = cf.findString("user:"******"year:")
#initialize edb objects
buEdb = Edb(dmn,userName,bottomupEdbName)
tdEdb = Edb(dmn,userName,topDownEdbName)
eEdb = Edb(dmn,userName,emissionsEdbName)
log.info("Reading/preparing EDB:s")
log.info("Reading subdb")
subdb = Subdb(eEdb)
subdb.read()
log.info("Reading subgrpdb")
subgrpdb = SubgrpStream(buEdb)
subgrpdb.read()
log.info("Reading facilitydb")
facilityIn = FacilityStream(buEdb)
log.info("Reading companydb")
companyIn = CompanyStream(buEdb)
facilityOut = FacilityStream(eEdb,mode="w")
companyOut = CompanyStream(eEdb,mode="w")
log.info("Writing company db to result edb")
companyOut.write(companyIn.read())
log.info("Writing facility db to result edb")
facilityOut.write(facilityIn.read())
if not buEdb.exists():
log.error("Edb " + buEdb.name + " does not exist for user " + userName +
" in domain " + dmn.name)
sys.exit(1)
if not tdEdb.exists():
log.error("Edb " + tdEdb.name + " does not exist for user " + userName +
" in domain " + dmn.name)
sys.exit(1)
if not eEdb.exists():
log.error("Edb " + eEdb.name + " does not exist for user " + userName +
" in domain " + dmn.name)
sys.exit(1)
keys = tdEdb.listGrids()
msg = "%i keys found in edb: %s" % (len(keys), tdEdb.name)
log.info(msg)
# sourcedb from bottom-up edb
with SourceStream(buEdb, mode='rb') as source_instream:
source_reader = ModelReader(source_instream)
bu_sources = list(source_reader)
log.info(
"%i point sources found in edb: %s" % (
len(bu_sources),
buEdb.name)
)
# Empty sourcedb of the result edb
if cf.findBoolean("emptyEmissionSourcedb:"):
eEdb.empty_sourcedb()
e_sources = []
log.info("Removed point sources from edb: %s" % (eEdb.name))
else:
# sourcedb from emission edb (result edb)
with SourceStream(eEdb, mode='rb') as source_instream:
source_reader = ModelReader(source_instream)
e_sources = list(source_reader)
msg = "%i point sources found in edb: %s" % (len(e_sources), eEdb.name)
log.info(msg)
if not path.exists(eEdb.rsrcPath()):
log.error("No edb.rsrc exists for emission edb")
sys.exit()
else:
rsrc = Rsrc(eEdb.rsrcPath())
acIndex = cf.findInt("acIndex:")
codeDepth = rsrc.ac[acIndex-1].depth
substances = cf.findStringList("substances:")
for subst in substances:
if subst not in subdb.substIndices:
log.error("Substance: " + subst + " not in Airviro substance list")
sys.exit()
# Initialize trace for debug and additional logging
if cf.findBoolean("trace:") == True:
log.info("Initializing trace for detailed logging")
trace = TraceDef(
active=True,
substances=cf.findStringList("trace.substances:"),
logfile=cf.findString("trace.logfile:"),
regdefgc=cf.findIntList("trace.regdef.gc:",
optional=True,
default=None),
gcDefRaster=cf.findExistingPath("trace.gcraster:")
)
else:
trace = TraceDef(active=False)
log.info("Initializing result table")
resTablePath = cf.findString("resTable:")
resTable = DataTable(desc=[{"id": "Code", "type": unicode}])
resTable.keys.append("Code")
for subst in substances:
resTable.addCol({"id": subst, "type": float, "unit": "%"})
# Create emission grid template (with geocodes)
log.info("Reading emission grid template")
eGridTemplatePath = cf.findExistingPath("emisGridTemplatePath:")
eGridTemplate = Egrid(eEdb,"name")
if eGridTemplatePath[-4:] == ".asc":
eGridTemplatePath=eGridTemplatePath[:-4]
eGridTemplate.readData(eGridTemplatePath)
eGridTemplate.substances = {}
eGridTemplate.par["SUBSTANCE"].val = []
dd = {"key": None,
"regstat": None,
"regdef": None,
"bu_sources": bu_sources,
"psIndices": [],
"units": natUnits,
"rsrc": rsrc,
"subdb": subdb,
"trace": trace,
"subgrpdb": subgrpdb
}
# Process all rows in the topdown table
for row in tdTable.data:
code = row[tdTable.colIndex["Code"]]
active = row[tdTable.colIndex["Active"]]
statType = row[tdTable.colIndex["Stat_type"]]
if active == "no":
continue
log.info("Code: "+code)
distributed=False
# Add '-' to the code to reach max length (fix for a GUI bug)
airviroCode = code
# while len(airviroCode.split(".")) < codeDepth:
# airviroCode += ".-"
tdrow = tdTable.data[tdTable.rowIndex([code])]
nrow = natTable.data[natTable.rowIndex([code])]
# Create a resTable row to fill with data
resrow = [None] * resTable.ncols
resrow[0] = code
# Check if national totals are non-zero
nonZero = False
for val in nrow:
if val != None:
if val > 0:
nonZero = True
break
# Filter out indices for pointsources with the current ac
# Also including sources coded with sub-codes
# This allows to estimate top-down emissions on a higher code-level
psIndices = []
for i, ps in enumerate(bu_sources):
codeMatch = False
for emis in ps.EMISSION:
# It is assumed that the first code is used while processing topdown
ac = emis.ACTCODE[0]
if ac[-1] == ".":
ac=ac[:-1]
# if ac[:len(code)] == code:
if ac == code:
codeMatch = True
break
if not codeMatch:
for emis in ps.SUBGRP:
# It is assumed that the first code is used while processing topdown
ac = emis.ACTCODE[0]
if ac[:len(code)] == code:
codeMatch = True
break
if codeMatch:
psIndices.append(i)
dd["psIndices"] = psIndices
keyName = row[tdTable.colIndex["Key"]]
#If no distribution key specified and no ps in bottom-up edb - cont.
if keyName is None and psIndices == []:
log.debug("No key and no point sources found for code: %s, skipping..." % code)
resTable.addRow(resrow)
continue
if psIndices!=[]:
msg = "--Found %i pointsources" % len(psIndices)
log.info(msg)
if keyName is not None:
if keyName not in keys:
log.error("No such key: " + keyName)
sys.exit()
msg = "--Key: %s" % keyName
log.info(msg)
keyGrid = Egrid(tdEdb, keyName)
keyGrid.readData()
log.debug("Read key: " + keyName + " from topdownEdb")
# create emission grid to store distributed emissions
eGrid = deepcopy(eGridTemplate)
eGrid.name = code.replace(".", "_")
eGrid.par["NAME"].val = code
eGrid.par["INFO2"].val = "Distribution key: " + keyGrid.par["NAME"].val
eGrid.par["ACTIVITYCODE"].val = [airviroCode.split(".")]
regstatName = row[tdTable.colIndex["Regstat"]]
regdefName = row[tdTable.colIndex["Regdef"]]
if regstatName is not None:
if regdefName is None:
log.error("No region definition given for regional statistics: " +
regstatName)
sys.exit(1)
regstatPath = path.join(dmn.domainPath(), "topdown", "regstat", regstatName)
regstat = DataTable()
log.info("regstatPath: "+regstatPath)
regstat.read(regstatPath, units=True, defaultType=float, delimiter=";")
if not "Geocode" in regstat.listIds():
log.error("No Geocode column found in regstat")
sys.exit(1)
regstat.convertCol("Geocode", int)
regstat.keys.append("Geocode") # Making Geocode the primary key
# create list of unique geo codes
geocodes = [row[regstat.colIndex["Geocode"]] for row in regstat.data]
geocodes = unique(geocodes)
for colId in regstat.listIds():
if colId.lower() == "year":
rows = []
regstat.convertCol(colId, int)
# Make it possible to accumulate year
regstat.setKeys(regstat.keys + [colId])
# Calculates the total emission for each geocode
# in case there are multiple rows for different fuels etc
colsToSum = regstat.listIds()
colsToSum.remove(colId)
colsToSum.remove("Geocode")
for gc in geocodes:
# sums all numeric values in colsToSum for
# rows matching row id [gc,year]
#returns an accumulated row and appends it to rows
rowId = regstat.dict2RowId({"Geocode": gc, colId: year})
rows.append(regstat.accumulate(rowId, "sum", colsToSum))
regstat.data = rows # replace original rows with accumulated rows
regstat.keys.remove(colId)
break
# dd["regstat"] = regstat
regdef = Raster()
regdefPath = path.join(dmn.domainPath(), "topdown", "regdef", regdefName)
regdef.read(regdefPath)
dd["regstat"] = regstat
dd["regdef"] = regdef
else:
dd["regstat"] = None
dd["regdef"] = None
if dd["regstat"] is not None and len(bu_sources) > 0 and statType == "fixed":
log.info("--Regionalizing pointsources")
dd = regionalizePS(dd, code)
if keyName is not None and nonZero:
regionalizedDefault = False
# Spatial distribution of emissions
for subst in substances:
sInd = subdb.substIndices[subst]
toUnit = dd["units"][subst] + "/year"
ntot = nrow[natTable.colIndex[subst]]
pstot = 0
for i in dd["psIndices"]:
source = dd["bu_sources"][i]
# TODO: should give reference to subgrps to include emis from them
pstot += source.get_emis(
sInd,
toUnit,
eEdb,
actcodes=[code]
)
if ntot is None or ntot == 0:
if pstot > 0:
# 9999 is used as marker for no national total
resrow[resTable.colIndex[subst]] = 9999.0
log.warning(
"Nattot is 0 but ps tot is: %f %s" % (pstot, toUnit))
continue
nrest = ntot - pstot
resrow[resTable.colIndex[subst]] = 100.0
if abs(nrest / ntot) < 0.0001:
nrest = 0
log.info(
"--Rest is < 0.01 % of national total, rounded to zero"
)
continue
elif nrest < 0:
log.warning(
"--National rest is below zero, %4.2f proc for %s" % (
-1 * nrest / ntot * 100,
subst)
)
dd["trace"].write()
# continue
log.info(
"---Substance: "+subst+
", rest is: " + str(nrest) +
toUnit + " = " + str(nrest / ntot * 100.0) + "%"
)
try:
keyRast = keyGrid.substances[sInd]
except KeyError:
keyRast = keyGrid.substances[subdb.substIndices["all"]]
dd["key"] = keyRast
if dd["regstat"] is not None:
if (subst not in regstat.colIndex and
sInd not in keyGrid.substances and not regionalizedDefault):
dd = regionalizeKey(dd, subst, code)
regionalizedDefault = True
else:
dd = regionalizeKey(dd, subst, code)
emisRast = distribute(dd["key"], nrest)
emisRast = emisRast * unitConvFac(toUnit, "ton/year")
eGrid.addData(emisRast, dd["subdb"].substIndices[subst])
distributed = True
else:
# resTable is filled
# In case all national totals are zero but there are ps
for subst in substances:
sInd = dd["subdb"].substIndices[subst]
toUnit = dd["units"][subst] + "/year"
ntot = nrow[natTable.colIndex[subst]]
pstot = 0
for i in dd["psIndices"]:
source = dd["bu_sources"][i]
# subgrps are not used!
pstot += source.get_emis(sInd, toUnit, buEdb,
actcodes=[code])
if ntot!=0 and ntot is not None:
resrow[resTable.colIndex[subst]] = pstot / ntot * 100.0
else:
resrow[resTable.colIndex[subst]] = -999.0
if len(dd["psIndices"]) > 0:
tmp_sources = (bu_sources[i] for i in dd["psIndices"])
with SourceStream(eEdb, mode='wb') as out_source_stream:
source_writer = ModelWriter(out_source_stream)
for source in tmp_sources:
source_writer.write(source)
log.debug("Wrote ps to emission edb")
if distributed:
eGrid.load()
log.debug("Wrote emission grid to emission edb")
dd["trace"].write()
resTable.addRow(resrow)
resTableFile = open(resTablePath,"w")
resTable.write(resTableFile)
log.info("Finished topdown process")