def processFile(domainName, user, edbName, filePath):
    logger=logging.getLogger('loadRUSedb.processFile')

    dmn=domain(domainName)
    edb=Edb(domainName,user,edbName) 

    if not dmn.edbUserExist(user):
        logger.error("Directory for user '"+user+"' does not exist in edb directory for domain '"+domain+"'")
        sys.exit()

    if not edb.exists():
        edb.create()
        logger.info("EDB '"+edb.name+"' for user '"+user+"' in domain '"+dmn.name+"' created")

    if filePath[-3:]==".gz":
        utilities.gunzip(filePath)
        filePath=filePath[:-3]
    filePath=filePath[:-5] #removing .data from the file name

    #converting from dos-format to linux
    utilities.dos2linux(filePath+".asc")
    utilities.dos2linux(filePath+".data")

    #create grid object
    gridname=path.basename(filePath)
    grid=Egrid(dmn.name,user,edb.name,gridname)

    grid.load(filePath)  #load grid into edb
Exemple #2
0
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))
Exemple #3
0
def main():
    #-----------Setting up and unsing option parser-----------------------
    parser=OptionParser(usage= usage, version=version)
    
    parser.add_option('--src_user',
                      action="store",dest="src_user",
                      help="Name of target edb user")

    parser.add_option("--src_edb",
                      action="store",dest="src_edb",
                      help="Name of target edb")

    parser.add_option('--tgt_user',
                      action="store",dest="tgt_user",
                      help="Name of target edb user")

    parser.add_option("--tgt_edb",
                      action="store",dest="tgt_edb",
                      help="Name of target edb")

    parser.add_option("--codeIndex",
                      action="store",dest="codeIndex",
                      help="Activity code index to use for filter")

    parser.add_option("--codeValue",
                      action="store",dest="codeValue",
                      help="Activity code value to filter by")

    parser.add_option("-l", "--loglevel",
                      action="store",dest="loglevel",default=2,
                      help="Sets the loglevel (0-3 where 3=full logging)")

    parser.add_option("-o", "--outfile",
                      action="store",dest="outfile",
                      help="Output filenam, leave out to write directly to target edb")

    (options, args) = parser.parse_args()
    
    #--------------------Init logger-----------------------
    rootLogger = logger.RootLogger(level=options.loglevel)
    global log
    log = rootLogger.getLogger(sys.argv[0])

    #-----------------Validating options-------------------

    if options.src_user is None:
        log.error("Need to specify --src_user <user>")
        return 1
    if options.src_edb is None:
        log.error("Need to specify --src_edb <edb>")
        return 1

    if options.tgt_user is None:
        log.error("Need to specify --tgt_user <user>")
        return 1
    if options.tgt_edb is None:
        log.error("Need to specify --tgt_edb <edb>")
        return 1


    if len(args)!=0:
        log.error("No argument expected")
        return 1

    dmn=Domain()
    src_edb=Edb(dmn,options.src_user,options.src_edb)
    tgt_edb=Edb(dmn,options.tgt_user,options.tgt_edb)

    if not src_edb.exists():
        log.error("Source edb %s does not exist") %options.src_edb
        return 1

    if not tgt_edb.exists():
        log.error("Target edb %s does not exist") %options.tgt_edb
        return 1

    if options.codeIndex is None:
        log.error("No code index specified")
        return 1

    if options.codeValue is None:
        log.error("No code value specified")
        return 1

    log.info("Reading subdb...")
    src_subdb=Subdb(src_edb)
    src_subdb.read()

    log.info("Reading sourcedb...")
    src_sourcedb=Sourcedb(src_edb)
    src_sourcedb.read()
    

    tgt_sourcedb=Sourcedb(tgt_edb)
    tgt_sourcedb.read()

    #log.info("Reading subgrpdb")
    #src_subgrpdb=Subgrpdb(src_edb)
    #src_subgrpdb.read()

    log.info("Reading edb.rsrc")
    src_rsrc=Rsrc(src_edb.rsrcPath())
    tgt_rsrc=Rsrc(tgt_edb.rsrcPath())

    log.info("Reading target sourcedb...")


    acCodeTables=[]
    for i in range(tgt_rsrc.numberOfCodeTrees("ac")):
        acCodeTables.append(CodeTable(tgt_rsrc.path,codeType="ac",codeIndex=i+1))

    gcCodeTables=[]
    for i in range(tgt_rsrc.numberOfCodeTrees("gc")):
        gcCodeTables.append(CodeTable(tgt_rsrc.path,codeType="gc",codeIndex=i+1))



    copiedSources=0
    readSources=0
    for src in src_sourcedb.sources:
        hasEmis=False
        src["PX"]=0
        src["PY"]=0
                      
        toBeRemoved=[]
        for substInd,emis in src.subst_emis.iteritems():
            ac= emis["ACTCODE"][int(options.codeIndex)-1]
            ac=[c for c in ac if c!="-"]
            emis["ACTCODE"][int(options.codeIndex)-1]=ac #Remove '-'from code

            ac=".".join(ac)
            
            if ac!=options.codeValue:
                toBeRemoved.append(substInd)

        for key in toBeRemoved:
            src.removeSubstEmis(key)

        if src.subst_emis!={}:
            hasEmis=True
        
        toBeRemoved=[]
        for subgrpInd,emis in src.subgrp_emis.iteritems():
            ac= emis["ACTCODE"][int(options.codeIndex)-1]
            ac=[c for c in ac if c!="-"]
            emis["ACTCODE"][int(options.codeIndex)-1]=ac #Remove '-'from code
            ac=".".join(ac)

            
            if ac!=options.codeValue:
                toBeRemoved.append(subgrpInd)

        for key in toBeRemoved:
            src.removeSubgrpEmis(key)

        if src.subgrp_emis!={}:
            hasEmis=True


        toBeRemoved=[]
        for actInd,emis in src.activity_emis.iteritems():
            ac= emis["ACTCODE"][int(options.codeIndex)-1]
            ac=[c for c in ac if c!="-"]
            emis["ACTCODE"][int(options.codeIndex)-1]=ac #Remove '-'from code
            ac=".".join(ac)
            
            if ac!=options.codeValue:
                toBeRemoved.append(actInd)

        for key in toBeRemoved:
            src.removeActivityEmis(key)

        if src.activity_emis!={}:
            hasEmis=True

        readSources+=1
        if hasEmis:
            copiedSources+=1
            tgt_sourcedb.sources.append(src)

    if options.outfile is None:
        log.info("Writing sources to target edb")
    else:
        log.info("Writing sources to file")

    tgt_sourcedb.write(filename=options.outfile,force=True)

    log.info("Successfully Copied %i out of %i sources" %(copiedSources,readSources))
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("-v","--viewports",
                      action="store",dest="viewports",
                      help="Comma-separated list of area id's to be cut out, default is all")

    parser.add_option("-y","--year",
                      action="store",dest="year",
                      help="Cut out for given year")
    
    parser.add_option("-l", "--loglevel",
                      action="store",dest="loglevel",default=2,
                      help="Sets the loglevel (0-3 where 3=full logging)")

    parser.add_option("-s","--suffix",
                      action="store",dest="suffix",default="v1",
                      help="Sets suffix to names of generated edb's to support version management, default is 'v1'")

    (options, args) = parser.parse_args()


    
    #--------------------Init logger-----------------------
    rootLogger = logger.RootLogger(level=options.loglevel)
    global log
    log = rootLogger.getLogger(sys.argv[0])

    #-----------------Validating options-------------------

    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.suffix)>4:
        log.error("Limit your suffix length to 4 characters")
        return 1

    if len(options.year)!=4:
        log.error("Year should be given with four digits")
        return 1

    dmn=Domain()
    viewports=[]
    if options.viewports is not None:        
        viewportIds=options.viewports.split(",")
    else:
        viewportIds=dmn.listViewports()
    for vpId in viewportIds:
        vp=ViewPort()
        vp.read(path.join(dmn.wndPath(),"modell.par"),vpId)
        viewports.append(vp)
    
    edb=Edb(dmn,options.user,options.edb)

    log.info("Reading sourcedb...")
    sourcedb=Sourcedb(edb)
    sourcedb.read()

    log.info("Reading emfacdb...")
    emfacdb=Emfacdb(edb)
    emfacdb.read()

    log.info("Reading subdb...")
    subdb=Subdb(edb)
    subdb.read()
    
    edbDotRsrc=edb.rsrcPath()
    
    for vpInd,vp in enumerate(viewports):        
        targetEdbName=vp.code+"_"+options.year+"_"+options.suffix     
        tEdb=Edb(dmn,options.user,targetEdbName)
        if tEdb.exists():
            log.info("Edb %s already exists, remove first to update" %targetEdbName)
            continue
        tEdb.create(edbRsrc=edbDotRsrc)
        log.info("Created empty edb %s" %targetEdbName)

        subdb.setEdb(tEdb)
        subdb.write()
        log.info("Wrote searchkeys")

        emfacdb.setEdb(tEdb)        
        emfacdb.write()
        log.info("Wrote emfacdb")
        
        tSourcedb=Sourcedb(tEdb)        
        
        log.info("Cutting out viewport %s (%i/%i)" %(vp.code,vpInd+1,len(viewports)))
        for srcInd,src in enumerate(sourcedb.sources):
            if includeShip(src,vp.code,src["Y1"],options.year):
                log.debug("Ship %i/%i included in %s" %(srcInd+1,len(sourcedb.sources),tEdb.name))
                tSourcedb.sources.append(src)
        tSourcedb.write()
        log.info("Wrote exatracted sources to %s" %tEdb.name)
        tEdb.setDesc("This edb has been extracted from %s under user %s, " %(edb.name,edb.user)+
                            "and includes all ships that have visited the map area %s (%s) during %s\n" %(vp.code,vp.name,options.year))
    log.info("Finished!")
    return 0
Exemple #5
0
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)
            )
Exemple #6
0
def extractByGeocode(argDict):
    """Extracts sources by geocode and saves to edb's for other users specified
    in a file with users, and optionally edb's, specified for each geocode"""

    edb = argDict["edb"]
    mapping = argDict["edbMapping"]
    codeLevel = argDict["codeLevel"]
    codeIndex = argDict["codeIndex"]

    # Processing grid sources
    for grd in argDict["grids"]:
        if not match(grd.par["NAME"].val, argDict["filters"]):
            continue

        # dict with (user,edb) as  key and a list of gc to be included as value
        subEdbs = {}

        allFound = True
        for gc in mapping:

            # The mapping can contain only users for each geocode, in that case
            # The users are stored as values and the original edb name is used
            # as default
            # If both user and edb name is given, the values are given in a
            # list for each gc
            if isinstance(mapping[gc], list):
                tmp_user = mapping[gc][0]
                tmp_edb = mapping[gc][1]
            else:
                tmp_user = mapping[gc]
                tmp_edb = edb.name
            subEdb = Edb(argDict["domain"], tmp_user, tmp_edb)
            if (tmp_user, tmp_edb) in subEdbs:
                subEdbs[(tmp_user, tmp_edb)].append(gc)
            else:
                subEdbs[(tmp_user, tmp_edb)] = [gc]

            # Check if edb exist, and if so if grid exist
            # (to save time when updating)
            if subEdb.exists():
                if grd.name not in subEdb.listGrids():
                    allFound = False
                    continue
            else:
                allFound = False

        if allFound:
            log.info(
                "Splitted grid " +
                "%s already exist for all" % grd.name +
                " geocodes, remove to update")
            continue

        log.debug("Reading grid: %s" % grd.name)
        grd.readData()
        for user_edb, gcList in subEdbs.iteritems():
            subEdb = Edb(argDict["domain"], user_edb[0], user_edb[1])
            subEdb.rsrc = edb.rsrc
            # extract data from grid that maches any of the geocodes in gcList
            subGrid = grd.sliceByGeoCode(
                codeLevel - 1,
                map(int, gcList),
                subEdb
            )
            if subGrid is not None:
                if not subEdb.exists():
                    subEdb.create(edbRsrc=edb.rsrcPath())
                    log.debug("Created edb for user %s" % subEdb.user)
                log.debug(
                    "Writing grid: %s to edb %s for user %s" % (
                        grd.name,
                        subEdb.name,
                        subEdb.user)
                )
                subGrid.load()

    if argDict["sources"] is not None:
        sub_source_writers = {}
#         subCompanydbs = {}
#         subFacilitydbs = {}

        # Read substance groups
        subgrpdb = Subgrpdb(edb)
        subgrpdb.read()

        facilitydb = Facilitydb(edb)
        facilitydb.read()
        companydb = Companydb(edb)
        companydb.read()

        for src in argDict["sources"]:
            if not match(src.NAME, argDict["filters"]):
                continue
#             gc = int(src.GEOCODE[codeIndex - 1][codeLevel - 1])
            gc = int(src.GEOCODE[codeIndex - 1].split(".")[codeLevel - 1])

            if src.PX is not None and src.PX != 0:
                facility = facilitydb.index[(src.PX, src.PY)]
                company = companydb.index[facility["COMPANY"]]

            # If no mapping is provided for a gc it is ignored
            if gc not in mapping:
                continue

            # create sub-edb and corresponding db:s
            if gc not in sub_source_writers:
                tmp_user = mapping[gc]
                subEdb = Edb(argDict["domain"], tmp_user, edb.name)

                # If a sub-edb does not exist,
                # create a new one with edb.rsrc from original edb
                if not subEdb.exists():
                    subEdb.create(edbRsrc=edb.rsrcPath())
                sub_source_writers[gc] = ModelWriter(
                    SourceStream(subEdb, mode='w')
                )

#                 subCompanydbs[gc] = Companydb(subEdb)
#                 subFacilitydbs[gc] = Facilitydb(subEdb)

                # Write substance groups to subEdb
                subgrpdb.setEdb(subEdb)
                log.info("Writing subgrpdb for user %s" % subEdb.user)
                subgrpdb.write()

#             subCompanydbs[gc].append(company, force=True)
#             subFacilitydbs[gc].append(facility,force=True)

            sub_source_writers[gc].write(src)

        # write all sources
        for gc in sub_source_writers:
            # log.info("Writing companies to edb for gc %i" % gc)
            # subCompanydbs[gc].write()
            # log.info("Writing facilities to edb for gc %i" % gc)
            # subFacilitydbs[gc].write()
            log.info("Writing sources to edb for gc %i" % gc)

    return argDict
Exemple #7
0
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")
def main():
    #-----------Setting up and using option parser-----------------------
    parser=OptionParser(usage= usage, version=version)

    parser.add_option(
        '-v',
        action='store_const', dest='loglevel',
        const=logging.DEBUG, default=get_loglevel(),
        help='produce verbose output'
    )
#     parser.add_option("-l", "--loglevel",
#                       action="store",dest="loglevel",default=2,
#                       help="Sets the loglevel (0-3 where 3=full logging)")
    
    parser.add_option("-t", "--template",
                      action="store",dest="cf",default=None,
                      help="Generate default controlfile") 
    
    (options, args) = parser.parse_args()

    #------------Setting up logging capabilities -----------
    logging.basicConfig(
        format='%(levelname)s:%(name)s: %(message)s',
        level=options.loglevel,
    )
    global log
    log = logging.getLogger(parser.prog)
    
    if options.cf!=None:
        generateCf(path.abspath(options.cf))
        log.info("Wrote default controlfile")
        sys.exit()

    if len(args)!=1:
        parser.error("Incorrect number of arguments")

    #Opening controlfile
    cf=ControlFile(fileName=path.abspath(args[0]))
    domain = Domain()
    edbName=cf.findString("edb:")
    userName=cf.findString("user:"******"Edb "+edbName+" does not exist for user "+userName+" in domain "+domain.name)
        sys.exit()

    year=cf.findInt("year:")
    substList=cf.findStringList("substances:")

    #Remove doubles from substance list
    substances={'nat':[]}
    for subst in substList:
        if subst not in substances['nat']:
            substances['nat'].append(subst)


    #Make dictionary for substance,units couples
    unitDict={}
    defaultUnit=cf.findString("nationalTotals.defaultUnit:")
    unitList=cf.findStringList("nationalTotals.units:")
    for i in range(len(unitList)):
        unitSubstances=cf.findStringList("nationalTotals.unit."+str(i)+":")
        for s in unitSubstances:
            unitDict[s]=unitList[i]
    for subst in substances['nat']:
        if subst not in unitDict.keys():
            unitDict[subst]=defaultUnit

    nationalTotalsTablePath=cf.findString("nationalTotalsTable:")
    nationalTotalsFile=codecs.open(nationalTotalsTablePath,'w',encoding="latin6")

    nationalTotalsTreePath=cf.findString("nationalTotalsTree:")
    nationalTotalsTreePath=path.abspath(nationalTotalsTreePath)
    nationalTotalTreeFile=open(nationalTotalsTreePath,'w')
    
    log.info("Reading emissions from CRF")
    CRFTree=CodeEmisTree(name="CRF")
    CRFPath=cf.findExistingPath("CRF.path:")
    CRFTree.parseCRFXML(CRFPath,year)
    CRFTree.calculateSums(byAttribute="Category")
    #CRFTree.write(sys.stdout)
    CRFTable=CRFTree.createTable(keys=["Category","Classification","AWMS"])
    HFC={
        "HFC-23":11700,
        "HFC-32":650,
        "HFC-41":150,
        "HFC-43-10 mee":1300,
        "HFC-125":2800,
        "HFC-134":1000,
        "HFC-134a":1300,
        "HFC-152a":140,
        "HFC-143":300,
        "HFC-143a":3800,
        "HFC-227ea":2900,
        "HFC-236fa":6300,
        "HFC-245ca":560
        }
    
    PFC={
        "CF4":6500,
        "C2F6":9200,
        "C3F8":7000,
        "C4F10":7000,
        "c-C4F8":8700,
        "C5F12":7500,
        "C6F14":7400
        }
    
    CRFTable.aggregateCols({"id":"HFC-tot","type":"str","units":"Gg"},HFC)
    CRFTable.aggregateCols({"id":"PFC-tot","type":"str","units":"Gg"},PFC)
    CRFTable.sortRows()
    CRFTableKeys=cf.findStringList("CRF.tableKeys:")
    CRFTable.keys=CRFTableKeys
    #CRFTable.write(sys.stdout)
    
    log.info("Reading emissions from NFR")
    NFRTree=CodeEmisTree(name="NFR")
    NFRPath=cf.findExistingPath("NFR.path:")
#     pdb.set_trace()
    NFRTree.parseNFRTable(filename=NFRPath)
    NFRTree.calculateSums()
    NFRTable=NFRTree.createTable()
    NFRTableKeys=cf.findStringList("NFR.tableKeys:")
    NFRTable.keys=NFRTableKeys
    colIndex=NFRTable.getColIndex("Code")
    for row in NFRTable.data:
        if isinstance(row[colIndex],str):
            row[colIndex]=row[colIndex].replace(" ","")

    log.info("Reading emissions from Mobile")
    MobilePath=cf.findExistingPath("Mobile.path:")
    MobileTable=DataTable()
    MobileTable.read(MobilePath,units=True)
    MobileTable=MobileTable.filtered({"Year":str(year)})
    MobileTableKeys=cf.findStringList("Mobile.tableKeys:")
    MobileTable.keys=MobileTableKeys
    #Remove biogenic emissions of CO2
    #Remove GHG from Bunkers, except for international aviation LTO
    for row in MobileTable.data:
#        if "Bio" in row[MobileTable.getColIndex("Codename")]:
        if "Bio" in row[MobileTable.getColIndex("Fueltype")]:
            row[MobileTable.getColIndex("CO2")]=None

#        if row[MobileTable.getColIndex("Localcodename")]=="Bunkers" and row[MobileTable.getColIndex("Localsubcodename")]!="LTO":
        if row[MobileTable.getColIndex("Subsector")]=="Bunkers" and row[MobileTable.getColIndex("Region")]!="LTO":
            row[MobileTable.getColIndex("CO2")]=None
            row[MobileTable.getColIndex("CH4")]=None
            row[MobileTable.getColIndex("N2O")]=None


    log.info("Reading emissions from Manual")
    ManualPath=cf.findExistingPath("Manual.path:")
    ManualTable=DataTable()
    ManualTable.read(ManualPath,units=True)
    ManualTableKeys=cf.findStringList("Manual.tableKeys:")
    ManualTable.keys=ManualTableKeys



    srcTables={'NFR':NFRTable,
               'CRF':CRFTable,
               'Mobile':MobileTable,
               'Manual':ManualTable}

    for tKey in srcTables.keys():
        substList=cf.findStringList(tKey+".substances:")
        substances[tKey]=substList    
    
    log.info("Reading dataSourceTable")
    dsPath=cf.findExistingPath("sourceTable:")
    dsTable=DataTable()
    dsTable.read(dsPath)

    acIndex=cf.findInt("nationalTotals.acIndex:")

    log.info("Reading activity codes from edb")
    natTree=CodeEmisTree("National totals","Gg")
    natTree.readActivityCodes(path.join(edb.edbPath(),"edb.rsrc"),acIndex)    

    log.info("Identifying column indices for sources in dataSourceTable")

    #codeHeaders={"Mobile":"Mobile CRFnodeprefix",
    #             "NFR":"NFR Code",
    #             "CRF":"CRF Code",
    #             "nat":"Code",
    #             "Manual":"Code"
    #             }

    codeHeaders={"Mobile":"Mobile UNFCCC",
                 "NFR":"NFR Code",
                 "CRF":"CRF Code",
                 "nat":"Code",
                 "Manual":"Code"
                 }
    
    codeInd={}
    for key,colName in codeHeaders.iteritems():        
        try:
            codeInd[key]=dsTable.getColIndex(colName)
        except DataTableException:
            log.error("Could not find '"+colName+"' in dataSourceTable")
            sys.exit()
        
    log.info("Collecting data from data sources")
    prevNatCode=""
    for row in dsTable.data:
        rowId={}
        srcRow={}
        for key in srcTables:
            rowId[key]=[]
            srcRow[key]=srcTables[key].ncols*[0]

        natCode=row[codeInd['nat']]
        if natCode!=prevNatCode:
            log.debug("\nProcessing: "+natCode)
            prevNatCode=natCode


        #Get row-id for each src table
        for tKey in srcTables.keys():
            colInd=codeInd[tKey]
            if row[colInd]!=None:
                for key in srcTables[tKey].keys:
                    try:
                        idPart=row[dsTable.getColIndex(tKey+" "+key)]
                    except ValueError:
                        log.error("No column named '"+tKey+" "+key+"' found in dataSourceTable")
                        sys.exit()
                    rowId[tKey].append(idPart)
                #If not all key values = None
                nonNones=[val for val in rowId[tKey] if val !=None]
                if len(nonNones)==0:
                    rowId[tKey]=None
                    srcRow[tKey]=None
                else:
#                     if tKey=="CRF":
#                         pdb.set_trace()
                    try:
                        srcRow[tKey]=srcTables[tKey].accumulate(rowId[tKey])
                    except DataTableException, err:
#                         import pdb; pdb.set_trace()
                        log.error("While trying to get data from "+tKey+" table: "+str(err))
                        sys.exit()


        #Creating code tree path for the current code
        codeParts=natCode.split(".")
        natPath=""
        for i in range(1,len(codeParts)+1):
            natPath+=".".join(codeParts[:i])
            if i < len(codeParts):
                natPath+="/"               
        natNode=natTree.root.find(natPath)
        if natNode is None:
#             import pdb; pdb.set_trace()
            log.error("Could not find national code path : "+natCode)
            sys.exit()

        for subst in substances['nat']:
            log.debug("Substance: "+subst)
            storedSubstances=[]
            foundSubstances=[]
            rec=None

            #Looping over all sourceTables
            for tKey in srcTables:
                st=srcTables[tKey]
                if rowId[tKey]!=None and subst in substances[tKey]:
                    try:
                        colIndex=st.getColIndex(subst)
                    except DataTableException:
                        log.error("Could not find column %s in table %s" %(subst,tKey)) 
                    emis=srcRow[tKey][colIndex]
                    unit=st.desc[colIndex].get('units',None)
                    storedSubstances.append(subst)
                    nodeExists=False
                    if emis!=0 and emis!=None:
                        if subst not in foundSubstances:
                            foundSubstances.append(subst)
                        else:
                            log.error("Substance found in multiple datasources for "+natCode)
                            sys.exit()
                        for node in natNode:
                            if node.tag=="Record" and node.attrib.get("substance",None)==subst:
                                rec=node
                                nodeExists=True
                                break
                        if not nodeExists:                        
                            rec=ET.SubElement(natNode,"Record")
                        break
                    if unit==None:
                        log.error("No units set for substance "+subst+" in source table "+tKey)
                        sys.exit()
            if storedSubstances==[]:
                log.error("No data source specified for substance "+subst)
                sys.exit()

            if rec!=None:
                if emis!=None and emis!=0:
                    try:
                        emis=convertUnits(fromUnit=unit,toUnit=unitDict[subst],val=emis)
                    except TypeError:
                        emis=0
                        
                    if not nodeExists:
                        rec.attrib["substance"]=subst
                    else:
                        emis+=float(rec.attrib["emission"])
                    rec.attrib["emission"]=str(emis)