netCDF文件内容提取

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/bnxf00000/article/details/54944797

      很久之前写的一个小程序，当初要提取大量netCDF文件中的部分数据。所以写了一个小程序来进行自动化处理。

      ncBrowse可以用来查看netCDF文件的中的字段和信息。

      我写的程序用到了netCDF4这个包，具体源码如下：

#!/usr/bin env python

#-*- coding:utf-8 -*-

# __author__ = 'ShadonSniper'

from netCDF4 import Dataset

import  os

import argparse

from threading import *

import time

"""

TEMP_FLAG = False

TEMP_ADJUSTED_FLAG = False

PSAL_FLAG = False

PSAL_ADJUSTED_FLAG = False

PRES_FLAG = False

PRES_ADJUSTED_FLAG=False

CYCLE_NUMBER_FLAG = False

PROJECT_NAME_FLAG = False

DATA_MODE_FLAG = False

LATITUDE_FLAG = False

LONGITUDE_FLAG = False

PI_NAME_FLAG = False

POSITIONING_SYSTEM_FLAG = False

"""

def nc2txt(nc_file): #nc_file is your nc file path

    TEMP_FLAG = False

    TEMP_ADJUSTED_FLAG = False

    PSAL_FLAG = False

    PSAL_ADJUSTED_FLAG = False

    PRES_FLAG = False

    PRES_ADJUSTED_FLAG = False

    CYCLE_NUMBER_FLAG = False

    PROJECT_NAME_FLAG = False

    DATA_MODE_FLAG = False

    LATITUDE_FLAG = False

    LONGITUDE_FLAG = False

    PI_NAME_FLAG = False

    POSITIONING_SYSTEM_FLAG = False

    print '[+] File: %s Processing!'%(nc_file)

    nc_fid = Dataset(nc_file, 'r')

    for attr in nc_fid.variables:

        #print attr

        if attr == 'TEMP': TEMP_FLAG = True

        elif attr == 'TEMP_ADJUSTED': TEMP_ADJUSTED_FLAG = True

        elif attr == 'PSAL': PSAL_FLAG=True

        elif attr == 'PSAL_ADJUSTED': PSAL_ADJUSTED_FLAG = True

        elif attr == 'PRES': PRES_FLAG = True

        elif attr == 'PRES_ADJUSTED': PRES_ADJUSTED_FLAG = True

        elif attr == 'CYCLE_NUMBER': CYCLE_NUMBER_FLAG = True

        elif attr =='PROJECT_NAME': PROJECT_NAME_FLAG = True

        elif attr =='DATA_MODE': DATA_MODE_FLAG = True

        elif attr == 'LATITUDE': LATITUDE_FLAG = True

        elif attr == 'LONGITUDE': LONGITUDE_FLAG = True

        elif attr == 'PI_NAME': PI_NAME_FLAG = True

        elif attr == 'POSITIONING_SYSTEM': POSITIONING_SYSTEM_FLAG = True

        else: pass

    PLATFORM_NUMBER=nc_fid.variables['PLATFORM_NUMBER'][:]

    attribute_tsize=nc_fid.variables['PLATFORM_NUMBER'][:].size

    attribute_size = nc_fid.variables['PLATFORM_NUMBER'][0].size

    attribute_sum = attribute_tsize/attribute_size

    if CYCLE_NUMBER_FLAG:

        CYCLE_NUMBER=nc_fid.variables['CYCLE_NUMBER'][:]

    DATE_CREATION = ''.join(nc_fid.variables['DATE_CREATION'][:])+ \

                            '('+nc_fid.variables['DATE_CREATION'].conventions+')'

    if PROJECT_NAME_FLAG:

        PROJECT_NAME=nc_fid.variables['PROJECT_NAME'][:]

    if PI_NAME_FLAG:

        PI_NAME=nc_fid.variables['PI_NAME'][:]

    if DATA_MODE_FLAG:

        DATA_MODE=''.join(nc_fid.variables['DATA_MODE'][:])

    if POSITIONING_SYSTEM_FLAG:

        POSITIONING_SYSTEM=nc_fid.variables['POSITIONING_SYSTEM'][:]

    if LATITUDE_FLAG:

        LATITUDE = nc_fid.variables['LATITUDE'][:]

    if LONGITUDE_FLAG:

        LONGITUDE = nc_fid.variables['LONGITUDE'][:]

    if TEMP_FLAG:

        TEMP = nc_fid.variables['TEMP'][:]

        TEMP_LEN = nc_fid.variables['TEMP'][0].size

    if TEMP_ADJUSTED_FLAG:

        TEMP_ADJUSTED=nc_fid.variables['TEMP_ADJUSTED'][:]

    if PSAL_FLAG:

        PSAL = nc_fid.variables['PSAL'][:]

    if PSAL_ADJUSTED_FLAG:

        PSAL_ADJUSTED=nc_fid.variables['PSAL_ADJUSTED'][:]

    if PRES_FLAG:

        PRES=nc_fid.variables['PRES'][:]

    if PRES_ADJUSTED_FLAG:

        PRES_ADJUSTED=nc_fid.variables['PRES_ADJUSTED'][:]

    nc_name = os.path.splitext(os.path.basename(nc_file))[0]

    nc_dir = os.path.dirname(nc_file)

    txt_name = nc_dir+'/'+nc_name+'.txt'

    txt_fp = open(txt_name,'a')

    txt_fp.write('**********************************************************************************HEADER**********************************************************************************\n')

    txt_fp.write('PLATFORM_NUMBER: ')

    txt_fp.write('\n')

    FormatOutput_Str(txt_fp,PLATFORM_NUMBER,10,attribute_sum)

    txt_fp.write('\n')

    if CYCLE_NUMBER_FLAG:

        txt_fp.write('CYCLE_NUMBER: ')

        txt_fp.write('\n')

        FormatOutput_Num(txt_fp,CYCLE_NUMBER,10,attribute_sum)

        txt_fp.write('\n')

    txt_fp.write('DATE_CREATION: ')

    txt_fp.write(DATE_CREATION)

    txt_fp.write('\n')

    if PROJECT_NAME_FLAG:

        txt_fp.write('PROJECT_NAME: ')

        txt_fp.write('\n')

        FormatOutput_Str(txt_fp,PROJECT_NAME,10,attribute_sum)

        txt_fp.write('\n')

    if PI_NAME_FLAG:

        txt_fp.write('PI_NAME: ')

        txt_fp.write('\n')

        FormatOutput_Str(txt_fp,PI_NAME,10,attribute_sum)

        txt_fp.write('\n')

    if DATA_MODE_FLAG:

        txt_fp.write('DATA_MODE: ')

        txt_fp.write('\n')

        txt_fp.write(DATA_MODE)

        txt_fp.write('\n')

    if POSITIONING_SYSTEM_FLAG:

        txt_fp.write('POSITIONING_SYSTEM: ')

        txt_fp.write('\n')

        FormatOutput_Str(txt_fp,POSITIONING_SYSTEM,10,attribute_sum)

        txt_fp.write('\n')

    txt_fp.write('**********************************************************************************LOCATION*********************************************************************************\n')

    if LATITUDE_FLAG:

        txt_fp.write('LATITUDE: ')

        txt_fp.write('\n')

        FormatOutput_Num(txt_fp,LATITUDE,10,attribute_sum)

        txt_fp.write('\n')

    if LONGITUDE_FLAG:

        txt_fp.write('LONGITUDE: ')

        txt_fp.write('\n')

        FormatOutput_Num(txt_fp,LONGITUDE,10,attribute_sum)

        txt_fp.write('\n')

    txt_fp.write('************************************************************************************DATA***********************************************************************************\n')

    #TEMP related value print

    if TEMP_FLAG:

        print >> txt_fp,"TEMP= "

        #print >> txt_fp,'\n'

        FormatOutput(nc_fid,txt_fp,"TEMP",10,attribute_sum)

        txt_fp.write('\n')

    if TEMP_ADJUSTED_FLAG:

        print >> txt_fp,"TEMP_ADJUSTED="

        #print >> txt_fp,'\n'

        FormatOutput(nc_fid,txt_fp,"TEMP_ADJUSTED",10,attribute_sum)

        txt_fp.write('\n')

    if PSAL_FLAG:

        print >> txt_fp,"PSAL="

        #print >> txt_fp,'\n'

        FormatOutput(nc_fid,txt_fp,"PSAL",10,attribute_sum)

        txt_fp.write('\n')

    if PSAL_ADJUSTED_FLAG:

        print >> txt_fp,"PSAL_ADJUSTED="

        #print >> txt_fp,'\n'

        FormatOutput(nc_fid,txt_fp,"PSAL_ADJUSTED",10,attribute_sum)

        txt_fp.write('\n')

    if PRES_FLAG:

        print >> txt_fp,"PRES="

        #print >> txt_fp,'\n'

        FormatOutput(nc_fid,txt_fp,"PRES",10,attribute_sum)

        txt_fp.write('\n')

    if PRES_ADJUSTED_FLAG:

        print >> txt_fp,"PRES_ADJUSTED="

        #print >> txt_fp,'\n'

        FormatOutput(nc_fid,txt_fp,"PRES_ADJUSTED",10,attribute_sum)

    nc_fid.close()

    txt_fp.close()

    print '[+] File: %s Process Completely!'%(nc_file)

#Traversal nc file

def GetFileFromThisRootDir(dir):

    allfiles = []

    ext='nc'

    for root,dirs,files in os.walk(dir):

        for filespath in files:

            filepath = os.path.join(root, filespath)

            extension = os.path.splitext(filepath)[1][1:]

            if extension in ext:

                allfiles.append(filepath)

    return allfiles

def FormatOutput(nc,fp,attribute,interval,sum):

    len = nc.variables[attribute][0].size

    for i in range(0,sum):

        if i!=0:

            #print >> fp,'\n\n',

            fp.write('\n\n')

        for j in range(0,len):

            #print  >> fp,nc.variables[attribute][i][j],

            #print >> fp,'\t\t',

            if j%interval==0 and j!=0:

                #print fp,'\n'

                fp.write('\n')

            fp.write(str(nc.variables[attribute][i][j]))

            fp.write('\t\t')

def FormatOutput_Num(fp,attribute_list,interval,sum):

    for i in range(0,sum):

        #print >> fp,attribute_list[i],

        #print >> fp,'\t\t',

        if i%interval==0 and i!=0:

            #print >> fp,'\n'

            fp.write('\n')

        fp.write(str(attribute_list[i]))

        fp.write('\t\t')

def FormatOutput_Str(fp,attribute_list,interval,sum):

    for i in range(0,sum):

        result = ''.join(attribute_list[i][:].data)

        #print >> fp,result,

        #print >> fp,'\t\t',

        if i%interval==0 and i!=0:

            #print >> fp,'\n'

            fp.write('\n')

        fp.write(result)

        fp.write('\t\t')

if __name__ =='__main__':

    start = time.time()

    parser = argparse.ArgumentParser(description='Process nc file')

    parser.add_argument('--nc_file',type=str,dest='nc_file',required=True,help='nc file path')

    parser.add_argument('--threads',type=int,dest='threads',default=1,help='set thread number,default is 1')

    args = parser.parse_args()

    nc_path =args.nc_file

    threads = args.threads

    nc_list=[]

    nc_list=GetFileFromThisRootDir(nc_path)

    nc_total=len(nc_list)

    # multithreading process nc file

    quotinet = nc_total/threads

    reminder = nc_total%threads

    all_threads=[]

    if quotinet==0:

        for nc_file in nc_list:

            t = Thread(target=nc2txt,args=(nc_file,))

            t.start()

            all_threads.append(t)

        #t.join()

    elif quotinet==nc_total:

        for nc_file in nc_list:

            #print repr(nc_file)

            nc2txt(nc_file)

    else:

        for k in range(0,quotinet):

            for i in range(0,threads):

                t = Thread(target=nc2txt,args=(nc_list[i+k*threads],))

                t.start()

                all_threads.append(t)

            #t.join()

        if reminder==0: pass

        else:

            for i in range(0,reminder):

                t = Thread(target=nc2txt,args=(nc_list[quotinet*threads+i],))

                t.start()

                all_threads.append(t)

    for t in all_threads:

        t.join()

            #t.join()

    print "Elapsed Time: %s"%(time.time()-start)

    print 'All Process Completely!'

      需要的朋友可参考修改。