python - 无法弄清楚导致 IOError 的原因是什么

Question

我已经使用 getopts 编写了一个脚本来接受四个用户输入项(两个输入文件和两个输出文件)。但由于某种原因，我不断收到此错误:

python2.7 compare_files.py -b /tmp/bigfile.txt -s /tmp/smallfile.txt -n /tmp/notinfile.txt -a /tmp/areinfile.txt
compare_files.py -b <biggerfile> -s <smallerfile> -n <notinfile> -a <areinfile> d
 I/O error: [Errno 2] No such file or directory: ''
(<type 'exceptions.IOError'>, IOError(2, 'No such file or directory'), <traceback object at 0x7f10c485c050>)

我无法理解哪个输入项变量给我带来了问题。请有人告诉我哪里出了问题？ 这是脚本:

import sys, getopt

def main(argv):
    binputfilename = ''
    sinputfilename = ''
    outputfilename1 = ''
    outputfilename2 = ''

    try: 
        opts, args = getopt.getopt(argv, "h:b:s:n:a", ["bfile=", "sfile=", "nfile=", "afile="])
    except getopt.GetoptError as (errno, strerror):
        print(str(sys.argv[0]) + " -b <biggerfile> -s <smallerfile> -n <notinfile> -a <areinfile> a")
        print("GetOpts error({0}): {1}".format(errno, strerror))
        sys.exit(2)

    for opt, arg in opts:
        if opt == '-h':
            print(str(sys.argv[0]) + " -b <biggerfile>  -s <smallerfile> -n <notinfile> -a <areinfile> b")
            sys.exit()
        elif opt in ("-b", "--bfile"):
            binputfilename = arg
        elif opt in ("-s", "--sfile"):
            sinputfilename = arg
        elif opt in ("-n", "--nfile"):
            outputfilename1 = arg
        elif opt in ("-a", "--afile"):
            outputfilename2 = arg
        elif len(sys.argv[1:]) > 4 or len(sys.argv[1:]) < 4:
            print(str(sys.argv[0]) + " -b <biggerfile> -s <smallerfile> -n <notinfile> -a <areinfile> c")
            sys.exit(2)
    #smallset = set(open(sinputfilename).read().split())
    #bigset = set(open(binputfilename).read().split()) 

    with open(sinputfilename, 'r') as smallsetsrc, open(binputfilename, 'r') as bigsetsrc, open(outputfilename1, 'w') as outfile1, open(outputfilename2, 'w') as outfile2:
        smallset = set(line.strip() for line in smallsetsrc)
        bigset = set(line.strip() for line in bigsetsrc)
        #find the elements in smallfile that arent in bigfile
        outset1 = smallset.difference(bigset)
        #find the elements in small file that ARE in bigfile
        outset2 = smallset.intersection(bigset)

        count = 0
        for msisdn in outset1:
            count += 1 
            #outfile1.write("%s, %s, %s, %s\n" % (str(count)+".", msisdn)) 
            print(str(count), msisdn) 
#        count = 0
#        for msisdn in outset2:
#            count += 1 
#            outfile2.write("%s, %s, %s, %s\n" % (str(count)+".", msisdn)) 

if __name__ == "__main__":
    try:
        main(sys.argv[1:])
    except IOError, e:
        print(str(sys.argv[0]) + " -b <biggerfile> -s <smallerfile> -n <notinfile> -a <areinfile> d")
        print("I/O error: {0}".format(str(e)))
        print(sys.exc_info())

Answer 1

这是使用 argparse 而不是 getopts 的代码，这应该可以更好地确保指定正确的名称。 (通常，我不喜欢必需的选项；我会使用位置参数。)

请注意，您不需要同时打开这些文件；只打开您需要的那个，并且只在您使用时打开。

from __future__ import print_function
import argparse

def main():
    p = argparse.ArgumentParser()
    p.add_argument("-b", "--bfile", required=True)
    p.add_argument("-s", "--sfile", required=True)
    p.add_argument("-n", "--nfile", required=True)
    p.add_argument("-a", "--afile" required=True)
    args = p.parse_args()

    with open(args.sfile) as smallsetsrc:
        smallset = set(line.strip() for line in smallsetsrc)

    with open(args.bfile) as bigsetsrc:
        bigset = set(line.strip() for line in bigsetsrc)

    outset1 = smallset.difference(bigset)
    outset2 = smallset.intersection(bigset)

    with open(args.nfile, "w") as out:
        for count, msisdn in enumerate(outset1):
            print("%d. %s" % (count, msisdn), file=out) 

    with open(args.afile, "w") as out:
        for count, msisdn in enumerate(outset2):
            print("%d. %s" % (count, msisdn), file=out) 


if __name__ == "__main__":
    main()