如何将变量值临时保留在内存中并在python中进行比较...

Question

Folks, I'm positive that I broke the logic by wrong indentation but now I can't fix it. Could you please help me?
# # analyzeNano.py - analyze XYZ file for 'sanity' # #

import csv
import sys
import os
import getopt

def main():
    '''
analyzeNano.py -d input-directory

analyzeNano.py analyzes a list of XYZ files inside input-directory. It counts for the number of consequitive DNA samples with identical ID and if it between 96 and 110 it treats it as 'good', otherwise 'bad'.
    input-directory    an input directory where XYZ files are located
    -d    flag for input-directory
At the end it creates 2 files: goodNano.csv and badNano.csv
Note: files that are not in goodNano.csv and badNano.csv have no DNA ID and therefore not listed
'''
    try:
        opts, args = getopt.getopt(sys.argv[1:],'d:')
    except getopt.GetoptError, err:
        print str(err)
        help(main)
        sys.exit(2)

    if len(opts) != 1:
        help(main)
        sys.exit(2)

    if not os.path.isdir( sys.argv[2] ):
        print "Error, ", sys.argv[2], " is not a valid directory"
        help(main)
        sys.exit(2)


    prefix = 'dna'
    goodFiles = []
    badFiles = []

    fileList = os.listdir(sys.argv[2])
    for f in fileList:
        absFile = os.path.join(os.path.abspath(sys.argv[2]), f )
        with open(absFile, 'rb') as csvfile:
            # use csv to separate the fields, making it easier to deal with the
            # first value without hard-coding its size
            reader = csv.reader(csvfile, delimiter='\t')
            match = None
            count = 0

            for row in reader:
                # matching rows
                if row[0].lower().startswith(prefix):

                    if match is None:
                        # first line with prefix..
                        match = row[0]

                    if row[0] == match:
                        # found a match, so increment
                        count += 1

                    if row[0] != match:
                        # row prefix has changed
                        if 96 <= count < 110:
                            # counted enough, so start counting the next
                            match = row[0] # match on this now
                            count = 0 # reset the count
                            goodFiles.append(csvfile.name)
                        else:
                            # didn't count enough, so stop working through this file
                            badFiles.append(csvfile.name)
                            break

                # non-matching rows
                else:
                    if match is None:
                        # ignore preceding lines in file
                        continue
                    else:
                        # found non-matching line when expecting a match
                        break
    else:
        if not 96 <= count < 110:
                    #there was at least successful run of lines
            goodFiles.remove(csvfile.name)

    # Create output files
    createFile(goodFiles, 'goodNano')
    createFile(badFiles, 'badNano')

def createFile(files, fName):
    fileName = open( fName + ".csv", "w" )
    for f in files:
        fileName.write( os.path.basename(f) )
        fileName.write("\n")


if __name__ == '__main__':
    main()

Could someone just browse and point me where I broke it?

Answer 1

Here's how I would rework your style:

with open("z:/file.txt", "rU") as file: # U flag means Universal Newline Mode, 
                                        # if error, try switching back to b
    print(file.name)        
    counter = 0
    for line in file: # iterate over a file object itself line by line
        if line.lower().startswith('dna'): # look for your desired condition
            # process the data
            counter += 1

Answer 2

All variables are held in memory. You want to hold onto the most recent match and compare it, counting while it matches:

import csv

prefix = 'DNA'

with open('file.txt','rb') as csvfile:
    # use csv to separate the fields, making it easier to deal with the
    # first value without hard-coding its size
    reader = csv.reader(csvfile, delimiter='\t')
    match = None
    count = 0
    is_good = False
    for row in reader:
        # matching rows
        if row[0].startswith(prefix):

            if match is None:
                # first line with prefix..
                match = row[0]

            if row[0] == match:
                # found a match, so increment
                count += 1

            if row[0] != match:
                # row prefix has changed
                if 96 <= count < 100:
                    # counted enough, so start counting the next
                    match = row[0] # match on this now
                    count = 0 # reset the count
                else:
                    # didn't count enough, so stop working through this file
                    break

        # non-matching rows
        else:
            if match is None:
                # ignore preceding lines in file
                continue
            else:
                # found non-matching line when expecting a match
                break
    else:
        if 96 <= count < 100:
            # there was at least successful run of lines
            is_good = True

if is_good:
    print 'File was good'
else:
    print 'File was bad'

Answer 3

From your description, the lines you're interested in match the regular expression:

^DNA[0-9]{10}

That is, I assume that your xyz is actually ten digits .

The strategy here is to match the 13-character string. If there's no match, and we haven't previously matched, we keep going without further ado. Once we match, we save the string, and increment a counter. As long as we keep matching the regex and the saved string, we keep incrementing. Once we hit a different regex match, or no match at all, the sequence of identical matches is over. If it's valid, we reset the count to zero and the last match to empty. If it's invalid, we exit.

I hasten to add that the following is untested .

# Input file with DNA lines to match:
infile = "z:/file.txt"

# This is the regex for the lines of interest:
regex = re.compile('^DNA[0-9]{10}')

# This will keep count of the number of matches in sequence:
n_seq = 0

# This is the previous match (if any):
lastmatch = ''

# Subroutine to check given sequence count and bail if bad:
def bail_on_bad_sequence(count, match):
    if 96 <= count < 100:
        return
    sys.stderr.write("Bad count (%d) for '%s'\n" % (count,match))
    sys.exit(1)


with open(infile) as file:
    for line in file:
        # Try to match the line to the regex:
        match = re.match(line)

        if match:
            if match.group(0) == lastmatch:
                n_seq += 1
            else:
                bail_on_bad_sequence(lastmatch, n_seq)
                n_seq = 0
                lastmatch = match.group(0)
        else:
            if n_seq != 0:
                bail_on_bad_sequence(lastmatch, n_seq)
                n_seq = 0
                lastmatch = ''

Answer 4

Please ignore my last request to review the code. I reviewed it myself and realized that the problem was with formatting. It looks that now it works as expected and analyze all files in the directory. Thanks again to Metthew. That help was tremendous. I still have some concern about accuracy of calculation because in a few cases it failed while it should not ... but I'll investigate it. Overall ... thanks a lot to everyone for tremendous help.