Question

For the following working code, I resorted to creating a class instance to store the name variable of my file output [a_string] and the file object itself [f_object]. I found that variables assigned inside the first if statement did not appear in the scope inside the following elif statement.

#Text file splitter, data bewteen the '*' lines are copied into new files.

class Output_file():
    def __init__(self,a_string='none',f_object='none'):
        self.name=a_string
        self.foutput=f_object

outputfile=Output_file()

n=0
filehandle=open('original file.txt')
for line in filehandle:

    if line[0]=='*': #find the '*' that splits the rows of data
        n+=1
        outputfile.name = 'original file_'+str(n)+'.txt'
        outputfile.foutput= open(outputfile.name,'w')
        outputfile.foutput.write(line)

    elif len(line.split()) ==5 and n > 0: #make sure the bulk data occurs in blocks of 5
        outputfile.foutput= open(outputfile.name,'a+')
        outputfile.foutput.write(line)


outputfile.foutput.close()

Do I have to use a class instance to store the file name and object or is there a better way?

Answer 1

Variables defined in either if or elif statement should appear in the other. For example:

>>> for i in range(5):
...  if i%2==0:
...   x = i
...   print(x)
...  else:
...   print(x)
... 
0
0
2
2
4

This would not be the case in block-scoped languages, but unfortunately python is not block-scoped, and thus this should work.

Do note however that for this to work, your name= must execute before you attempt to use it. That is, your if statement MUST execute at least once before your elif statement.

Your code lacks comments, but I assume your data looks somewhat like:

 This is a header
 blah blah blah
 **************
 16 624 24 57 32
 352 73 47 76 3
 25 6 78 80 21 331
 **************
 234 234 4 64 7
 **************
 **************
 86 57 2 5 14
 4 8 3 634 7

And you want to split them into separate files, but only if it’s “valid” data. I’d code it like so if I wanted to imitate your style:

def isSeparatorLine(line):
    return line[0] = '*'
def isValidLine(line):
    return len(line.split())==5

groupNum = 0
outputFile = None
with open('original file.txt') as original:
    for line in original:
        if isSeparatorLine(line):
            groupNum += 1
            outputFilename = 'original file_{}.txt'.format(groupNum)
            if outputFile:
                outputFile.close()
            outputFile = open(outputFilename, 'w')
            outputFile.write('New file with group {}'.format(groupNum))
        elif group>0 and isValidLine(line):
            outputFile.write(line)

I’d however personally prefer to write it like so:

from itertools import *

FILENAME = 'original file.txt'
FILENAME_TEMPLATE = 'stanza-{}.txt'

def isSeparatorLine(line):
    return all(c=='*' for c in line)
def isValidLine(line):
    return len(line.split())==5
def extractStanzas(text):
    """
        Yields: [stanza0:line0,line1,...], [stanza1:lineN,lineN+1,...], [stanza2:...]
        where each stanza is separated by a separator line, as defined above
    """
    for isSeparator,stanza in groupby(text.splitlines(), isSeparatorLine):
        if not isSeparator:
            yield stanza

with open(FILENAME) as f:
    stanzas = list(extractStanzas(f.read()))

for i,stanza in enumerate(stanzas[1:]):
    assert all(isValidLine(line) for line in stanza), 'invalid line somewhere: {}'.format(stanza)
    with open(FILENAME_TEMPLATE.format(i), 'w') as output:
        output.write('\n'.join(stanza))