Simple script to merge multiple text files

[borthouth]

Hi,

I have just started using Python and I am slowly getting into it. I wanted to make a little script to merge all files in a directory into one.

All of these files will be text files.

I know I need to use os.path.walk() to walk through all the files, and if not os.path.isdir(filename) with open(filename,'r').read() to read all the files that are not directories.

Can someone give me a clue or a very simple example that I can work with please?

Thanks in advance.

[bvdet]

Hi,

I have just started using Python and I am slowly getting into it. I wanted to make a little script to merge all files in a directory into one.

All of these files will be text files.

I know I need to use os.path.walk() to walk through all the files, and if not os.path.isdir(filename) with open(filename,'r').read() to read all the files that are not directories.

Can someone give me a clue or a very simple example that I can work with please?

Thanks in advance.

I think this code will do what you want:

import os
 
def dir_list(dir_name, subdir, *args):
    '''Return a list of file names in directory 'dir_name'
    If 'subdir' is True, recursively access subdirectories under 'dir_name'.
    Additional arguments, if any, are file extensions to add to the list.
    Example usage: fileList = dir_list(r'H:\TEMP', False, 'txt', 'py', 'dat', 'log', 'jpg')
    '''
    fileList = []
    for file in os.listdir(dir_name):
        dirfile = os.path.join(dir_name, file)
        if os.path.isfile(dirfile):
            if len(args) == 0:
                fileList.append(dirfile)
            else:
                if os.path.splitext(dirfile)[1][1:] in args:
                    fileList.append(dirfile)
 
        # recursively access file names in subdirectories
        elif os.path.isdir(dirfile) and subdir:
            # print "Accessing directory:", dirfile
            fileList += dir_list(dirfile, subdir, *args)
    return fileList
 
def combine_files(fileList, fn):
    f = open(fn, 'w')
    for file in fileList:
        print 'Writing file %s' % file
        f.write(open(file).read())
    f.close()
 
if __name__ == '__main__':
    search_dir = "C:/directory"
    fn = "output_file.txt"
    combine_files(dir_list(search_dir, False, 'txt'), fn)

[borthouth]

Thanks , you have helped a lot :)

[ghostdog74]

another way

import os,shutil
f=open("/tmp/fileappend.txt","a")
for r,d,fi in os.walk("/home/me"):
    for files in fi:
        if files.endswith(".txt"):                         
            g=open(os.path.join(r,files))
            shutil.copyfileobj(g,f)
            g.close()
f.close()
 

[borthouth]

I think this code will do what you want:

import os
 
def dir_list(dir_name, subdir, *args):
    '''Return a list of file names in directory 'dir_name'
    If 'subdir' is True, recursively access subdirectories under 'dir_name'.
    Additional arguments, if any, are file extensions to add to the list.
    Example usage: fileList = dir_list(r'H:\TEMP', False, 'txt', 'py', 'dat', 'log', 'jpg')
    '''
    fileList = []
    for file in os.listdir(dir_name):
        dirfile = os.path.join(dir_name, file)
        if os.path.isfile(dirfile):
            if len(args) == 0:
                fileList.append(dirfile)
            else:
                if os.path.splitext(dirfile)[1][1:] in args:
                    fileList.append(dirfile)
 
        # recursively access file names in subdirectories
        elif os.path.isdir(dirfile) and subdir:
            # print "Accessing directory:", dirfile
            fileList += dir_list(dirfile, subdir, *args)
    return fileList
 
def combine_files(fileList, fn):
    f = open(fn, 'w')
    for file in fileList:
        print 'Writing file %s' % file
        f.write(open(file).read())
    f.close()
 
if __name__ == '__main__':
    search_dir = "C:/directory"
    fn = "output_file.txt"
    combine_files(dir_list(search_dir, False, 'txt'), fn)

This seemed to work for a while for me now it comes up with "NameError: name 'False' is not defined"

[bvdet]

This seemed to work for a while for me now it comes up with "NameError: name 'False' is not defined"

>>> bool(0)
False
>>> bool(1)
True
>>> 

'0' and '1' can be substituted for 'False' and 'True' respectively. The question is what happened to 'False' on your system?

[borthouth]

>>> bool(0)
False
>>> bool(1)
True
>>> 

'0' and '1' can be substituted for 'False' and 'True' respectively. The question is what happened to 'False' on your system?

Did a quick reset on my machine and is working now. very strange.