python文件连接和组合文件

Question

My main problem is this:

I have a set of files, and I am concatenating them this way in python:

   sys.stdout=open("out.dat","w")
filenames = ['bla.txt', 'bla.txt', 'bla.txt']
with open('out.dat', 'w') as outfile:
    for fname in filenames:
        with open(fname) as infile:
            outfile.write(infile.read())
with open('out.dat') as f:
    print "".join(line.strip() for line in f)  
sys.stdout.close()

The bla.txt file looks like

aaa

and the intention is to make it look like

aaaaaaaaa

(3 times the same string, not on a new line each time...)

for some reason what I do produces an output that looks like

aaaaaa

a

I am not sure why this is happening and if there is a simpler/more elegant solution.

More second problem is that eventually, my plan is to have a number of different files (letter triplets for example) that I could concatenate in all possible combinations: aaabbbccc,aaacccbbb, ...,etc

Any guidance appreciated! Thank you!

Answer 1

There are some confusing things about your code, I'll leave some comments on the respective places:

# Not sure what is reason for this
sys.stdout=open("out.dat","w")

filenames = ['bla.txt', 'bla.txt', 'bla.txt']

# This does what you need
with open('out.dat', 'w') as outfile:
    for fname in filenames:
        with open(fname) as infile:
            outfile.write(infile.read())

# Here, you open `out.dat` and rewrites it content back into it - 
# because you made `sys.stdout = open("out.dat", "w")` above.
# All these lines could be removed (along with `sys.stdout` assignment above)
with open('out.dat') as f:
    print "".join(line.strip() for line in f)  
sys.stdout.close()

The most minimalistic approach I could think of:

# Open output
with open('out.dat', 'w') as outfile:
    # Iterate over each input
    for infilename in ['bla.txt'] * 3:
        # Open each input and write it to output
        with open(infilename) as infile:
            outfile.write(infile.read())

As for your error, it should not be happening, could you confirm that the content of bla.txt is exactly aaa ?

Answer 2

Nihey Takizawa post almost answers why you've got this error. First, let's see what is going on on each step of the program execution.

sys.stdout=open("out.dat","w")

This is pretty important. Because you replace stdout with file handler to "out.dat", every internal function or statement that use it will write to "out.dat" from now on.

with open('out.dat', 'w') as outfile:
    for fname in filenames:
        with open(fname) as infile:
            outfile.write(infile.read())

After this block, content of the file "out.dat" is:

aaa
aaa 
aaa

...or in other words: aaa\\naaa\\naaa\\n where \\n is single character standing for newline. Number of chars: 12 (9 times a and 3 times newline \\n ).

with open('out.dat') as f:
    print "".join(line.strip() for line in f)

Here is important thing. Remember, that because in step 1 you've changed sys.stdout to "out.dat" internal function print writes output to "out.dat".

You strip each line and join them, so you write "aaaaaaaaa" to "out.dat".

1  2  3  4  5  6  7  8  9 10 11 12
a  a  a \n  a  a  a \n  a  a  a \n  # this is content of the file before print
a  a  a  a  a  a  a  a  a \n       # that you write, 9 a chars + \n
                                   # which is added by print function by default

Note, that you've replaced 10 out of 12 characters and close the file, so 11 and 12 chars would remain the same. Result is your output.

Solution? NEVER mess with things like by changing sys.stdout file handler unless you know what you're doing.

EDIT: How to fix your code. I thought that Nihey Takizawa nicely explained how to fix your code, but it's actually not completely correct as I see. Here's solution:

filenames = ['bla.txt', 'bla.txt', 'bla.txt']
with open('out.dat', 'w') as outfile:
    for fname in filenames:
        with open(fname) as infile:
            outfile.write(infile.read().strip())

Now your out.dat file contains aaaaaaaaa string only without newlines.