Python ftplib does not download completely

Question

I am using python ftplib to download builds from the builds ftp server. The files are about 1.5-1.6 GB. I use a batch file to run the program. This way it is easier to schedule the download. The problem is, downloaded files do not seem to work. They throw a compatibility error(windows). If I download the files using FileZilla, the files work fine. Also, there is a difference of a few hundred B on the source file and the downloaded file. What is going on?

import ftplib, sys, os

ftp = ftplib.FTP("<server_name")
try:
    ftp.login(user= "<user>", passwd = "<password>")
except:
    sys.stderr.write('Could not login.')

data=[]
ftp.dir(data.append)
builds=[]

trg=0           
trg_bld=""
for i in data:
    if len(i.split(" "))>12:
        if len(i.split(" ")[12].split("_"))>2:
    #this line is to find the version on the server
            if (i.split(" ")[12].split("_")[1]== "3.1.0"):
                if int(i.split(" ")[12].split("_")[2])>trg:
                    trg_bld = i.split(" ")[12]

trg_file = trg_bld
print trg_file
if os.path.isfile(trg_file):
    sys.stderr.write('File already exists.')
    sys.exit(1)
f= open(trg_file, "w")
ftp.retrbinary('RETR '+trg_bld, f.write)
f.close()
ftp.close()
sys.stdout.write("File download successful.")

Batch file: E: cd E:\Builds python ftp_sch.py pause

Answer 1

f= open(trg_file, "w") -> f= open(trg_file, "wb")

Without the b flag Python thinks you're writing ASCII and it's changing the line-endings (hence the size difference and broken binaries).