Download through FTP from Mainframe

Question

I am trying to parse / download some of the files from Mainframe using ftplib but it's unsuccesful after few attempts.

My code Till now is :

import ftplib
ftp = ftplib.FTP('host','username','password')
ftp.retrlines("File To be Downloaded")

This works fine and I can see the required file.

However when I use following code.

ftp.retrbinary("RETR 'File_Name'",open('ww.txt','wb').write)

It fails with following error.

error_reply: 200 Representation type is Image

Based on reponse of Dag, I tried the following code to download file through "retrlines".

ftp.retrlines("RETR 'File_Name'",open('ww.txt','wb').write)

But it gives me following error.

error_reply: 200 Representation type is Ascii NonPrint

Answer 1

What is the expected result? Retrlines reads the file as text, which is correct for text files. Retrbinary retrieves the file as a binary file. This is a bad idea if you are targetting a mainframe using EBCDIC instead of ASCII, and might be the reason you get this result code (which according to FTP spec is not an error).

Answer 2

Try explicitly setting "TYPE" in your Python script:

http://publib.boulder.ibm.com/infocenter/zos/v1r12/index.jsp?topic=%2Fcom.ibm.zos.r12.halu001%2Ftype.htm

I'm guessing unless you specify a type, the default is probably "ASCII". A Pop Favorite for most binary FTP transfers is "BINARY". I'm not sure what's most appropriate for your system: "IMAGE", perhaps?

Also: look at this link: Downloading text files with Python and ftplib.FTP from z/os.