IBM Mainframe Forum

Hi All,

I am transfering some files using ftp JCL . On the Unix side I see that some spanish characters are showing as some escape character. I need to load this file into Oracle database. I tried Manually converting from ebcdic to UTF but I am loosing delimited values after the conversion using the DD command and iconv. so My load to Oracle db is failing. Can someone help me what parameters should I set in the FTP JCL so that EBCDIC to ASCII or UTF8 conversion is taken care by the FTP on the mainframe side.

Thanks,
Nagesh.

Hi Used ascii in the ftp JCL, But still the spanish characters are showing as Unicode Character 'NOT SIGN' (U+00AC) .

And, as Mark Zuckerberg would call you, but than a lot stronger, dumb piece of homo sapiens, what have your questions to do with the "Forum Rules" forum?

Standard ASCII does not have the Spanish characters, so you are asking for the impossible if you want the Spanish characters reflected in standard ASCII.

Start reading https://www.ibm.com/support/knowledgece ... ftpcon.htm as you need to indicate to the system to do a translation from EBCDIC to (most likely) code page 284 to allow use of the Spanish characters. Hint: find out about the QUOTE LOCSITE SBDATACONN(1037,284) subcommand of FTP.

And, since you do not seem to know this, FTP text transfers from a z/OS system to a Windows or Linux machine will AUTOMATICALLY convert EBCDIC to (standard) ASCII -- nothing is needed to do that. When you want non-standard ASCII, such as Spanish characters, you MUST either translate on the mainframe before the transfer and then do the transfer as a binary transfer, or indicate to FTP that you want the translation to be something other than to standard ASCII.

Thanks alot Rob. for quickly answering this. I appreciate all your help.

Thanks,
Nagesh.

If you are talking to me, my name is "Robert" -- not "Rob".

Sorry Robert. Thanks alot for quick reply. I still had no luck . I am using below syntax in FTP. Can you help me if I am missing any options that needs to be changed at TCP parms on ZOS system.

servername
username
password
locsite sbd=(IBM-1047,ISO8859-1)
cd target direcroty
put 'dsaname.on.zos' gg.test.data
quit

IBM-1047 is IBM EBCDIC (at least, one of the versions of IBM EBCDIC) while ISO8859-1 is standard ASCII. I would think you would want SBD(IBM-284,ISO8859-1) to get Spanish data translated to standard ASCII. Is your mainframe data using code page 284 already?

Hi Robert,

Thanks alot for quick reply. Yes. My mainframe dataset already showing Spanish characters clearly without any errors when I see through browse. I am not sure how to check my mainframe code page. Can you give me the command so that I can check.

Thanks,
Nagesh.

Hi Robert,

I appreciate all your help. Using SBD(IBM-284,ISO8859-1) had changed N tilde(Spanish) character to # at the destination and other # to N tilde(Spanish). I want to send the Mainframe dataset to linux server without changing the spanish characters . Since we have many files to transfer we are planning to automate this process, But I am not able to get the FTP syntax correct. Even if we get at destination with UTF8 , that would also suffice. when I used SBD(IBM-284,UTF-8) I get message saying not all characters are transalated from 047 to utf8.

Thanks,
Nagesh.