IBM Mainframe Forum

Hi,

I have tried a lot and am not able to achieve this.
Using DFSORT SORT step, I have created an xml output file. But if my output file had any of these characters '&', '<', '>', ''' (single quotes) or '"' (double quotes) then the xml fails to open.

Option 1:
If my data is wrapped between '<![CDATA[' and ']]>' tags, then I won't have a problem. But when I did this, the '[' and ']' characters came out as junk characters and xml failed again.



This is what I get:


Option 2:
Instead of using CDATA tags, there will not be any problem, if I replaced:
'&' with '&amp;'
'<' with '&lt;'
'>' with '&gt;'
''' with '&apos;'
'"' with '&quot;'
I tried using CHANGE= option, but if I do not know the position of these characters in my file (say, these characters can be anywhere from 1 to 20th position in my input file), then how do I achieve this?

Can any of you help me with any of my options (or is there any new option for me)?

Much appreciated.

I believe DFSORT is outputting the characters you used as input, but your display of the output can't handle them.

On the input side [ is X'BA'. When I run your job, I get the line with [ as:



and it looks like the [ is missing. But when I display this line in hex with DFSORT, it's:



You can see that the X'BA' is actually there (I put ** below it to identify it).

So I'd say that DFSORT is most likely outputting the characters you told it to and the problem lies somewhere else.

Hi Frank,

You are right. The hex characters X'BA' is in the output, but still, when I FTP the file and open with notepad/excel/IE, it shows something else. So, I tried the other way. I transferred the file with chars '[' and ']' to mainframe and was surprised to see that the char '[' was translated to X'AD' and ']' to X'BD'. Now, I reran my DFSORT with these hex chars and it worked.

Just curious about my option 2. Do you know how to replace the ";" with "&amp;" and so on (as mentioned in the original post) using DFSORT?

Thanks.

Just curious about my option 2. Do you know how to replace the ";" with "&amp;" and so on (as mentioned in the original post) using DFSORT?

If you're asking about finding a character (e.g. '>') anywhere in a record and replacing it with a string (e.g. '&gt;'), DFSORT does not have any built-in functions for that.

Thanks Frank.

If you're asking about finding a character (e.g. '>') anywhere in a record and replacing it with a string (e.g. '&gt;'), DFSORT does not have any built-in functions for that.

DFSORT now has a find and replace capability with z/OS DFSORT V1R5 PTF UK90013. For example:



For complete details on the new FINDREP function and the other new functions available with PTF UK90013, see:

http://www.ibm.com/systems/support/stor ... /mvs/ugpf/