DSXchange

I'm reading a EBCDIC coded COBOL file that has a binary field S9(9) comp, using the CFF stage. As as record delimiter the UNIX LF is used. Everything works fine until a record is encountered where the binary field contains a hex 0A value which is interpreted a a LF.
How can I circumvent this error? Because whatever hex value I use for a record delimiter, this value will appear in the binary field some day.

I seem to recall having a similar problem many (many) moons ago when the packed field was the last one in the record. Is that the case here or does it not matter where in the record the comp field is?

Sorry, didn't mean to imply it had been discussed here although it could very well have been. Been too long for me to recall that or whatever in the heck we did to get around that but it would have been with Server jobs and not using the CFF stage.

Hopefully that added bit of detail will spur some other PX folks with CFF experience to chime in.

No problem, Craig. Just solved the problem, I shouldn't have used a lifefeed at all to read the file, I can read the EBCDIC file now.
This post can be closed.