On Fri, 23 Sep 2005 13:19:27 -0500, Chase, John <[EMAIL PROTECTED]> wrote:
>Seems to me that a .PDF should "qualify" as an "unstructured" character >string of arbitrary length, as far as JES is concerned. Whatever utility we >use to "copy" it could "break it up" into, e.g., fixed-length records of >(say) LRECL=133 and a suitable block size, which it indeed does. "Problem" >is that when it writes the fixed-length records, "something" is prefixing >each record with an ANSI carriage-control character. From yours and other >replies, I gather that it is not possible to "copy" the .PDF in the way we >want if the target dataset is a SYSOUT dataset. So, it looks like we'll >have to code a VPS exit to "strip" the first byte off each record to >"reassemble" the original .PDF file. I believe that as other posters have suggested, the sysout writing process does not provide any carriage control by default -- the CC byte has to be provided by the application. If your app (IEBGENER) does not provide the byte, it won't be part of the spool dataset. The SDSF display shows a CC byte, even when it's not part of the spool dataset. The CC behavior you are seeing is most likely being created by a downstream process. For VPS, make sure you specify "NCCSPACE=0" for the printer. This specifies the line spacing for datasets that don't have CC. The default is "1", single spacing, which will cause VPS to insert a CRLF in each line. You may have to play with other VPS options as well, specifically AUTOEJCT, AUTONL & TCPOPTS to get VPS "hands off" on the formatting. Other intermediate players can introduce unwanted "newline processing" as well. For example, if VPS is sending to MS SNA Server, SNA Server will break lines by default. There is an option to control that behavior. Getting a byte stream to pass unmolested through a print process can be tricky. I've worked on several challenging installations. If you need further help, please provide the details of the print process -- VPS COMMTYPE, what VPS is talking to, etc. -Rob ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
