The native data format used by the Barr Enterprise Print Server is EBCDIC Line Data wrapped in mainframe NJE format. Line Data is simple text with ANSI (ASA) or 3211 machine carriage control. This is essentially the same format used by the JES2 spooling system on IBM mainframes, which is logical because the Barr Enterprise Print Server is, in many ways, emulating a JES2 spooling system.
Most of the conversions performed by the Barr Enterprise Print Server can be logically deduced once you understand this starting point. For example, if you receive a job from a mainframe through BARR/NJE and then print that job to a channel-attached printer through BARR/PRINT390, clearly no conversions will occur because all three devices, the mainframe, the Barr spooler, and the channel printer use EBCDIC Line Data by default. You can also deduce that translations will occur by default when the source or destination is something that does not use EBCDIC Line Data.
The Record-Oriented Inputs and Record-Oriented Outputs in the following diagram are related to emulation of various IBM Mainframe devices, which use EBCDIC Line Data natively. The File-Oriented Inputs and File-Oriented Outputs in the diagram, on the other hand, are all more native to the computer or UNIX An operating system for workstations developed by AT&T Bell Laboratories that features multiprogramming in a multiuser environment. world, and they typically use ASCII data.
This table summarizes the main features of record-oriented and file-oriented interfaces.
Record-Oriented Interfaces |
File-Oriented Interfaces |
Usually associated with IBM Mainframe interfaces like RJE and 3211 emulation. |
Common in the computer or Inter-networking arenas. |
EBCDIC text is assumed. |
ASCII text is assumed. |
EBCDIC Line Data format with ASA or 3211 machine carriage control is used. |
ASCII text with ASCII carriage control is generally assumed, but any format is possible. |
Files are transmitted as a series of records with some agreed upon-record structure (that is, delimiters and carriage control). |
Files are transmitted as a sequence of bytes with no inherent record structure. |
True binary pass-through IS NOT possible. |
True binary pass-through IS possible. |
Four basic data conversion possibilities can be used with the Barr Enterprise Print Server.
A B: Receive from a record-oriented input and send to record-oriented output
A D: Receive from a record-oriented input and send to file-oriented output
C B: Receive from a file-oriented input and send to record-oriented output
C D: Receive from a file-oriented input and send to file-oriented output
Receiving jobs from an IBM mainframe through NJE and printing the jobs on channel-attached printers, such as Xerox laser and IBM line printers.
Little conversion is necessary because both the source and destination are record-oriented. Only the carriage control undergoes conversions between ASA and 3211 machine, depending on the individual case, but the text is left alone by default.
EBCDIC Line Data will pass through cleanly. Schemes that hide more advanced printer commands inside of Line Data also pass through cleanly, such as Xerox LCDS (with DJDEs) and Xerox Metacode.
Some data conversion can be forced using code pages when desired. Here are two examples of when you might want to translate the data.
Sometimes Xerox printers are programmed for ASCII data, even when emulating a 3211 printer. With the Barr software you can specify any code page, including an ASCII one, for any PRINT390 channel-attached printer.
On an IBM mainframe, the JES Acronym for Job Entry Subsystems of the IBM MVS operating system. These subsystems are used for entering jobs into the MVS operating system and dispensing the output from the jobs. can be set to mask all non-printable characters going to a printer and also convert lowercase to uppercase (called a FOLD). In JES2, this is controlled by the TRANS=YES parameter. This behavior can be emulated by selecting a custom EBCDIC code page that leaves the data in EBCDIC, but makes the same changes as TRANS=YES.
Receiving jobs from an IBM Mainframe through NJE or RJE and printing them on computer or network printers. Also, receiving jobs from an IBM mainframe and writing them to a computer or network drive or long-term storage device.
EBCDIC to ASCII translation is the norm in these cases, because the source is almost certainly EBCDIC, and the destination is generally ASCII.
The EBCDIC to ASCII conversion only takes place on the output side. No conversion occurs when receiving, so the job is still in EBCDIC while stored in the Barr spooler.
The EBCDIC to ASCII translation can be customized by marking the job with the appropriate EBCDIC code page. You can use the override table feature to automate this process.
You can achieve the TRANS=YES behavior (to blank out non-printable characters and convert lowercase to uppercase) by selecting the same custom EBCDIC code page as described in the previous section.
The ASCII code page is determined by the printer driver you select and cannot be controlled directly by the Barr software.
In general, only text print jobs (Line Data) will survive the EBCDIC to ASCII translation. However, with careful planning, some printer codes can get through. For example, PCL Acronym for Printer Control Language. Hewlett-Packard uses this language in its LaserJet, DeskJet, and RuggedWriter printer lines. codes are primarily made up of text characters, so it is possible to embed some PCL codes in the data in EBCDIC text.
The Barr File port can be used on the output side to generate special formats. For example, if you want to leave the data in EBCDIC, you can use the Barr File Port to write the file to disk in an EBCDIC fixed-length format.
Receiving jobs from any TCP/IP-based host (UNIX systems, HP 9000s, AS/400s, PCs, and so on.) and printing them on channel-attached printers, such as Xerox Lasers and IBM line printers. Also, printing to the channel-attached printers from the computer or LAN environment.
This case also includes sending print jobs generated in the computer and LAN environment to an IBM mainframe through NJE or RJE.
This case is probably where the most confusion occurs, and it is partly because there is not one clear, standard way of operating.
The critical factor in this case is that the destination is record-oriented, so the source data must be in a supported record-oriented format.
For the file-oriented inputs (section C in the diagram), you must indicate the type of file when configuring the input. This is performed with the Format tab.
The ASCII record-oriented formats supported in this case are listed below. All of these will be converted to EBCDIC Line Data (NJE or .tnj format) as they are read into spool. This process includes an ASCII to EBCDIC translation on the text.
Standard ASCII Text
ASCII Text with ASA (ANSI) carriage control
ASCII spool files from the Barr DOS products
A couple of EBCDIC Line Data formats are also supported with LPD, socket, and Print Utility, and are listed below. In these cases, the records will be re-wrapped in NJE format, with the appropriate carriage control. No translation is performed on the text, because it is assumed the text is EBCDIC or otherwise in the character set expected by the printer. This means that Xerox LCDS and Xerox Metacode can be passed to Barr in these formats.
VBM (Variable Blocked with machine carriage control, a mainframe standard)
Barr S/370 (the format used by the DOS Barr products for EBCDIC Line Data)
One other possibility in this case is the Barr archive format. Data originally spooled to one Barr Enterprise Print Server can be sent out in the Barr archive format, and then received on a second Barr Print Server through LPD, socket, or Print Utility. This method of exchanging jobs preserves both the data and all the header information, and no conversion or translation will occur in the exchange.
Using the Barr Enterprise Print Server to manage your computer and Network printing.
For file-oriented inputs (section C in the diagram), you must indicate the type of file when configuring the input.
This is the only case where true binary pass through is possible because both source and destination are file-oriented.
To configure binary pass-through, select Binary (raw) for the input file type. For your spool printer, from the Advanced Printer Options dialog box, select the Special parameter: Pass data to printer without conversion.
If you do not configure binary pass-through, then, by default, the data will be converted to EBCDIC during input to spool, and then back to ASCII during output. This might be useful in some cases, for example, to convert the data to upper case for a line printer, support ASCII with ASA carriage control, or read in an EBCDIC Line Data file (VBM or Barr S/370) and print it to an ASCII printer.