home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!dove!cam!jhazen
- From: jhazen@cam.nist.gov (Jonathan Hazen)
- Newsgroups: comp.sys.sun.admin
- Subject: Re: deciphering printcap fc,fs,xc,xs fields
- Message-ID: <9963@fs3.cam.nist.gov>
- Date: 28 Dec 92 19:42:45 GMT
- References: <9960@fs3.cam.nist.gov> <16088@auspex-gw.auspex.com>
- Sender: news@cam.nist.gov
- Organization: National Institute of Standards and Technology, Gaithersburg, MD
- Lines: 67
-
- Having read the printcap, ttcompat, and termio manual pages I'm still
- quite confused about how to decipher the following fields from my
- printcap file:
-
- fc#0000374:fs#00000003:xc#0:xs#0040040:
-
- I need to translate these magic numbers into an equivalent ms field.
-
- My first thought was to simply forget about these magic numbers
- and use ms instead. To figure out what to use for the ms field, I
- experimented by sending a postscript file directly out the serial port.
- The following worked perfectly:
-
- (stty -parity crtscts; cat root.ps ) > /dev/ttya
-
- I was able to observe flow control in action
- on my breakout box and the entire postscript file printed correctly.
-
- However, when I used the above stty values as the values for the ms
- field in /etc/printcap, nothing printed, and after a while lpq reported
- a printer error.
-
- Watching the breakout box,
- I saw that a dialog takes place between the Sun and the PostScript printer.
- I compared the dialog that takes place when I use the magic numbers fields
- with the dialog that takes place when I use the ms field.
-
- When I use the magic numbers, the Sun sends something, a second or so
- later the printer sends something back, then transmission of the job
- commences. When I use ms=-parity,crtscts, the Sun sends something, a second or
- so later the printer sends something back but the job doesn't then commence
- transmission.
-
- Thus, my hypothesis is that I need to include some stty input option in
- the ms field so that the Sun can understand the reponse from the printer.
-
- The options I've tried are: igncr, and -istrip. Neither of these in any
- combination work. Sun suggested that I try -cread and -clocal. I'm not
- completely clear what these options do, but they didn't work.
-
- I figure I could spend a long time guessing at what input options are necessary.
- However, clearly the magic numbers configure the port so that this initial
- dialog works. Thus, if someone can tell me how to map the magic numbers
- into the equivalent ms field, I should be able to spare myself a great deal of trouble.
-
- Below is the actual dialog that occurs. As you can see, the
- printer sends back a response with lines separated by CRLF, which is what made
- me try the igncr option.
-
- I sure would appreciate some help, because I'm stumped.
-
- Thanks,
- Jonathan
-
-
- I connected a data communications monitor between the Sun and the printer
- and captured the dialog. The Sun sends:
-
- LF(%%[ pagecount: )print statusdict/pagecount get exec( )
- cvs print(, 2398 1 ]%%)=flushLFET
-
- To which the printer replies:
-
- %%[ pagecount: 10132, 2398 1 ]%%CRLFET
-
- (I've used LF for line feed, CR for carriage return and ET for end of
- transmission.)
-