Pentax PocketJetII/200 Linux Printer Driver
October 25, 2002

A. Installation instructions
1. run "make" to build the driver
2. run "./Install" and follow screen instructions
3. run "pentaxsetup" and set the printer settings as desired.
4. To choose PJ200 printer instead of PJII, use at the "Advanced" screen
5. For PJ200, there is an option to print 'text only' data. This requires
   that you have a new version of the PJ200 printer. This option will not work
   on older versions of the PJ200 printer.

B. Using the driver with LPD spooler
1. The "pentaxsetup" program will set the printer to use the LPD spooler.
   It will modify the /etc/printcap file, and restart LPD.
   The driver filter is called "pentaxpj".
2. To print from any print-enabled Linux application, you should choose
   to print to "lpr", and select POSTSCRIPT as the data format.
3. IMPORTANT: In any Linux application, you MUST be certain that the PostScript data
   is formatted properly for the same PAPER SIZE that the "pentaxsetup" is configured to use.
   To do this, you should check the options available in the application for setting
   the paper size.
4. When the driver is used this way, it will find the default settings you created with the
   "pentaxsetup" program stored in the file "/etc/pentaxpj.conf". You may see the "command line"
   automatically generated on the very first line of this config file, after the "#".

C. Using the driver at the command line
1. You may NOT use "pentaxpj" at the command line. If it is called this way, then all comand line
   parameters are ignored, and it will load the default configuration from the config file, and send
   the print data to "stdout", which will likely not be desirable.
2. Instead, you MUST use "pentaxpj_sh".
   The driver looks to see if it is called as "pentaxpj_sh", and if so, then it will
   first load the "default" settings from the config file. Then, it will check your extra 
   command line parameters and use these to override any settings from the config file.
   Mostly, you will want to specify the "output port" with "-p device" when using the driver this way,
   otherwise, data will be sent to stdout.

3. Here is a summary of the "command line options" available:
	usage: pentaxpj_sh -[2RPTvdWHopcwlinh] file
	option summary:
	-2         set 200x200 pixel resolution
	-R         reset printer and exit
	-P         print raw pbm file
	-T         text quality (default graphics)
	-d dens    set density
	-f mode    set feed mode (0 (No Feed),1 (Fixed Page), or 2 (End of Page))
	-a lines   "extra feed" for "No feed"(-f 0) mode only
	-W width   set paper width (* 8 pixels units) (use only for custom size page)
	-H height  set paper height in pixels (use only for custom size page)
	-g geom    set ghostscript geometry <width>x<height>
	-o file    send output to a file
	-p device  send output to a device (e.g. /dev/lp0)
	-s paper   set paper size (letter, a4, or legal) (please use this to set standard paper sizes)
	-v, -vv    verbose, more verbose
        -?         list the command line options available
	other options specific of lpd (ignored)

4. Notice that -P will allow printing raw PBM data instead of the normal PostScript data.
This is useful to print already existing bitmap data, such as may be useful for receiving a FAX.


D. Other tidbits about the Driver

1. The printer data format requires "bitmap" data as is specified by the PJII/PJ200 specification.
   So, generally, the driver will take "PostScript" as input, use "GhostScript" to convert the data to raw bitmap,
   and then the driver additionally will format the bitmap as required by the PJII/200 raster language.

2. GhostScript is used to format the data according to a complete pre-defined standard page size, using the GS "geometry".
   Then, data from the bitmap is "discarded" according to the printer's defined "unprintable regions".
   In some cases, valid data may be discarded, but every effort has been made to minimize this data loss.
   On Windows print architecture, the printable area may be specified so that this does not occur.
   Linux printing architecture does not seem to have such a concept available.

3. It is VERY important that the PS data generated matches the paper size as selected by the driver.
   Otherwise, you may see some very strange results with significant amounts of data being discarded.