SwiftExtract is a standalone executable batch program entirely controlled using its command line, making it trivial to integrate into any other application.  The program is called 'spext.exe' on Windows and 'spext' on UNIX.  The Windows version is a DOS console application, not a Windows GUI program - if you execute it from Windows instead of in a DOS box, nothing interesting will happen!  Here is result from 'spext -h':

The SwiftExtract PCL page extractor version 2.0.0 
Copyright (c) 1989-2003 SwiftView, Inc.
All Rights Reserved

This program licensed to:
ABC Industries
on Tue Jul 17 16:33:57 2001 (995412837)
usage: spext -finputfile [-pstartpagenum] [-nnumpages] -llic [-o[outputfile]]
             -c[cachefile] -t [-i[infofile]] -h -v

   -f  = The input PCL file name.
   -p  = Start extraction at this page number (numbers start at 1).
   -l  = License Password: required if not an evaluation copy.
   -v  = Verify licensing and exit
   -n  = Extract this number of pages; defaults to end of file.
   -o  = Write extracted pages to outputfile if outputfile is given, else to stdout.
   -c  = Private page info cache filename.
         If cachename not given, default to  prefixed with "."
   -t  = Don't check cache file date.
   -i  = Write pcl file info to infofile if infofile is given, else to stdout.
   -h  = Help--print this message and exit.

   example: spext -ftest.pcl -opage10.pcl -p10 -n5 -lYourLicense

   We use the host name/IP address when we issue a license:
   Your host name  is: ourserver
   Your IP address is: 192.168.0.33

Info File

The infofile created by the -i parameter is a simple text file providing information about the original PCL file.  The format of the infofile is as follows: 

It is a standard text file, with lines separate by 'newlines' on UNIX and separated by CR-LF on Windows.

The first line identifies the file and gives the information file version number (not the version number of the program).  Currently, this is the following:
    #SwiftExtract PCL Page Extractor Info File V1.0

All lines starting with "#" should be ignored when parsing the file - they are comments.

All other lines each contain a single set of data of the form:  parameter=value

Values are strings or ascii base 10 integers as appropriate.  Parameter order in the file is undefined except as noted below.

Valid parameters:

numpages=n
The total number of pages in the input PCL file

extractedpages=n
The total number of pages in the output PCL file

offset=n
One of these parameters is given for each page in the extracted PCL file, listed in the info file in extracted page order.  The value is the offset to the byte after the end of each page in the extracted file.  Therefore the first offset is the size of the first extracted page, the second is the size of the first two pages, etc.