   Start by reading the "Introduction" under "Basics" on the main menu.  Then
we recommend that you first run the simulated demonstration of record and file
locking before executing the code which uses the actual dMILL All-Networks
Productivity Kit library routines.  Choose "Demos" and then "Network
Simulation".

   Before you enter the interworkstation demo, you will be able to set or
verify the values of Kit control memvars.  In particular, you can enable the
various multiple-user tracing options to see how the Kit supports debugging
and verification.  Choose "Demos" and then "Concurrent Demo".

   To execute the code which uses the actual dMILL All-Networks Productivity
Kit library routines, do the following:

 1. Create a directory on the server to hold the files of this demo; for the
    purposes of these instructions we'll assume that the directory is
    identified as C:\KitDemo on the server and S:\KitDemo on the workstations.
    Copy all of the files of the demonstration to this directory.

 2. Give each workstation which is to run the demo read-write access to this
    directory.  The demonstration code erases and creates files.

 3. Execute a "SET NETNAME=<value>" statement at each workstation which is to
    run the demo.  For example, at Frank's workstation, he might enter "SET
    NETNAME=Frank" and Susan might enter "SET NETNAME=Susan".

 4. If the server is not dedicated, you can run the demo on the server by
    making C:\KitDemo the current directory and entering "demomenu".  On a
    workstation other than the server, you can run the demo by making
    S:\KitDemo the current directory and entering "demomenu".

 5. You may invoke the Summer '87 Clipper debugger by typing ALT-D.  You can
    use the debugger to examine the locking activity of the demonstration on
    its one test database with its single test record.  In the debugger,
    select "Display" on the top menu and then select "Database" on the pull-
    down menu.

 6. Note that the Clipper Summer '87 debugger does not show records locked
    following an RLock() of a record in an exclusively-opened file, nor does
    it show a file as locked following an FLock() of an exclusively-opened
    file.  However, RLock() and FLock() return true when entered in the
    "Display Expression" window.

 7. Note that the Clipper 5.01 and 5.2 debuggers do NOT show lock status.  The
    demo is compiled under the Summer '87 version for that reason.

 8. Each workstation must run DOS 3.1 or higher.  NetBIOS is not required, but
    you must load the DOS SHARE program, unless your network operating system
    emulates it.  Novell NetWare emulates SHARE and you need not load it on
    Novell nets.

 9. The locking features of this demonstration code are the same for all
    dialects supported by the dMILL All-Networks Productivity Kits.  For
    example, if you ask for a record or file which is unavailable, the results
    will be the same for all dialects except dBASE III PLUS, which lacks a
    SAVE SCREEN/RESTORE SCREEN facility.  Resource contention messages must be
    displayed in a more manual fashion with dBASE III PLUS; however, the
    message text is the same for all dialects.

   You will experience a difference in so-called "dirty reads" depending on
the dialect.  For example, when it is properly configured, dBASE IV is able to
show changes made to record data on your screen when that data is changed by
another user.

   This demo is written in Clipper 4.0 (Summer '87) and dirty reads are not
fully supported.  When you choose choice "D" (Display record 1.) on the
"Network Locking Demonstration for Two Workstations" submenu, you won't
necessarily see the latest changes made by another user.

   There are two ways to force Clipper to show you the latest data.  First,
you may choose "S" (Use test database KitDemo.DBF in shared mode.) or "E" (Use
test database KitDemo.DBF in exclusive mode.) before choosing option "D".
Second, you may choose option "R" (Lock record 1 and edit it.); if you receive
access to the requested record, you will see the latest data.

               GOOD LUCK AND THANK YOU FOR YOUR SUPPORT!!!
























































