Conference cookie::mru

Another user has tried running MRU V1.1 with JUDriver.  I will have to make
it as BOLD AS POSSIBLE and RIGHT UP FRONT that users shouldn't do this.

	The behavior of the MRU_UNLOAD_DELAY logical has been
	changed to match the original intent.  MRU_UNLOAD_DELAY
	is the number of seconds that mrd_unload() will retry
	trying to remove a tape from a drive.  An elapsed time
	loop is used so that it doesn't matter if each command
	takes 1 second or 10.  The default is the intended 180
	seconds.

	A new logical, MRU_READY_DELAY, has been added.  When
	the Inport door is closed after a Ready IOD, a SCSI
	Check Condition is generated, eventually.  If this
	Check Condition is seen by a Move Medium command, the
	command may appear to fail, even when it succeeds.  To
	work around this, the VMS implementation of Ready IOD
	waits for the Port Accessed condition to appear before
	it returns.  The amount of time to wait is MRU_READY_DELAY,
	with a default of 65 seconds.

	The timer ensures that it doesn't wait for ever, in case
	the Check Condition doesn't show up.  But, it is also
	long enough to allow time to open the door, put a tape
	in and close it towards the end of the one minute that
	the light stays on.

	I've added stubs to the Errors help file for the following
	new messages:

		Errors Audit Mouse2
		Errors Audit Mouse3
		Errors FindMachineBroken

	In case they weren't mentioned previously, I also added
	audit messages and errors messages for Position, Home
	and Find.  I think the new Show reused existing messages.

	Here are some more messages and a description of what caused
	it.  I'm updating the Errors. file with tags, titles and
	basic stub text.

	Errors EnterVolTag

		When a Find is first started, a dialog is started
		to prompt for the volume tag that is to be used as
		the search key.  If an empty string is provided to
		this dialog, this message is displayed.  The dialog
		is not dismissed.

	Errors AlreadySearching

		Only one "Find" is allowed a time.  This message
		occurs when an attempt is made to start a 2nd
		search.

	Errors CantFind

		After an exhaustive search the desired volume
		tag couldn't be found.

	Errors CantElement

		A Show Element was attempted, but the interface
		could determine for which element.  Each element
		should have a brick_t data structure associated
		with it, but the brick for this element was NULL.


		

	The GUI and CLI both support a new logical; MRU_CASE_SENSITIVE.

	If this logical is not defined, volume tags (where used) will
	be converted to uppercase.  If it is defined, the case of
	the volume tag will be preserved.

	In the CLI, this is used for arguments provided via the
	CARTRIDGE qualifier.

	In the GUI it is used for the Cleaning tape prefix and
	the Find command.

	Do we want to fully document uses of the CARTRIDGE feature?

	On the commands, Load, Unload, Inject, Eject, Move and Home
	if the CARTRIDGE feature is used, the CLI will check that
	the named cartridge is actually at the source element.  If
	not, the command will fail.  Example:

		robot move slot 2 port 0 cart DEC000
	or:

		robot load slot 15 drive 0 cartridge CLN967

    Just a slight change to the installation guide for VMS.
    
    Under the section on DAT and DLT loaders, the guide says that
    the loaders always show up at LUN 1 (i.e. GKA201).  This is no longer true.
    The not-yet-released loader for the TZ89 shows up at a separate 
    SCSI ID, like the TL8xx libraries do.

	Then we'll call it a library...  And this this yet unannounced
	device has a bar-code reader, it is much more like the larger
	libraries than it is the previous loaders.

FIXED: robot.8, ROBOT$HELP.HLP, MRUINFO.PS


FIND CARTRIDGE {tape_label} [ROBOT robot_name]
   
    Locates medium identified by tape_label.

-s