FILE(<numeric expression from 0 to 63>)
This function is used to specify the number of the file
being used.  It is not used when loading or saving programs,
only data.  The initial file number is zero.  Since the file
number stays the same until the FILE function is used, it is
not necessary to use this function until it is required to
access a data file different than the last one accessed.
The file number can be any arbitrary number from 0 to 63.
It should be used in the OPEN statement for the file, and
any time a different file number is required.
Examples:  OPEN FILE(34),DISK(1),3,"DATAFILE"
	   PUT FILE(34),TESTDATA$
	   GET FILE(34),TESTDATA$
	   CLOSE FILE(34),3

* TYPE(<numeric expression of 0 or 1>)
The value of the expression is placed in location TYPE,
described in appendix C.  This function is normally used to
specify the type of file which is being OPENed.  The
following conventions will be used by the I/O sections
provided by Tarbell Electronics:  0 for sequential,
1 for random.  TYPE is initialized to 0 by the I/O section.
The TYPE function only has an effect during an OPEN
statement.  The TYPE of a program transfer (LOAD, SAVE) is
always zero (sequential).
Examples:	SEQ=0:RAN=1
	OPEN DISK(1),FILE(7),TYPE(RAN),RECORD(80),2,"RANFILE"
	GET FILE(7),RECORD(N),X,Y$,Z

RECORD(<numeric expression>)
The value of the expression is placed in location RECORD,
described in appendix C.  This function is normally used to
specify the record number of a random file, as part of a
GET or PUT operation, or to specify the number of bytes
per record, as part of an OPEN operation.  When the RECORD
function is used as part of a GET or PUT statement, the
transfer begins at a location in the file determined by
decrementing the record number by one, then multiplying it
by the record size (set in OPEN statement).  RECORD numbers
start at 1, so RECORD(1) would always start with the first
byte in the file.  If RECORD(0) is used, or if the RECORD
function is not used with a random file, the next character
transferred will be immediately after the last character
transferred, similar to a sequential operation.  See the
example above.

A random file can be thought of as a continuous stream of
characters, with a dividing line at each n characters,
where n is the number of bytes per record, specified by
the RECORD function in an OPEN statement.  The records
are the groups of characters between the lines.  Record
number 1 is the first group, record 2 the second group,
and so on.  The random feature allows the programmer to
specify any record number for immediate access, without
having to search through the whole file.




				5-2