136,137



     the "DIM" statement (see Your Atari 400/800, p.346). Originally,
     it was to be used to implement "common" variables.

     In all cases, the first byte is always one of the number listed on the 
     chart above (you will seldom, if ever, see the undimensioned 
     values in a program). This number defines what type of variable 
     information will follow. The next byte, var # (variable number), is
     in the range from zero to 127. Offset is the number of bytes from
     the beginning of STARP at locations 140 and 141 ($8C, $8D).
     Since each variable is assigned eight bytes, you could find the
     values for each variable by:

     1000  VVTP = PEEK(134) + PEEK(135) *
          256: INPUT VAR: REM VARIABLE NUM
          BER
     1010  FOR LOOP = 0 TO 7: PRINT PEEK(V
          VTP + LOOP + 8 * VAR): NEXT LOOP

     where VAR is the variable number from zero to 127.
     If you wish to assign the same value to every element in a DIMed
     string variable use this simple technique:

     10   DIM TEST$(100)
     20   TEST$ = "*": REM or use TEST$(1)
     30   TEST$(100) = TEST$
     40   TEST$(2) = TEST$: PRINT TEST$

     By assigning the first, last and second variables in the array in
     that order, your Atari will then assign the same value to the rest of
     the array. Make sure you make the second and last elements
     equal to the string, not the character value (i.e don't use
     TEXT$(2) = "*").
     See De Re Atari for an example of SAVEing the six-byte BCD
     numbers to a disk file -- very useful when dealing with fixed
     record lengths.

136,137        88,89          STMTAB
     The address of the statement table (which is the beginning of the
     user's BASIC program), containing all the tokenized lines of
     code plus the immediate mode lines entered by the user. Line
     numbers are stored as two-byte integers, and immediate mode
     lines are given the default value of line 32768 ($8000). The first
     two bytes of a tokenized line are the line number, and the next is
     a dummy byte reserved for the byte count (or offset) from the start
     of this line to the start of the next line.

     Following that is another count byte for the start of this line to the
     start of the next statement. These count values are set only when
     tokenization for the line and statement are complete.