function mload,nums,F12=F12,yaxis=yaxe,norm=norm,mask=maskstring,multiy=multiy,tol=tol

 LAMP function to create a workspace from a series of datasets with different flipper states

NECESSARY VARIABLES

                nums=EITHER a string with the filenumbers containing the data.

                                Consecutive files are separated by a colon (:)

                                Adjacent files are separated by a comma (,)

                                The string must be enclosed in quotation marks (')

                                e.g. '90230:90240,90250:90260'

                     OR an array of numbers

OPTIONAL VARIABLES:

                F12 = two element array with flipper states (=0 for off, =1 for on)

                                e.g. F12=[0,1] is F1=off, F2=on

                 norm=your choice of normalization, 'time', 'mon', or 'none'

                                                The default (if no scantime is entered) is 'time'

                yaxe=your choice for the yaxis, 'san','dan','tra', or 'ran'

                                                The default (if no yaxe is entered) is 'san'

                maskstring=your choice of masking the detector, written as a string

                                                The default is no mask

NOTE:  The data is NOT normalized to the number of y-pixels in the mask, i.e. circular and elliptical masks will have less weighting away from the centre.  Divide by a masked water run to perform this action

                multiy=your choice to combine y data with the same SAN/DAN value or not

                                A string variable, ='av' to combine, 'noav' to leave uncombined

                                Default is 'av'

                tol=a tolerance in y for the y-combination. 

                            Default is 0.004 the appropriate y-axis units

OUTPUT VARIABLES

                wout is a workspace with the data of the desired flipper states.  If the data is 2D, it will be integrated vertically 

                                The X-axis is given in pixel number

                                The Y-axis is given as SAN or DAN, user's choice

                                The Z-axis is the flipper state.  If flippers are used, the resulting array will be 3D, each layer containing data of a specific flipper state.  If a flipper that is 'off' is labelled with a 0 and a flipper that is 'on' is labelled with a 1, the arrays will be stacked in binary fashion.

                                e.g. if all four flipper states are used,

                                              The first value of Z will be 00 (flipper1 = 'off', flipper 2  = 'off')

                                              The second value of Z will be 01 (flipper1 = 'off', flipper 2  = 'on')

                                              The third value of Z will be 10 (flipper1 = 'on', flipper 2  = 'off')

                                              The fourth value of Z will be 11 (flipper1 = 'on', flipper 2  = 'on')

                                The monitor is given as time, monitor, or as unity (for 'none'), user's choice

                                There is a new array of variables in the 'pv' field.  This is a M rows * 8 columns array, where M is the number of steps in the scans.  The columns are:

                                   [SAN, DAN, mean date, counting time, monitor, normalization factor, S2W, S3W]

                                The normalization factor should be multiplied with the data for any further operations (e.g. addition of workspaces) and the errors for the correct propagation of errors

                e.g.. type at the LAMP interface:

     w1=mload('68263:68300')

               This will read all the scans from runs 68263 to 68300.  It will sort the data by flipper state, integrate vertically over the detector (if necessary), normalise the data to time, average all data at the same SAN and return a 2- or 3D array (depending on how many flipper states were used) of the data with x-pixel vs. SAN.

NOTE:  If more than one flipper state was used, the number of measurements for each flipper state must be EQUIVALENT!  If they are not, the number of rows in the data arrays will not match and mload will complain.

     w1=mload(68263+indgen(37),F12=[0,0])

               Similar to above, but will return a 2D array containing only the data with both flippers off.

     w1=mload('68263:68300',F12=[0,0],yaxis='san',norm='time',mask='box,[190,200,63,222]',multiy='av')

               Similar to above, with a box mask applied to the 2D detector before integrating over the vertical direction.