a # 1 61 R ' z # can a;s;l                                               '
a # a,[1]  '"can - edit multiprecision integers into canonical form.     '
a # a,[1]  '"from ''the apl handbook of techniques'' copyright 1978, ibm.  '
a # a,[1]  '"-input a is edited into the canonical multiprecision        '
a # a,[1]  '" integer format as described under function add.            '
a # a,[1]  '"-this function is used by every other function in the       '
a # a,[1]  '" multiprecision integer and floating point arithmetic       '
a # a,[1]  '" packages, except rounds.                                   '
a # a,[1]  '"-uses:  rounds.                                             '
a # a,[1]  ' z#1 rounds a                                                '
a # a,[1]  '"propagate carries leftward                                  '
a # a,[1]  'l1:z#(s,0)+0,z_10000000Xs#(Xz)XD|z%10000000                  '
a # a,[1]  ' $(V/0^=s)/l1                                                '
a # a,[1]  '"drop leading zeroes (but prevent 0$I0)                      '
a # a,[1]  'l2:$(1=Rz#(((0^=-1Uz)I1)_Lio)Uz)/l3                          '
a # a,[1]  '"make all terms (except the exponent) the same sign          '
a # a,[1]  ' $(~V/s#(_l#1Ys)=s#Xz)/l3                                    '
a # a,[1]  ' $l2,Rz#z+(lX10000000Xs)+(1U_lXs),0                          '
a # a,[1]  'l3:z#0,z                                                     '
Lfx a