We use the following format to exchange aliquot sequences.

- the first line of the file contains the starting value
- the second line contains the small prime bound B.
  All prime factors less than B are not printed.
- the following lines contain all factors (except the largest one)
  of each term, each factor appearing only once.
- each line can end with a comment starting with # up to the end of the line

As an example, here is the file describing the sequence 88 with B=2:

------------------------------------------------------------------------------
88
2
2
2
2
2
3
1
------------------------------------------------------------------------------

Running the 'aliq' program on this file (say 88.fmt) gives as output:

% aliq < 88.fmt
88:0 = 88 = 2^3*11
88:1 = 92 = 2^2*23
88:2 = 76 = 2^2*19
88:3 = 64 = 2^6
88:4 = 63 = 3^2*7
88:5 = 41 = 41
largest term has 2 digits at index 1

As a larger example, the formatted file 276.fmt for the sequence 276 up
to index 1119 with B=2 uses only 31288 bytes (and 14345 bytes when
compressed with gzip), while the plain file obtained by

% gunzip -c < 276.fmt.gz | aliq

requires more than 150000 bytes (69657 when compressed).

How to compile the aliq program.

1) get and install GMP on your computer
   (see http://www.matematik.su.se/~tege/gmp)
2) compile the program using
   % cc -O -L. aliq.c -o aliq -lgmp -lm
   where the libgmp.a library is supposed to be in the current directory,
   as well as the gmp.h header file.
   If you want to generate Magma output, just add -DMAGMA after 'cc',
   and you'll get:

% aliq < 88.fmt
F := [
/* 88 0 ( 2 ) */
 [ <2, 3>, <11, 1> ] 
/* 88 1 ( 2 ) */,
 [ <2, 2>, <23, 1> ] 
/* 88 2 ( 2 ) */,
 [ <2, 2>, <19, 1> ] 
/* 88 3 ( 2 ) */,
 [ <2, 6> ] 
/* 88 4 ( 2 ) */,
 [ <3, 2>, <7, 1> ] 
/* 88 5 ( 2 ) */,
 [ <41, 1> ] 
];

/*
88 5 ( 1 ) 
1
largest term has 2 digits at index 1
*/