main

Description of main.py-module. This is the main entry point to the application.

Usage:

usage: main.py [-h] [--lattice_dimension LATTICE_DIMENSION] [--entry_bound ENTRY_BOUND] [--bkz_version {1,2,3}] [--svp_solver {1,2,3}] [--block_size BLOCK_SIZE] [--precision PRECISION]
               [--repetitions REPETITIONS]

Run lattice reduction algorithms.

options:
  -h, --help            show this help message and exit
  --lattice_dimension LATTICE_DIMENSION
                        Desired lattice dimension. (default: 10)
  --entry_bound ENTRY_BOUND
                        Bound for basis entry values (default: 73)
  --bkz_version {1,2,3}
                        Specify the version of bkz implementation: 1: bkz_se, 2: bkz_se_bfp_track, 3: bkz_se_sum_track (default: 1)
  --svp_solver {1,2,3}  Specify the svp_solver utilized during bkz execution: 1: enum_se_og_solver, 2: enum_se_solver, 3: enum_sh_solver (default: 1)
  --block_size BLOCK_SIZE
                        Desired block size for bkz. (default: 5)
  --precision PRECISION
                        Precision of floating point arithmetic: high, default, low. (default: default)
  --repetitions REPETITIONS
                        Number of random bases to operate on. (default: 5)

`positive_integer(value)`

Helper function to check if given integer is positive.

Parameters:

Name	Type	Description	Default
`value`	`int`	Integer to check.	required

Returns:

Type	Description
`bool`	True if value is positive, otherwise False.

`run_bkz(basis, block_size, bkz_version, svp_solver)`

Executes a BKZ (Block Korkine–Zolotarev) reduction on a given lattice basis. This function serves as a unified entry point for invoking one of the available BKZ variants registered in BKZ_ALGORITHMS. The selected BKZ routine will repeatedly call the provided SVP solver on local blocks, update the basis, and maintain Gram–Schmidt data as required by the algorithm.

Parameters:

Name	Type	Description	Default
`basis`	`ndarray`	A 2D NumPy array of shape (n, n) representing a lattice basis, where each column is a basis vector.	required
`block_size`	`int`	An integer specifying the block size in the BKZ algorithm, which determines the dimension of sublattice for the SVP solver calls.	required
`bkz_version`	`str`	A string key selecting the BKZ algorithm variant from `BKZ_ALGORITHMS`.	required
`svp_solver`	`str`	A string key referring to an entry in `ENUM_ALGORITHMS`.	required

Returns:

Name	Type	Description
`bkz_reduced_basis`	`ndarray`	A 2D NumPy array of shape (n, n) representing the BKZ-reduced lattice basis, where each column is a basis vector.

`run_lll(basis)`

Calls the LLL-reduction algorithm.

Parameters:

Name	Type	Description	Default
`basis`	`ndarray`	A 2D NumPy array of shape (n, n) representing a lattice basis, where each column is a basis vector.	required

Returns:

Name	Type	Description
`lll_reduced_basis`	`ndarray`	A 2D NumPy array of shape (n, n) representing a LLL-reduced lattice basis, where each column is a basis vector.