}

  /**
   * Returns an immutable copy of the provided table.
   *
   * <p>The {@link Table#cellSet()} iteration order of the provided table determines the iteration
   * ordering of all views in the returned table. Note that some views of the original table and the
   * copied table may have different iteration orders. For more control over the ordering, create a

  }

  /**
   * Returns an immutable copy of the provided table.
   *
   * <p>The {@link Table#cellSet()} iteration order of the provided table determines the iteration
   * ordering of all views in the returned table. Note that some views of the original table and the
   * copied table may have different iteration orders. For more control over the ordering, create a

For context, there are 3 basic iptables binaries we use.

- iptables
- iptables-save
- iptables-restore

But with the kernel legacy/nft split, there are parallel binaries for each of those, and again for ipv6 tables, so the actual potential binary set on any given linux machine can be some combination of:

- iptables-legacy
- iptables-nft
- ip6tables
- iptables-legacy-save
- iptables-nft-save
- iptables-legacy-restore

which creates documents, for example, SchemaHTML, HistoryHTML
to the "output/doc" directory.

manage.bat(sh) => 23 (generate):
A execution command of Generate task
which generates classes corresponding your tables,
for example, entities, condition-beans to specified
directories by DBFlute properties on "dfprop" directory.

Generated structures (directories and classes) are like this:

apiVersion: release-notes/v2
kind: bug-fix
area: traffic-management
issue:
- 46625
releaseNotes:
- |

all
rule 'MD002', :level => 1
rule 'MD007', :indent => 4
rule 'MD013', :line_length => 160, :code_blocks => false, :tables => false
rule 'MD026', :punctuation => ".,;:!"
exclude_rule 'MD013'
exclude_rule 'MD014'
exclude_rule 'MD030'
exclude_rule 'MD032'
exclude_rule 'MD033'
exclude_rule 'MD041'

        }
      };

  @Override
  protected Table<String, Integer, Character> create(@Nullable Object... data) {
    Table<String, Integer, String> table = HashBasedTable.create();
    checkArgument(data.length % 3 == 0);
    for (int i = 0; i < data.length; i += 3) {
      String value = (data[i + 2] == null) ? null : (data[i + 2] + "transformed");
      table.put((String) data[i], (Integer) data[i + 1], value);
    }

	ProtocolIPv6 Protocol = "IPv6"
)

// Table represents different iptable like filter,nat, mangle and raw
type Table string

const (
	// TableNAT represents the built-in nat table
	TableNAT Table = "nat"
	// TableFilter represents the built-in filter table
	TableFilter Table = "filter"
	// TableMangle represents the built-in mangle table
	TableMangle Table = "mangle"
)

}

// castagnoliShift computes the CRC32-C of K1 or K2 zeroes (depending on the
// table given) with the given initial crc value. This corresponds to
// CRC(crc, O) in the description in updateCastagnoli.
func castagnoliShift(table *sse42Table, crc uint32) uint32 {
	return table[3][crc>>24] ^
		table[2][(crc>>16)&0xFF] ^
		table[1][(crc>>8)&0xFF] ^
		table[0][crc&0xFF]
}

func archUpdateCastagnoli(crc uint32, p []byte) uint32 {

## Create the tables

In this case, we are creating the tables in the same Python file, but in production, you would probably want to create them with Alembic, integrated with migrations, etc.

Here, this section would run directly, right before starting your **FastAPI** application.

* Create an `engine`.
* Create all the tables from the `metadata` object.

```Python hl_lines="25-28"