RFE: long DHCPv4 options (RFC 3396)

[evverx]

Brief description

Long DHCPv4 options longer than 255 bytes aren't supported out of the box. For example

raw(DHCP(options=[('captive-portal', 'a'*256)]))

fails with

error: While dissecting field 'options': ubyte format requires 0 <= number <= 255

but according to https://www.rfc-editor.org/rfc/rfc3396#section-1 it can be split and then it should be joined on the receiving side

when more than
one option with a given type code appears in the DHCP packet, all
such options should be concatenated together.

Wireshark started concatenating options according to that RFC a couple months ago: https://gitlab.com/wireshark/wireshark/-/merge_requests/18079

https://www.ietf.org/archive/id/draft-tojens-dhcp-option-concat-considerations-00.html was also mentioned there.

(It's possible to do all that manually. It's just something I noticed in the context of suboptions in NetworkConfiguration/dhcpcd#404. It came up again when I took a look at #4638 and tried to stuff quite a few IP addresses into a reply :-))

Scapy version

7106b01

Python version

Python 3.13.1

Operating system

Fedora 41

Additional environment information

No response

How to reproduce

raw(DHCP(options=[('captive-portal', 'a'*256)]))

and

DHCP(b'\x06\x02\x01\x02\x06\x02\x03\x04')

should be the same as

DHCP(b'\x06\x04\x01\x02\x03\x04')

Actual result

No response

Expected result

No response

Related resources

No response

[StrugglingKeyboard]

Hi!

I just opened #4950 to address this. The PR implements RFC 3396 option splitting in i2m (for encoding) and option concatenation in m2i (for decoding), along with support for the aggregate option buffer when option overload is present. Tests are included.

This is my first contribution to Scapy, so any feedback on the approach or the code is very welcome!
Many thanks!