Add The Elder Scrolls Online support #2376
Conversation
|
I you feel brave, you might try to add something like "ndpi_memmem" in |
Yea, it's a good idea and I thought about it, but the problem is that many simple copypasteable implementations have O(n^2) complexity. Well, if use it as in my case (just to find some sequence in the first 100 bytes of payload), then the complexity doesn't matter, but it's not good for something more serious. |
|
Here is a bit better implementation published under Unlicense, so I guess we can take it. |
|
@0xA50C1A1, could you add the usual entry into |
src/lib/protocols/teso.c
Outdated
| ndpi_set_bitmask_protocol_detection("TES_Online", ndpi_struct, *id, | ||
| NDPI_PROTOCOL_TESO, | ||
| ndpi_search_teso, | ||
| NDPI_SELECTION_BITMASK_PROTOCOL_TCP_WITH_PAYLOAD_WITHOUT_RETRANSMISSION, |
There was a problem hiding this comment.
NDPI_SELECTION_BITMASK_PROTOCOL_V4_V6_TCP_WITH_PAYLOAD_WITHOUT_RETRANSMISSION
There was a problem hiding this comment.
NDPI_SELECTION_BITMASK_PROTOCOL_V4_V6_TCP_WITH_PAYLOAD_WITHOUT_RETRANSMISSION
This game doesn’t seem to use IPv6, but I guess nothing bad will happen if I change selection bitmask to V4_V6.
There was a problem hiding this comment.
Regardless of this specific protocol, generally speaking, it is better to handle ipv6 traffic too, even if right now the server/service is only ipv4, because:
- it might support ipv6 in the future
- we might capture the traffic before a NAT64 translation
Yea. For some reason I thought that I had already done it. |
|
|
The error in the fuzzing is always #2258... |
Please sign (check) the below before submitting the Pull Request:
Describe changes:
In favor of The Elder Scrolls Online's 10th anniversary, which it had 4 days ago. This MMORPG is still popular even nowadays.
I'll leave it as a draft for now, because I'm not sure about using
memmem(). Although this is a GNU extension from glibc, this function is available in BSD, macOS and musl libc, also it's works pretty fast. But I'm not sure about the MSVC has this function, also it's not in the Linux kernel space, which will cause some issues for @vel21ripn's fork.