Victor Julien

I still need to review the commits affecting code, but a couple of comments already: 1. I'm fine with all commits in a single PR 2. I think we can...

One concern I have is around expectations of matching each record on the wire vs the "transaction" representation we have in Suricata, that are not necessarily a 1 on 1...

Think it would be nice to use the kernel "fixes" mechanic here: ``` $ git fixes 53fc70a Fixes: 53fc70a9a73c ("protodetect: fix int warnings") ``` where `fixes` is an alias: ```...

> > Think it would be nice to use the kernel "fixes" mechanic here: > > ``` > > $ git fixes 53fc70a > > Fixes: 53fc70a9a73c ("protodetect: fix int...

Merged in #7730, thanks!

> @victorjulien is this part of the stuff you are working on ? No, working on the address side of rule parsing only.

> Thanks for this contribution. I uncommented the reference to the Suricata-Verify PR so that it gets run with CI I fixed your edit :-P

I need to look at this again. I looked earlier w/o understanding the logic and not seeing any test cases pass/fail in relation to this PR. SSL3/TLS records can certainly...

@catenacyber can you dig a bit deeper into this?

CI will need fixing.