Hi Neels,
On 16.04.2023 06:37, Neels Hofmeyr wrote:
I fixed something in SCCP_Emulation.ttcn (from Ericsson), and trying out what I'm allowed to do, it happened so that I just pushed the fix onto https://github.com/osmocom/titan.ProtocolEmulations.SCCP master
Seehttps://github.com/osmocom/titan.ProtocolEmulations.SCCP/commit/17a894fc6620...
The fix is trivial enough, but I'd like to note that I pushed something that no-one reviewed, and hope that's ok.
the patch looks good to me. As I said in the IRC, make sure to update the 'titan.ProtocolEmulations.SCCP_commit' in osmo-ttcn3-hacks.git, so that `make deps` will pull your new commit.
my github user is part of that Osmocom group, I set the remote to a git@... URL, so that's why I could just push onto master.
Now that you mentioned GitHub, where I have no access to anymore, I am wondering if we should host titan.* forks on Osmocom's own Gitea instance rather than on a politically affiliated service like this one? If we agree on that, I could take care of the migration.
For the future, do we have a process to review fixes for Ericsson's ttcn code? Send it to this ML?
IMO, ideally we should try submitting fixes upstream (to [1] in this case), so that the TITAN maintainers can be involved in doing code review too. They're generally open for contributions and already accepted lots of patches from Harald.
Worth mentioning that we have titan.ProtocolModules.BSSMAP in Gerrit, but not titan.ProtocolEmulations.SCCP. We could add it to Gerrit, but I doubt we will see such patches too often, so probably not worth adding.
I am fine with sending patches to this ML and reviewing them here. If we go the Gitea route, we could use its pull request feature.
Just thinking out loud, let's see what others think.
[1] https://gitlab.eclipse.org/eclipse/titan/titan.ProtocolEmulations.SCCP
Best regards, Vadim.