Vsevolod Stakhov

https://github.com/freebsd/pkg/blob/master/libpkg/pkg_audit.c#L637 - n = 0 https://github.com/freebsd/pkg/blob/master/libpkg/pkg_audit.c#L641 - n = 0, allocate ret of 1 element https://github.com/freebsd/pkg/blob/master/libpkg/pkg_audit.c#L641 - n = 1, access ret[1] -> crash Literally, it is off-by-one error for...

Long time ago, I have designed a library for that purposes: https://github.com/vstakhov/asignify

Yes, exactly! I've just noticed this patch that might have attracted my attention being submitted as a PR. The problem is that identical rules are not a problem for PicoSAT...

Whilst it looks fine I think it breaks POLA principle as users can use the existing symbol name in their composites. There are many poorly named symbols in Rspamd and...

The problem with `task_timeout` is quite known. There are different stages and different timeouts throughout the task processing. Currently, each module has it's own timeout disregarding the existing ones and...

Well, I have been watching closely for IPv6 progress for more than a decade, as I was utterly surprised by such a huge effort to resolve non-existing problem. And so...

> So, no. Support IPv6, end of story. Look, there is a `fork` button in Github - you are free to fork Rspamd and maintain it in the way you...

> We as a company have one /24 IPv4 allocation, and there is little to no prospect of getting any more from RIPE. All our services communicate with each other...

> We see over 70% of email traffic via IPv6 nowadays (in germany, for mostly private customers), all internal infrastructure only uses IPv6. Only the machines at the "edge" have...

> Could be trivially done by connecting each address and retrying with the next on failure. Huh, trivial? Then submit a PR.