fix #1898

NS SAREGISTER would fail due to a nil dereference of `client`;
add two safeguards against this.

distrib: add OpenRC init scripts

Signed-off-by: Sam James <sam@gentoo.org>

Written originally for Gentoo but should work
on Alpine and other OpenRC-consuming distros too.

Signed-off-by: Sam James <sam@gentoo.org>

add a warning about NS UNREGISTER unregistering channels

See discussion on #1891

update translations

* version and changelog updates for 2.9.0-rc1

* review fixes to changelog and documentation

add ip-check-script.exempt-sasl

This reverts commit 3b3c5b591f.

disable languages in default.yaml

fix the reopened #1844

* optionally protect against multiple starts with flock

Fixes #1823

* use traditional .lock extension

* move config key to top level

README: Add link tothe LDAP plugin

fix #1876

INVITE did not exempt from +b unless the channel was coincidentally also +i.
This was a regression introduced in v2.4.0.

correctly account for nickname in CAP LS arithmetic

The arithmetic was assuming that the nickname is * (which it is
pre-registration). However, we were sending the actual nickname
post-registration. It would be simpler to always send *, but it
appears that the nickname is actually required by the spec:

>Replies from the server must [sic] contain the client identifier name or
>asterisk if one is not yet available.

+I should allow unregistered users to join a +R channel

See #1858: this was the intent all along, but I missed this issue.

fix the atheme side of #1864

Import stored certfps from Atheme as well.

There is no change in behavior since committing the transaction
already write(2)'s all the data to disk. But let's comply with
the official buntdb API.

This should have been done in 8b2f6de3e0, since we updated both
the database schema and the importer then.

* Advanced certfp support

Signed-off-by: Georg <georg@lysergic.dev>

* Moving certfp logic

Signed-off-by: Georg <georg@lysergic.dev>

* Cleaning up certfp logic

Signed-off-by: Georg <georg@lysergic.dev>

send `*` for WHOX o (oplevel) instead of `0`

Jobe points out that 0 is a valid oplevel in some contexts,
* is a better placeholder for "unimplemented".

anope2json, atheme2json: handle non-UTF8 data

Also ignore an unrecognized field type in anope

fix #1860