The built-in permissions system

Hiho, i read an article on sk's news about the built-in permissions system. It is under development since december 2010, someone stopped working on it, can i or the community have some news about it, @EvilSeph @lukegb @Dinnerbone @Ryan @Selbram @Grum ? would be nice, thanks in advance, andre

 

well it looks like nobody is interested in the built-in permissions... I started a thread also and no reply yet... I guess we will get it and than all people will moan about it... http://forums.bukkit.org/threads/bukkit-permissions-the-original-ones.16687/

 

<3 The yeti 2.7.x

 

because I want to focus the posts in this thread.. a copy from my original thread...


ok after reading http://minecraftnews.sk89q.com/2011/05/the-story-of-bukkit-permissions

I have several questions and perhaps we get some answers and a discussion before it is to late.

a) Is there anything already discussed about bukkit permissions except on sk89q blog?
b) Will Permissions and GroupManager cease to work?
c) Will I need a mySQL/SQLLite interface to edit permission?
d) What will the features be?

I already posted it on the blog webpage, the permissions implementation lags atm, and as nothing specific has been posted about the bukkit permissions (at least how they work from the admin side) I would like to see some improvements being done.

Database support may be ONE, but without a nice way to manage it it is NOT an option for the beginning. You can't expect people to setup mysql and a webfrontend just to edit permissions.

And leaving that aside how will the permissions work?

The way atm to put the users and groups together for every world isn't a good way. There can only be ONE user with a given name on a server and there might be several groups in perhaps different worlds perhaps with different rights (permissions) on the server.

So dividing users and groups into separate "files" and referencing the groups from the user makes the most sense.

So I ask the question again:

so any ROUGH idea when build in permissions are ready and how they will look like?
will it be ONE user.yml per user and one world.yml per world, so that the user.yml can reference groups in the world.yml in the form of world.admins, anotherworld.guests ?
Could you guys give us a better idea what to except when you have finished the framework and start getting rid of the bugs?

(stupid copy wrangling of fonts and size.....)

Also if it is only sqllite and mssql, how will we be able to edit permissions without a webfrontend? Per sql statements or only ingame?

 

nobody interested?

 

I'm very interested, but we don't know a heck of a lot about what is going to happen yet.

There is also this plugin which looks very promissing: http://forums.bukkit.org/threads/admin-dev-permissionsex.13396/

 

I am also quite interested, but I have not time, the best you can do is pm a mod....

Hope someone explains what is going on

 

Yeah, sql based permissions would be so great. I'm tired of having a huge messy yaml file :l

 

I do not like very much mysql... I would prefer sqlite, its backupable...

 

well you can backup everything, the problem is how to get data into the database and which one..

I don't want to run mysql just to run a minecraft server and I don't want to run apache to run a webinterface..

and can you think about adding all the permission nodes via sql statements?

 

I dont think its going to happen any time soon, same as fill.

 

That's what web interfaces are for. The whole reason (or one of) for choosing YAML was that eventually it would be hidden behind a front-end GUI. The backend database could be mysql, slqlite, flatfile, you wouldn't know the difference.

 

Yeah, but requiring a webserver and a sql server just to run bukkit isn't a really good idea as lot of people don't need that setup.

IF and only IF the webinterface and sql would make configuring bukkit a lot easier it would make sense... but atm there isn't a framework for it yet.

A roadmap or at least some ideas where bukkit is going would be nice.

 

I think maybe you missed my point.

I don't think the intention is to require any specific setup like a full blown webserver or mysql. A webinterface is easy to provide, just look at the dynmap plugin, it has an integrated light weight webserver, no need for apache to host your map. Most plugins already integrate with mysql directly, no need to add anything manually. Even then MySQL isn't hard to interface with directly usingweb frontends like phpMyAdmin

And my main point, is that with a flexible front end, it doesn't matter what data base sits on the back end. You could use SQLlite and not have to think about it, or you could install mysql IF you wanted to. Just like it is now (minus the fact that we currently dont have any front end interface other than yaml files or in game commands).

There are already front ends like McMyAdmin that provide a gui interface to manage the server.