Let's consider - Auto Plugin Converter

here’s a key thing to the idea.
the api comes out way before the implementation does.
so the question when it comes to porting plugins is as to why rush something that doesnt need to be rushed?

Personally:
I wrote ~80 plugins, and ~14 of them needs to be ported. Some of them are large, and I am still adding new contents to it continuously.
Wait we have Pour.
But… Is it a good idea to stick with Bukkit API + Pour?
Things will change, someday, eventually.

I see it as a bandaid to help get sponge off the ground

love that banhappy tag.


i can’t really see it as that much of a bandaid with the whole release eta thing, honestly. enough plugins can be pushed out before the first version is released for it to be really necessary as anything other than a revival utility to me, but its good to have it nonetheless.

in any case i’d leave it as a matter of preference whether or not people use Pore or rewrite their plugin. its their project and it’ll be their job to fix up what they need to if such problems arise from using it.

It’s not ‘Pour’ it’s ‘Pore’

If Pore can be 90% effective, I see that as a better route than writing a plugin converter. I think we will have to wait and see how well the mapping will go once the Sponge API is in a more complete form (which should be soon).

2 Likes

I think that it’s a great opportunity for developers to get familiar with the API. If you have a converter do all the work, then people will never learn to use the Sponge API properly :wink:

That might be the case, but the chances of a machine creating unreadable code are very high. The converter would have to insert newly generated code in places where Bukkit’s API doesn’t match Sponge’s API. Generally, the problem with generated code is that it’s not nice to read. Using “preprogrammed” patterns doesn’t solve the problem as well, because the combination of those patterns might also create unreadable and inefficient code.

Personally, I would rather focus on creating Pore as a temporary mediator because its easier to develop and the code created by a converter probably wouldn’t be usable.

By that I meant porting, not a converter.

Oh, sorry; I accidentally missed the word “never”. Nevermind!

no problem :smiley:

If pore and this actually become a thing then this will become the backup to pore.

Just use pore! You don’t need to use convert!

We don’t even know if pore will work

It’s made by LapisBlue of course it will work!