copy mrbgem infomation on mruby wiki to libraries on mruby.github.io

[matsumotory]

I think we should copy mrbgem infomation on mruby wiki to libraries on mruby.github.io for now.

How do you think? If you agree, I'll send pull-request.

[bovi]

What about moving https://github.com/bovi/mgem-list repo under mruby organisation? Then we could generate most parts of the site automatically. I understand that @matz doesn't want to have to many repositories under mruby organisation. So in the end it is up to him.

[matz]

It is OK to have permanent repository. I thought mruby-logo was only useful during the vote.
Shall I create a new repo named mgem-list?

[bovi]

I transfer ownership to mruby. Just a moment!

[matsumotory]

I agree with moving mgem-list. mgem-list should be under mruby organisation.

[matsumotory]

👍

[bovi]

ok. here it is: https://github.com/mruby/mgem-list

[carsonmcdonald]

@matsumoto-r I've started the process of pulling the library information into a page, I just haven't finished it yet. That is what gen_mgemdata.rb does. Github executes Jekyll in safe mode so we can't use a plugin to regenerate it all the time. There are a few options to getting the data to update for the site each time something new is added but I'm not sure what will be best yet.

[bovi]

I saw that @mattn has somehow http://mruby.heroku.com which is able to execute data on the fly. Should we ask him to have a dynamic page which fits into our layout to which we then integrate?

[carsonmcdonald]

@bovi I think something like that is going to be the best option. I'm not sure if it is better to keep the data in the mruby.github.io repo or instead create a remote static file and pull that in using javascript. If the mruby.github.io repo "owns" the data then the remote update has to be able to commit/push an update for every change and that doesn't seem great. Someone could just do it by hand every once in a while but again that doesn't seem great. If we let Javascript do all the work then that makes things easier but I keep thinking it would be nice to keep everything contained within the Jekyll system. Anyone have more ideas/thoughts?

[carsonmcdonald]

I've gone ahead and added a quick list using the data generated from gen_mgemdata.rb so that it is hopefully a little more clear how the parts fit together in Jekyll.

[bovi]

I don't know how well it works with GitHub. But with GitLab I used web hooks to trigger a web service which then processed some shell processes. I saw that GitHub has some kind of Webhooks which might be useful to trigger on a commit at mruby/mgem-list. The hook would then notify mruby.heroku.com (or whatever dynamic webspace is available) to push the data to the mruby.github.io repo.

@carsonmcdonald what is your objective against pushing automatically changes to mruby.github.io? Security? Or are there other problems you are worried of?

[carsonmcdonald]

@bovi that is the track I was thinking of taking and yes it was the security of it that I'm not completely comfortable with. I may be more paranoid than I should be. I think the script is pretty simple so hopefully if @mattn wants to give it a shot he can use it or replace it with something that does the same thing. I'm also willing to set something up but I was going to think on it a bit more.

[bovi]

@matsumoto-r what do you think about http://mruby.github.io/libraries/ ? Is this what you had in mind?

@carsonmcdonald great work. Just notice that all libraries are already online. Now I have to be fast and update the mgem gem to use the new repository ^_^

[matsumotory]

@bovi great! If you have done about this issue, I'll close.

[matsumotory]

We are mruby.glthub.io owners, but I think commits should be pushed by pull-request with a purpose. Then, we review and comment the pull-request. If that is OK, a owner will merge.

How do you think?