automate Chapter wiki creation (eventually blogs too)

[Nelson Pavlosky]

I should be able to create a wiki for a chapter without bugging Asheesh, and it should spawn customized for the chapter, using information from the Chapters database.

* I've created a series of Templates on the example wiki, including {{Chapter subdomain}}, {{Chapter name}}, {{School url}}, and {{School name}}. Each of these templates should be filled in using the data in the appropriate field from the entry in the chapters database when the wiki is created.

* There should be some web interface or command line interface, where I can specify which chapter I want to create a wiki for, and (if the chapter is active), I will be allowed to create a wiki for it using the info from the chapters database.

[Nelson Pavlosky (nelson-freeculture)]

If it would be easier to separate these steps, we could start off by simply creating a script which takes an existing chapter wiki (which Asheesh has spawned in his magic way) and fills in the templates from the database. I would like that a lot, and I think we should prioritize it.

[Parker Phinney (gameguy43)]

just a small note: make sure that you also set the icon for the wiki (the image at the top left) to be the fc.o "legos" by default.

[Nelson Pavlosky (nelson-freeculture)]

This is semi-important b/c otherwise Asheesh is the bottleneck for getting new chapter wikis created, and I'm sure Asheesh dislikes being bugged as much as we dislike bugging him ;-) But as long as Asheesh doesn't mind creating new wikis when we ask him this isn't critically important.

[Parker Phinney (gameguy43)]

update:

so we have the wikis on a farm, and the blogs are being handled by lyceum. we have a process by-which one can create chapter webspace:
http://wiki.freeculture.org/Website_documentation#Chapter_hosting

it works, and it's not all that difficult, but it does require some comfort with the command line.

i'm leaving this bug open because this process could probably be much more automatized if someone wrote a nice python script or something.

[The Half Hollow Hacker (halfhollowhackers)]

Well, I am trying to write that "python script" As soon as I can get into the MediaWiki API, everything should be alright. I only have MediaWiki and backups left to tackle. Everything else should work with the script.

[Asheesh Laroia (paulproteus)]

On Mon, 27 Apr 2009, The Half Hollow Hacker wrote:

> Well, I am trying to write that "python script" As soon as I can get
> into the MediaWiki API, everything should be alright. I only have
> MediaWiki and backups left to tackle. Everything else should work with
> the script.

You rule.

[The Half Hollow Hacker (halfhollowhackers)]

Okay. Status update: I only have to write the new skeleton UI and some code to add the mysql wiki db to the backup system. Other than that, everything is done. Hopefully everything will be ready to go by this weekend.

[Parker Phinney (gameguy43)]

what is the status on this? what are the next action tasks?

[Parker Phinney (gameguy43)]

i'm especially interested in the status of moving blogs to wpmu.freeculture.org ... it looks like they've all been done, but we should probably run whatever shell script we had used one more time, to pull in any recent changes. here're some notes that i have from a couple months ago:

<tonyb486> [tonyb@tonybox fcdumps]$ scp fcdumps.tar.bz2 <email address hidden>:/home/freecult

to get a cookie:
wget --save-cookies cookie2 --post-data="log=admin&pwd=[INSERTPASSWDHERE]&rememberme=forever" http://freeculture.org/lyceum/login

<tonyb486> pyrak: 16:53 < tonyb486> [tonyb@tonybox fcdumps]$ for i in `cat blags`; do wget --load-cookies cookie2 http://$i.freeculture.org/admin/export.php?download=true -O $i.xml; sleep 3; done
<tonyb486> then just to check them tail -n1 *.xml and see if each one ends in </rss> or not

[The Half Hollow Hacker (halfhollowhackers)]

Almost done, the only thing left to fix is the mediawiki settings file. After that, it'll continue to be improved.