At work we’re preparing to move data from one WordPress install to another, as part of an overall “let’s consolidate all our blogs under one roof” move. Of course, it’s not as easy as it should be.
First off, let me say this: I truly believe that WordPress is one of the best examples of open source on the planet, if not simply for its breakout popularity. While I don’t always agree with how certain bits were done in certain ways, there is more than one way to skin a cat.
That being said, however, I believe, as it stands, the migration utilities available could use some work. We’re moving Blog X from a 3.0.1 install to another 3.0.1 install, but it’s not as simple as just hitting export in the WordPress admin. Why? The importer seems to freak out once the import file becomes to large. Even though we’ve gone through the settings for WordPress and the underlying system, it begins to freak around 800 posts in. So, as a solution, we’re linking up the databases and running queries directly on the backend to migrate the data. Fortunately, however, this is pretty darn easy…
Even more stink comes in when one notices that Blog X is currently a standalone blog… and we’re trying to put it into another WordPress 3.0.1 install… one that we’re using as an MU system, and Blog X is no longer going to be the ‘main event’. I realize this isn’t a common issue, but it certainly would be nice if this worked a bit better. So, like I said, it’s not that bad. But, as the importer will tell you, “Upload your WordPress eXtended RSS (WXR) file and we’ll import the posts, pages, comments, custom fields, categories, and tags into this site.” That’s just great. However, I’ve got users. 2000+ users that need to move as well. This is no longer easy. I have had to come up with a way (along with a colleague’s generous assistance) to not only migrate the data on the DB side for each user (not so bad, honestly) but also to generate the correct associated folders and files in the filesystem (for avatars and the like) which turned out to be a complete pain in the rump.
However, launch day is next Tuesday. Maybe we’ll get lucky and all of our planning (3 test runs later) will pay off and showing up at 2AM to do this thing will prove to be far too cautious for what was actually needed. I certainly hope so.
In the end, all I really want is a user import feature to be a part of WordPress’ core functionality. I realize plugins exist, but we all know they need to update along with WordPress to stay stable and accurate. Please hear my plea, WordPress Development Team! Add a way to import users to the core so we don’t face the headaches again!