Worth writing an article on how to clone WP sites properly?

[davecoveney]

Cloning or moving a WP site properly, without a loss of widgets, options from plugins and so on, is a surprisingly tricky job to get right.

I wrote tools such as a mySQL search and replace utility for serialised PHP to help with this, but I figure that it could well be worth writing a step by step guide on how to use this tool plus others to help move a WordPress site from one location to another.

Do you think there'd be much interest in this? Obviously it takes time to document the process in a clear way, but I wondered whether it was a problem many people experienced?

[JohnM]

+10 ;)

http://johnmyr.wordpress.com/2009/08...a-test-plugin/

In June Peter Westwood released a WordPress Beta Tester plugin. The plugin enables you to upgrade your blog to the latest Beta or Release candidate at the click of a button using the built in upgrader.
If we look at the stats for the plugin well see that its not very much used. Its downloaded 230 times in two months, and one might think that this isnt a good idea after all.

However, in my opinion a WordPress Beta Tester plugin is a great idea. It just needs more features to make it attractive for normal people to use. The hardcore WP devs dont need it.

Beta testing needs to be done in an environment as close to real installs as possible, with the same plugins, hacks and content as live sites. To achieve this the plugin should include features and documentation to help and motivate WordPress users to run a cloned version of their live site on localhost or a subdomain like beta.mydomain.tld.

I asked if it was appropiate for dev chat and it sneaked in without me actually knowing. Anyways heres the discussion from dev chat:

https://irclogs.wordpress.org/chanlog.php?channel=wordpress-dev&day=2009-08-20#m24289

A step by step tutorial would be very nice.

A "Ultimate guide to cloning WordPress sites" and a follow up with "Betatest 2.9 on your cloned site" should make some buzz and would probably be a popular article around the WordPress blogosphere bringing backlinks and good karma ;)

[Laura]

I'd be really interested in that, Dave. For me it'd be useful for moving sites from a local dev environment to online.

I reckon that kind of thing would make a great talk at WordCamp UK ;)

[JohnM]

but I wondered whether it was a problem many people experienced?

According to Matt there were 10.000 betas of WP2.8 running before it was released. Still obvious bugs related to Widget UI where not detected. I guess this is because people doesnt bother to run full content and setup with widgets/plugins when running nightlys and betas.

A clone tutorial would help the community making beta testing better, and WP better.

Obviously it will also help hundreds or thousand of people moving WP from dev server to live sites, and people having live sites which they want to test changes on.

[Otto]

According to Matt there were 10.000 betas of WP2.8 running before it was released. Still obvious bugs related to Widget UI where not detected.

Could you be more specific here? I was unaware that there were any widget bugs.

The only problems I know regarding widgets were people having issues with the javascript compression and concatenation features in 2.8. People with those issues saw problems with widgets (and virtually everything else) because their browsers were not loading any javascript properly. But I'd not call these bugs in the widgets UI itself.

[JohnM]

I agree Otto. Thats not a precise statement. Anyways, if more people ran cloned sites with WP betas, problems may be detected earlier ( if its a bug or some bad plugin code or other problem )

[andrea_r]

It comes up often enough that if you wrote a good article on it, I'd link to it regularly. :)

[Ryan]

Why not write a plugin which automatically ports things over?

Presumably most things like transferring the database content and files over should be fairly basic and already documented elsewhere, the complicated bit left would be the MySQL queries to change all the URL's to the new ones and a plugin to automate the process would be terrific IMO.

[davecoveney]

Ryan - I've already done the hard work of search and replace to get the URLs right.

Thing is, the move of the database is the easy part, as is dealing with all the files. What then needs to happen is to run my script against the fresh DB install to change all the URLs safely. You couldn't have that script as a plugin because you wouldn't be able to get to it until it had done its work - chicken and egg problem there.

So... I would need to do something like write a database dumping utility that does the URL editing there and then, so giving you a DB you can use elsewhere. But ultimately you still have a lot of steps to get through that need to be documented carefully.

There's also the fact that I could write the guide far more quickly than a polished plugin, it would bring much more link love than a plugin ever gets, and it would be more flexible.

I think for the moment I'll stick with the guide, but if someone used my code or principles to help write a plugin then I wouldn't object. And who knows, I may do it myself one day! But really, once download/upload time is taken into account my process is very quick anyway.

[davecoveney]

@Laura - good to see you here! And yes, definitely a good one for WordCamp - may flag that one up for next year!

@Andrea - links are always nice. Yay :-)