I’ve managed to get a text-file output of my old Armedliberal.com site, and am trying to get it to import into the MT 3.3 instance we’re running here.
When I try, I get a ‘file too big’ error.
When I break the file up into little files and try again – same thing.
Any ideas, anyone?
When I had a similar problem (just the HTML files, not the database), I wrote a small perl script to take the entries, break them into their component parts, and then add them back to the database via the API. The only shortcoming is that I couldn’t save the comments and trackbacks. You might try a similar approach.
Jeff’s on the right track although for a one-shot it would probably be a little easier to generate a SQL script that would insert the necessary entries into the databsae.
No luck getting a database dump, eh?
Nope. Does. Not. Exist.
Dave, there are Perl libraries for loading the entries into the database. It took me about 20 minutes to write. That said, a SQL script would work just fine; it just requires more knowledge of the MT internals.
I just looked for the script, and couldn’t find it (it was so fast to do, I likely didn’t keep it), or I would post it.
I was on vacation — did you ever get this fixed? If not, you might check the directory. As I understand the process, MT will import every file in that directory, so if you’ve left a big file there, you’d get the error even if you broke it up in to other little files. Otherwise, drop me e-mail off line.