In case anyone finds this via a search, I want to conclude this...so if anyone is in the same situation as me, they'll see a way out. And/or to offer up to everyone a poor-man's way of backing up one's site...not in the most useful way, but in a way easy to effect and one that one might as well have as a backup to more malleable backups.
So, essentially, my "issue" was that my host's server somehow lost the Posts table of my WordPress database (but the Comments table was somehow safe). They had no backups of my site/database and the last backup I had was from 11/2008. The question is: is there any way to recover my lost year of Posts and, if so, re-enter them in a way that will allow me to also re-integrate my Comments for that lost year of posts.
There was no hope for me to recover my posts via stuff like cached Google searches or the Wayback Machine because I have always used robots.txt to block spiders.
However, I found out that Google Reader will archive all entries in an RSS feed...essentially forever. Regardless of whether or not the user has read the entries or not. At any point, the user can click on an "All Entries" view and go back to every single post that was ever scraped by Google Reader, dating back to when the subscription was established.
Luckily for me, I subscribed to my own feed on a date preceding my last backup *and* I had set my RSS feed to display full entries. So I can recover all my lost text/HTML...which is all I need, since my host's server never lost any of my other files.
The question for me is now: how to do it in an automated way.
- Obviously, I can load up all my lost entries in Google Reader and then save that page as a flat HTML file.
- Then I can just copy-and-paste all the entries, one-by-one.
- And, if I add them in the right order (so that each post retains its original Post ID #), I imagine I can copy my complete Comments table over onto my reconstructed database.
I don't know my ass from a MySQL query, so I'll need to find a programmer to help me with this, but I'm sure a script could be written to break up the Google Reader flat file into separate posts and "inject" them into the database (or into an importable file) in a way so that they're bearing the correct Post ID #s. Then just copy the complete Comments table over upon the resulting database.
Hopefully I can find a freelancer who'll do it for a reasonable price, as I imagine that someone with the right knowledge base could take care of this situation pretty quickly.