Tumblr Archiver

How does it work?

This application reads your Tumblr RSS feed (since it's consistent, independent of themes and relatively small) and extracts all posts, letting you save them as a zip archive on your computer. It can also save the pictures locally. Whether you want to delete, start fresh and keep the memories, or are worried about the possibility of Tumblr being shut down, this will help you protect your posts.

Limitations

• Sets - due to server limitations (speed and bandwidth), you can only download 300 posts at a time - which could take a while for some veterans
• Credit - it may be difficult to tell whether the post was a reblog, or where it came from, depending on the format
• Questions - Tumblr asks will be the question as the title and the response as the body, as though it's a text post (blame their RSS feed, I can't differentiate)
• Videos and Music - it's too much to ask my server to make copies of these - use the post editor to find these and copy them yourself (or the links) if you need them
• Notes, responses and reblogs - it's also too much to ask for my server to scan for responses in the notes - especially when notes aren't shown in the RSS feed

Can't I do this on Tumblr?

As of the time of making this on 24/01/2012 there is no known way to get a local copy of all of your posts - at least none that's any easier than this tool that I am aware of. However, it would not surprise me if they made this a feature a week or two after I go to the trouble of making this...

Check your settings on Tumblr

You should tweak the settings for your blog to ensure full compatibility

• Ensure your blog has an RSS Feed
• Do NOT truncate your blog's RSS feed
• Set the number of posts per page to 15 (highest possible)

Generate and download the archive

1. Enter your blog URL into the first textbox
2. Enter the part number into the second textbox - this refers to a clump of 20 pages (300 posts) that will be grouped in the archive.
3. Click "Get Archive" (expect to wait between 10 and 30 seconds)
4. Click "Download Here" (the download should be no more than a few seconds)

Read the archive

Simply open the html file (called "Part x [xxxxxx-xxxx]") - it should open automatically in your web browser.

Downloading Images (Optional)

The archives consist of HTML code - text, links and formatting. No images will show up as you scroll through, though the links for image posts will take you to the original on Tumblr's server. To fill in the images, you need to carefully complete this list:

1. Extract the archive (a zip file)
2. Download and install Python 2.7.2
   • Download for Windows XP, Vista, 7
   • Mac OS X 10.6 and 10.7
   • If you run something else, you probably know what you're doing
3. Open the file called download_images.py and wait for it to download all images

No Limits!

Because this software was difficult to write, and would use up a LOT of server resources (bandwidth + space + processing power) that have to be paid for, I might remove the limitations of doing one page at a time for a $5 day-pass. Details will be ironed-out in the future, but I'm curious if there'll be any demand for it. Basically you would pay $5 through PayPal and then get a code that you enter in an optional textbox in the application to unlock automatic downloads.

I'm well aware that most Tumblr users may not be in a position where they're even capable of buying things online, and the $5 fee might be asking a bit much of some people. I've always tried to make my work freely and publicly available but in this case it's not just the initial cost of developing the application that has to be covered (usually by ads or the fact that this is my hobby) - but the server costs and maintenance as well will sky-rocket.

I'm sure that the current version should be enough for most tasks, but I can see the obvious hurdle to people wanting to use it. Please email me at billymoffat92@gmail.com if you're interested or have any ideas.