campfire_export

$ gem install stackbuilders-campfire_export
$ campfire_export

This gem is a fork from the original version of this project by Marc Hedlund. While almost all of the work is still his, this fork contains fixes and updates for user-facing issues, including the following:

• Upgrade gem dependencies and add support for Ruby versions >= 1.9.3
• Retry up to 5 times before giving up on a resource. Makes campfire_export usable for large exports.
• Fixes file size validation when file contains Unicode.
• Allow export of a specific room instead of exporting all rooms on a Campfire account.

When you install the gem following the instructions above (i.e., with the name stackbuilders-campfire-export) you will get these fixes, and the project will run on modern Ruby versions.

We will try to respond to issues and pull requests on the Stack Builders fork of this repository (https://github.com/stackbuilders/campfire_export), and to push new releases in a timely manner to https://rubygems.org/gems/stackbuilders-campfire_export.

I had an old, defunct Campfire account with five years' worth of transcripts in it, some of them hilarious, others just memorable. Unfortunately, Campfire doesn't currently have an export function; instead it provides pages of individual transcripts. I wanted a script to export everything from all five years, using the Campfire API.

I found a Gist that looked pretty good:

• https://gist.github.com/821553

but it wasn't quite right. So this is my modification, converted to a GitHub repo and a Ruby gem.

• Saves HTML, XML, and plaintext versions of chat transcripts.
• Exports uploaded files to a day-specific subdirectory for easy access.
• Reports and logs export errors so you know what you're missing.
• Obsessively confirms that everything was exported correctly.

Ruby 1.9.3 or later is required. RubyGems is also required -- I'd recommend having the latest version of RubyGems installed before starting.

Once you are set up, to install, run the following:

$ gem install campfire_export

There are a number of configuration variables required to run the export. The export script will prompt you for these; just run it and away you go. If you want to run the script repeatedly or want to control the start and end date of the export, you can create a .campfire_export.yml file in your home directory using this template:

# Your Campfire subdomain (for 'https://myco.campfirenow.com', use 'myco').
subdomain:  myco

# Your Campfire API token (see "My Info" on your Campfire site).
api_token:  your-campfire-token

# OPTIONAL: Export start date - the first transcript you want exported.
# Uncomment to set. Defaults to the date each room was created.
#start_date: 2010/1/1

# OPTIONAL: Export end date - the last transcript you want exported.
# Uncomment to set. Defaults to the date of the last comment in each room.
#end_date:   2010/12/31

# OPTIONAL: Campfire room - the room you want exported. All other rooms
# will be skipped.
# Uncomment to set. Defaults to all rooms.
#room_name:   Stack Builders Website Dev

The start_date and end_date variables are inclusive (that is, if your end date is Dec 31, 2010, a transcript for that date will be downloaded), and both are optional. If they are omitted, export will run from the date each Campfire room was created, until the date of the last message in that room.

Just run campfire_export and your transcripts will be exported into a campfire directory in the current directory, with subdirectories for each site/room/year/month/day. In those directories, any uploaded files will be saved with their original filenames, in a directory named for the upload ID (since transcripts often have the same filename uploaded multiple times, e.g. Picture 1.png). (Note that rooms and uploaded files may have odd filenames -- for instance, spaces in the file/directory names.) Errors that happen trying to export will be logged to campfire/export_errors.txt.

The Gist I forked had a plaintext transcript export, which I've kept in as transcript.txt in each directory. However, the original XML and HTML are now also saved as transcript.xml and transcript.html, which could be useful.

Days which have no messages posted will be ignored, so the resulting directory structure will be sparse (no messages == no directory).

First, thanks a ton to Jeffrey Hardy from 37signals, who helped me track down some bugs in my code as well as some confusion in what I was getting back from Campfire. His patient and determined help made it possible to get this working. Thanks, Jeff!

Also, thanks much for all the help, comments and contributions:

• Brad Greenlee
• Andre Arko
• Brian Donovan
• Andrew Wong
• Junya Ogura
• Chase Lee
• Alex Hofsteede
• Justin Leitgeb

As mentioned above, some of the work on this was done by other people. The Gist I forked had contributions from:

• Pat Allan
• Bruno Mattarollo
• bf4

Thanks, all!

• Marc Hedlund, marc@precipice.org