Accurate Backup Script for Google Reader

A script to accurately backup all data from Google Reader. Note: Google Takeout might give you inaccurate data. So you'd better use this instead of that.

Data to be dumped

  • All subscriptions (feeds.json)
  • Starred items (stars.json)
  • All items (all other .json files)


(TLDR for hackers: Use pip to resolve dependencies. See requirements.txt file.)

  1. You need Python 2.7 first. (Mac includes a builtin Python.)
  2. Install pip. If you don't know how to install pip, type curl -O https://raw.github.com/pypa/pip/master/contrib/get-pip.py; sudo python get-pip.py and return, then type your password to give permission.
  3. Download the source code zip from here and extract it.
  4. Install requirements using pip: pip install --user -r requirements.txt.


(TLDR for hackers: try python backup.py -h.)

Type the following command in Terminal ($ is just a prompt):

$ python backup.py -u YOURNAME@GMAIL.COM BACKUP_DIR

It will ask you to type password as well. Type the password of your Google account. Note that your typings are never displayed:


It will download several *.json files into BACKUP_DIR you typed above.

Export backup data to Earth Reader repository

You can export all Google Reader data backup.py downloaded to Earth Reader repository using toearth.py:

$ toearth.py -o "Your Name" e "you@email.com" BACKUP_DIR REPOSITORY_DIR

The above command will create a new directory (REPOSITORY_DIR) which is an Earth Reader repository.


Public License except of toearth.py, is distributed under GPLv3+.