Being an avid reader and user of Anki, I've historically struggled to come up with a way to take text I want to remember from books and turn them into flashcards. With physical books, I had exactly one option: type out the text on my computer. This didn't sit well with me at all, so several years ago I became a devout user of e-books.
E-books give me the ability to copy and paste text directly on my computer, transport lots of books all at once and are generally excellent for anyone who cares about maintaining and organizing a large library. Unfortunately, most e-readers have only a very basic ability to highlight text and store it on the device. It wasn't until I bought a Kobo e-reader that I found my solution. These devices hold data (including highlights) on a SQLite database that is easily accessible.
Since buying my Kobo, I've been using a simple SQLite extension in Firefox, but it came up short in some key areas (namely in my ability to quickly and easily remove highlights I don't need anymore). My solution is this program (which may or may not transform into a set of programs). It looks at a Kobo database, collects all the highlights into a list and then presents them to the user, along with a choice: do you want to keep this highlight?
If you don't, you type n and it will switch the Hidden field in the database row to true - meaning it will no longer show up on the reader, but it will still be there for future reference. And, because of the query used to gather the data, it won't show up again in the execution of this program.
All you need to do to run the program is the place KoboReader.sqlite file that can be found on a Kobo reader within the .kobo directory into the same directory as the script. You can run the file within any arguments, which will present highlights from first to last. Or, you can add reverse or random to have last-first ordering or random ordering, respectively.
If you want to skip the hassle of manually copy-pasting the Kobo DB file, you can now use the get and push commands with this script. get will pull the DB from the device and place it in the directory you're running the script. push will send the KoboReader.sqlite file you have in your current directory to the device. backup will create a compressed zip file backup of your DB. This is only for MacOS users at the moment, although I may make it cross-platform later.
If you want to change which file you manipulate with the highlights command, change the KOBO_DB environment variable (default is KoboReader.sqlite).
If you want to change where the program looks for your Kobo reader's DB file once it's plugged in via USB, change the KOBO_READER environment variable (defaults to /Volumes/KOBOeReader/.kobo/KoboReader.sqlite, which is where it appears in Mac OS X).
Sample usage (from the command line):
~$ ./kobo.py highlights
~$ ./kobo.py highlights random
~$ ./kobo.py highlights reverse
~$ ./kobo.py get
~$ ./kobo.py push
~$ ./kobo.py backup
~$ ./kobo.py help
Output looks like this:
file:///mnt/sd/<book_title>.kepub.epub
This is a block of sample text. It is being used in place of real text from a book.
Do you want to keep this bookmark (y/n/stop)?
This process will continue until you type stop and press Enter. Once you do that, changes will be committed to the DB and the connection to the DB will be closed. Any changes made before typing stop will not be saved.