Go to file
2018-08-23 16:10:16 +02:00
anki-bundled@b5a8d3f47c Update anki-bundled to 2.1.1 2018-08-20 15:39:59 +02:00
ankisyncd Remove debug print 2018-08-22 14:22:54 +02:00
tests Use mapping protocol access instead of legacy configparser API 2017-11-06 21:06:18 +01:00
.gitignore Add gitignore 2017-10-28 23:18:18 +02:00
.gitmodules Apply @ndl's patches 2016-11-03 22:32:19 +07:00
ankisyncctl.py Use cwd in ankisyncctl.py for now 2018-08-20 04:05:18 +02:00
ankisyncd.conf Make built-in server actually use the "host" config field 2017-11-06 17:36:00 +01:00
COPYING Renamed LICENSE.txt to COPYING 2013-10-13 16:10:00 -05:00
README.md Add information about pyaudio 2018-08-23 16:10:16 +02:00

ankisyncd

Anki is a powerful open source flashcard application, which helps you quickly and easily memorize facts over the long term utilizing a spaced repetition algorithm. Anki's main form is a desktop application (for Windows, Linux and macOS) which can sync to a web version (AnkiWeb) and mobile versions for Android and iOS.

This is a personal Anki server, which you can sync against instead of AnkiWeb. It was originally developed to support the flashcard functionality on Bibliobird, a web application for language learning.

This version is a fork of jdoe0/ankisyncd. It supports Python 3 and Anki 2.1.

Installing

  1. Install Anki. The server currently works well with 2.1.1. Running the server with other versions might work, but things might break, so do it at your own risk. Anki 2.0.x definitely isn't going to work, as it's written in Python 2, whereas ankisyncd runs only on Python 3.5 and later. If for some reason you can't get the supported Anki version easily on your system, you can use anki-bundled from this repo:

     $ git submodule update --init
     $ cd anki-bundled
     $ pip install -r requirements.txt
    

    Keep in mind pyaudio, a dependency of Anki, requires Python 3 and PortAudio headers to be present before running pip. ankisyncd doesn't use the audio recording feature of Anki, so if you don't want to install PortAudio, you can edit anki-bundled/anki/sound.py and anki-bundled/requirements.txt to exclude pyaudio:

    ed version:

     $ echo '/# Packaged commands/,$d;w' | tr ';' '\n' | ed anki/sound.py
     $ echo '/^pyaudio/d;w' | tr ';' '\n' | ed requirements.txt
    

    sed -i version:

     $ sed -i '/# Packaged commands/,$d' anki/sound.py
     $ sed -i '/^pyaudio/d' requirements.txt
    

    Manual version: remove every line past "# Packaged commands" in anki/sound.py, remove every line starting with "pyaudio" in requirements.txt

  2. Install the dependencies:

     $ pip install webob
    
  3. Modify ankisyncd.conf according to your needs

  4. Create user:

     $ ./ankisyncctl.py adduser <username>
    
  5. Run ankisyncd:

     $ python -m ankisyncd ankisyncd.conf
    

Setting up Anki

Anki 2.1

Create a new directory in ~/Anki/addons21 (name it something like ankisyncd), create a file named __init__.py containing the code below and put it in ~/Anki/addons21/ankisyncd.

import anki.sync

anki.sync.SYNC_BASE = 'http://127.0.0.1:27701/%s'

Anki 2.0

To make Anki use ankisyncd as its sync server, create a file (name it something like ankisyncd.py) containing the code below and put it in ~/Anki/addons.

import anki.sync

anki.sync.SYNC_BASE = 'http://127.0.0.1:27701/'
anki.sync.SYNC_MEDIA_BASE = 'http://127.0.0.1:27701/msync/'

Replace 127.0.0.1 with the IP address or the domain name of your server if ankisyncd is not running on the same machine as Anki.