Introduction

This is a short guide how to install Eclectus and additional data.

From package

Download the appropriate package from the download section. And follow the respective paragraphs below. For Ubuntu you can directly add the PPA to your sources.list.

Ubuntu

Add the following two lines to your /etc/apt/sources.list (for Ubuntu Karmic):

deb http://ppa.launchpad.net/cburgmer/ppa/ubuntu karmic main deb-src http://ppa.launchpad.net/cburgmer/ppa/ubuntu karmic main

Make sure you also have the Ubuntu Universe added to your sources and run:

$ sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 2B553E3C $ sudo aptitude update $ sudo aptitude install eclectus

If you don't want to pull in 160 MB of data files, open aptitude and de-select the packages manually.

Debian

Download the .deb for cjklib, too, and run:

$ su $ dpkg -i eclectus_0.2beta~unstable-1_all.deb python-cjklib_0.2~unstable-1_all.deb

Additional data files are currently only available as source packages.

Source package

To install from the source package, un-tar the contents, change to the directory and install it using superuser privileges. But make sure you download and install cjklib first.

$ tar -xzf eclectus-0.2beta.tar.gz $ cd eclectus-0.2beta $ sudo python setup.py install

For data files (suggested are bw.png.segment-0.2beta.tar.gz and cmn-caen-tan-0.2beta.tar.gz) run:

$ tar -xzf bw.png.segment-0.2beta.tar.gz $ cd bw.png.segment-0.2beta $ sudo make install

From SVN/GIT

To install from the Subversion repository or GIT repository at Github follow the steps below.

Eclectus

Before installing Eclectus make sure the dependencies are covered. Especially note that cjklib is also needed.

Do a $ svn checkout http://eclectus.googlecode.com/svn/branches/eclectus-0.2beta eclectus-read-only

(alternatively $ git clone git://github.com/cburgmer/eclectus.git )

to get the latest sources and change into the resulting directory.

Generate the database and install the program:

$ python libeclectus/buildtables.py build base \ --database=sqlite:///libeclectus/libeclectus.db $ python setup.py install

Data

The Kanji Stroke Order font just needs to be available to Qt for Eclectus to find it, while the others need to be downloaded and built. For a local installation the data should be placed under libeclectus/files/ for Eclectus to find it. The Makefile in libeclectus/files/ takes care of downloading and building the data. The final install step can be started from the generated build directories.

Stroke order images

The Wikimedia Commons Stroke Order Project offers three different image types: * -bw.png http://upload.wikimedia.org/wikipedia/commons/6/6e/中-bw.png' height='60' /> * -red.png http://upload.wikimedia.org/wikipedia/commons/a/af/中-red.png' height='60' /> * -order.gif http://upload.wikimedia.org/wikipedia/commons/8/8a/中-order.gif' height='60' />

Eclectus currently uses two stroke order sources, one for the upper right corner, next to the headword, one in its own section.

The stroke order displayed in the section is installed by $ make bw.png.segment

If you decide not to use the Kanji Storke Order font for the upper right corner, you may want to install $ make order.gif or $ make red.png which both currently have only a small repetoire though.

Finally you need to install the files:

$ cd build/THE_SOURCE_JUST_BUILT $ make install

Eclectus will automatically find the sources if placed correctly. If you have more than once choice installed, you might need to edit the eclectusrc file under .kde/share/config/ and set the option strokeOrderType to the preferred stroke order source for the image shown in the upper right corner, e.g. order.gif or kanjistrokeorderfont.

Chinese audio

The SWAC collection has two sets for Mandarin Chinese audio of single words. To install a collection of approx. 8500 words from a female speaker, run $ make cmn-caen-tan $ cd build/cmn-caen-tan $ make install which will download the audio files from http://download.shtooka.net/cmn-caen-tan_ogg.tar and run the index generation process. Replace cmn-caen-tan by chi-balm-hsk1 for a male voice of approx. 1000 samples.