Installing and Using Cjklib

Contents

• Introduction
• Dependencies
• Porting to Python 3
• Installing
  • Windows
  • Unix
• Documentation & Usage
  • Examples
• Database
  • SQLite
  • MySQL
• Contact

Introduction

Cjklib provides language routines related to Han characters (characters based on Chinese characters named Hanzi, Kanji, Hanja and chu Han respectively) used in writing of the Chinese, the Japanese, infrequently the Korean and formerly the Vietnamese language(s). Functionality is included for character pronunciations, radicals, glyph components, stroke decomposition and variant information.

Dependencies

• Python 2.4 or above (currently no support for Python3)
• SQLite 3+
• SQLAlchemy 0.5+
• pysqlite2 (already ships with Python 2.5 and above)

Alternatively for MySQL as backend:

• MySQL 5+
• MySQL-Python

Porting to Python 3

conda create -n py37 python=3.7
conda activate py37
git clone <git-repo-url>
cd cjklib3
pip install 2to3
2to3 -w .
curl http://ftp.unicode.org/Public/UNIDATA/Unihan.zip -o Unihan.zip
python -m cjklib.build.cli build cjklibData --attach= --database=sqlite:///cjklib/cjklib.db
pip install .
installcjkdict --download CEDICT
python -m cjklib.build.cli build fullCEDICT --attach=sqlite:///cjklib/cjklib.db --database=sqlite:///cjklib/cjklib.db
pip install .

Installing

Windows

Install cjklib using the provided .exe installer. Make sure above dependencies are satisfied.

Three scripts cjknife.exe, buildcjkdb.exe, and installcjkdict.exe will be added to the Python Scripts sub-directory. Make sure this directory is included in your PATH environment variable to access these programs from the command line.

CJK dictionaries are not included by default. If you want to install any of those run the following (with an Internet connection) from the root directory of the source package:

$ installcjkdict CEDICT

This will download CEDICT, create a SQLite database file and install it under the directory given by the APPDATA environment variable, e.g. C:\windows\profiles\MY_USER\Application Data\cjklib. Just substitute CEDICT for any other supported dictionary (i.e. EDICT, CEDICT, HanDeDict, CFDICT, CEDICTGR).

Unix

If you are installing from the source package you need to deploy the library on your system:

$ sudo python setup.py install

Also make sure above dependencies are satisfied. CJK dictionaries are not included by default. If you want to install any of those run the following (with an Internet connection):

$ sudo installcjkdict CEDICT

This will download CEDICT, create a SQLite database file and install it to /usr/local/share/cjklib. Just substitute CEDICT for any other supported dictionary (i.e. EDICT, CEDICT, HanDeDict, CFDICT, CEDICTGR).

Documentation & Usage

Documentation is available online. Also see the project page and its wiki. There is a small command line tool cjknife that offers some of the library's functions. See cjknife --help for an overview.

Examples

• Get stroke order of characters:

  >>> from cjklib import characterlookup
  >>> cjk = characterlookup.CharacterLookup('C')
  >>> cjk.getStrokeOrder(u'说')
  [u'㇔', u'㇊', u'㇔', u'㇒', u'㇑', u'㇕', u'㇐', u'㇓', u'㇟']
  

• Access a dictionary (here using Jim Breen's EDICT):

  >>> from cjklib.dictionary import EDICT
  >>> d = EDICT()
  >>> d.getForTranslation('Tokyo')
  [EntryTuple(Headword=u'東京', Reading=u'とうきょう',
  Translation=u'/(n) Tokyo (current capital of Japan)/(P)/')]
  

Database

Packaged versions of the library will ship with a pre-built SQLite database file. You can however easily rebuild the database yourself.

First download the newest Unihan file:

$ wget ftp://ftp.unicode.org/Public/UNIDATA/Unihan.zip

Then start the build process:

$ sudo buildcjkdb -r build cjklibData

SQLite

SQLite by default has no Unicode support for string operations. Optionally the ICU library can be compiled in for handling alphabetic non-ASCII characters. Cjklib can register own Unicode functions if ICU support is missing. Queries with LIKE will then use function lower(). This compatibility mode has negative impact on performance and as it is not needed for dictionaries like EDICT or CEDICT it is disabled by default. See cjklib.conf for enabling.

MySQL

With MySQL 5 the following CREATE command creates a database with utf8 as character set using the general Unicode collation (MySQL from 5.5.3 on will support full Unicode given character set utf8mb4 and collation utf8mb4_bin):

CREATE DATABASE cjklib DEFAULT CHARACTER SET utf8 COLLATE utf8_bin;

You might need to set access rights, too (substitute user_name and host_name):

GRANT ALL ON cjklib.* TO 'user_name'@'host_name';

Now update the settings in cjklib.conf.

MySQL < 5.5 doesn't support full UTF-8, and uses a version with max 3 bytes, so characters outside the Basic Multilingual Plane (BMP) can't be encoded. Building the Unihan database thus might result in warnings, characters above U+FFFF can't be built at all. You need to disable building the full character range by setting wideBuild to False in cjklib.conf before building. Alternatively pass --wideBuild=False to buildcjkdb.

Contact

For help or discussions on cjklib, join [email protected].

Please report bugs to the project's bug tracker.