Package cjklib

source code

Han character library. 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.

Supported

The following functions are supported by this library:

• Character readings (pronunciation):
  • Pinyin, Gwoyeu Romatzyh, Wade-Giles, IPA, Braille (all Mandarin)
  • Jyutping, Cantonese Yale (both Cantonese)
  • Hangul (Korean)
  • Hiragana, Katakana (Japanese)
• Conversion of readings from one into another
• Mapping between character and reading
• Mapping between character and Kangxi radical (radical mapping as defined by the Unihan database)
• Mapping between character and multiple radical forms
• Mapping between character and stroke sequence
• Mapping between character and stroke count
• Mapping between Kangxi radical forms, radical variant forms and equivalent characters
• Character variant lookup (including mapping from traditional to Chinese simplified forms)

Tools

The following tools come with this library:

• cjknife, provides most functions from the library on the command line.
• buildcjkdb, builds the database from source files.

Data

The library comes with its own set of sources on:

• Pinyin syllables
• Gwoyeu Romatzyh syllables including rhotacised forms and abbreviations
• Wade-Giles syllables
• Jyutping syllables
• Cantonese Yale syllables
• Pinyin to Gwoyeu Romatzyh mapping
• Wade-Giles to Pinyin mapping
• Pinyin to IPA mapping
• Pinyin to Braille mapping
• Jyutping to Cantonese Yale mapping
• Jyutping to IPA mapping
• mapping of Mandarin syllables to onset and rhyme
• mapping of Cantonese syllables to onset and rhyme
• Kangxi radical forms
• stroke count and stroke order
• stroke names
• character decomposition

See the data files for comparison with other sources.

This project makes use of the Unicode Han database provided by the Unicode Consortium: Unicode Standard Annex #38 - Unicode Han database (Unihan): http://www.unicode.org/reports/tr38/tr38-5.html, 28.03.2008.

The following data is used:

• Character Kangxi radical information (from kRSKangxi)
• Radical residual stroke count (from kRSKangxi) and total stroke count (from KTotalStrokes)
• Mandarin character readings in Pinyin (from kMandarin, kHanyuPinlu, kXHC1983)
• Cantonese character readings in Jyutping (from kCantonese)
• Korean character readings in Hangul (from kHangul)
• Character variant forms (from kCompatibilityVariant, kSemanticVariant, kSimplifiedVariant, kSpecializedSemanticVariant, kTraditionalVariant, kZVariant)

This includes dictionary data from:

• kXHC1983: Xiàndài Hànyǔ Cídiǎn (现代汉语词典). Shāngwù Yìnshūguǎn, Beijing,
• kHanyuPinlu: Xiàndài Hànyǔ Pínlǜ Cídiǎn (現代漢語頻率詞典). 北京語言學院語言教學研究所編著, First edition 1986/6, 2nd printing 1990/4, ISBN 7-5619-0094-5.

Currently no data validation scheme is implemented as this library is still in early development. Rather than specifying few data cjklib tries to support as much options as possible. The library tries to be as accurate as possible but mistakes do happen, especially for data which differs on different locales.

Dependencies

cjklib is written in Python and is well tested on Python 2.5. Apart from this dependency it needs a database back-end for most of its parts and library SQLAlchemy. Currently tested are:

• SQLite, tested on SQLite 3
• MySQL, tested on MySQL 5.0 (works only with characters from the Basic Multilingual Plane in Unicode, BMP)