Kanpyo is Japanese morphological analyzer written in Rust inspired by ikawaha/Kagome.
This is a work in progress. I would break the API without notice.
The easiest way to install kanpyo is with the embedded dictionary. No additional setup required.
cargo install kanpyo --features mecab-ipadicor from git:
cargo install --git https://github.com/togatoga/kanpyo kanpyo --features mecab-ipadicThe dictionary will be automatically downloaded from GitHub Releases during the build process and embedded into the binary.
If you prefer a smaller binary size or want to use a custom dictionary:
cargo install kanpyoYou need to build and install a dictionary manually:
cd kanpyo-dict
tar xvf resource/mecab-ipadic-2.7.0-20070801.tar.gz -C resource
cargo run --release --bin ipa-dict-builder -- --dict resource/mecab-ipadic-2.7.0-20070801The dictionary is installed in the following directory:
- Linux:
$HOME/.config/kanpyo/ - macOS:
$HOME/Library/Application Support/kanpyo/ - Windows:
%APPDATA%\kanpyo\
You're ready to use kanpyo!
kanpyo --help
Japanese Morphological Analyzer
Usage: kanpyo [COMMAND]
Commands:
tokenize Tokenize input text
graphviz Output lattice in Graphviz format
help Print this message or the help of the given subcommand(s)
Options:
-h, --help Print help
-V, --version Print versionkanpyo tokenize "すもももももももものうち"
すもも 名詞,一般,*,*,*,*,すもも,スモモ,スモモ
も 助詞,係助詞,*,*,*,*,も,モ,モ
もも 名詞,一般,*,*,*,*,もも,モモ,モモ
も 助詞,係助詞,*,*,*,*,も,モ,モ
もも 名詞,一般,*,*,*,*,もも,モモ,モモ
の 助詞,連体化,*,*,*,*,の,ノ,ノ
うち 名詞,非自立,副詞可能,*,*,*,うち,ウチ,ウチ
EOSkanpyo
自然言語処理
自然 名詞,形容動詞語幹,*,*,*,*,自然,シゼン,シゼン
言語 名詞,一般,*,*,*,*,言語,ゲンゴ,ゲンゴ
処理 名詞,サ変接続,*,*,*,*,処理,ショリ,ショリ
EOS
形態素解析
形態素 名詞,一般,*,*,*,*,形態素,ケイタイソ,ケイタイソ
解析 名詞,サ変接続,*,*,*,*,解析,カイセキ,カイセキ
EOSecho "自然言語処理" | kanpyo
自然 名詞,形容動詞語幹,*,*,*,*,自然,シゼン,シゼン
言語 名詞,一般,*,*,*,*,言語,ゲンゴ,ゲンゴ
処理 名詞,サ変接続,*,*,*,*,処理,ショリ,ショリ
EOSPrint lattice in Graphviz format for debugging.
kanpyo graphviz "自然言語処理" | dot -Tpng -o lattice.png- Support various dictionaries(Sudachi, UniDic, neologd, etc.)
- Support server mode
- Support search mode
- Tests for load dictionary and tokenize