MREP: Morpheme Regular Expression Printer

MREP is a regular expression matcher for morpheme sequences. You can find morpheme sub-sequences that match a given pattern, such as noun sequences.

Requirement

• Python >=2.7
• mecab-python ( https://github.com/SamuraiT/mecab-python3 )

Install

$ pip install mrep

If you do not have a dictionary for MeCab, install unidic-lite.

$ pip install unidic-lite

If you want to install it from its source, use setup.py.

$ python setup.py install

Usage

usage: mrep [-h] [-o] [--color {never,auto,always}] [-n] [--mecab-arg MECAB_ARG]
            PATTERN [FILE [FILE ...]]

positional arguments:

PATTERN:	pattern
FILE:	data file

optional arguments:

-h, --help	show this help message and exit
-o, --only-matching
	print only matching
--color COLOR	color mode. select from "never", "auto" and "always". (default: auto)
-n, --line-number
	Show line number
--mecab-arg MECAB_ARG
	argument to pass to mecab (ex: "-r /path/to/resource/file")

Pattern

.

matches all morphemes

<surface=XXX>

matches morphemes whose surface are XXX

<pos=XXX>

matches morphemes whose POS are XXX

<feature=XXX>

matches morphemes whose features are XXX

<feature=~XXX>

matches morphemes whose features maches a RegExp pattern XXX

X*

matches repetiion of a pattern X

X|Y

matches X or Y

(X)

matches X

Example

<pos=名詞>

matches a noun

<pos=名詞>*

matches repetition of nouns

<pos=名詞>*<pos=助詞>

matches repetition of nouns and a particle

(<pos=名詞>|<pos=動詞>)*

matches repetition of nouns or verbs

License

This program is distributed under the MIT license.

Copyright

(c) 2014, Yuya Unno.