tris/recoll
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Korean splitter script: use python-mecab-ko if possible, else konlpy

Browse Source
This commit is contained in:
Jean-Francois Dockes 2020-04-10 14:27:06 +02:00
parent 1bd5babcb8
commit b63cc1b712
1 changed files with 20 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
src/filters/kosplitter.py
View File

@ -28,7 +28,15 @@
import sys import sys
import cmdtalk import cmdtalk
from konlpy.tag import Okt,Mecab,Komoran # We can either use konlpy, which supports different analysers, or use
# the python-mecab-ko, a direct interface to mecab, with the same
# interface as konlpy https://pypi.org/project/python-mecab-ko/
try:
import mecab
usingkonlpy = False
except:
from konlpy.tag import Okt,Mecab,Komoran
usingkonlpy = True
class Processor(object): class Processor(object):
def __init__(self, proto): def __init__(self, proto):
@ -38,13 +46,21 @@ class Processor(object):
self.tagsKomoran = False self.tagsKomoran = False
def _init_tagger(self, taggername): def _init_tagger(self, taggername):
global usingkonlpy
if not usingkonlpy and taggername != "Mecab":
from konlpy.tag import Okt,Mecab,Komoran
usingkonlpy = True
if taggername == "Okt": if taggername == "Okt":
self.tagger = Okt() self.tagger = Okt()
self.tagsOkt = True self.tagsOkt = True
elif taggername == "Mecab": elif taggername == "Mecab":
# Use Mecab(dicpath="c:/some/path/mecab-ko-dic") for a if usingkonlpy:
# non-default location # Use Mecab(dicpath="c:/some/path/mecab-ko-dic") for a
self.tagger = Mecab() # non-default location. (?? mecab uses rcfile and dicdir not
# dicpath)
self.tagger = Mecab()
else:
self.tagger = mecab.MeCab()
self.tagsMecab = True self.tagsMecab = True
elif taggername == "Komoran": elif taggername == "Komoran":
self.tagger = Komoran() self.tagger = Komoran()