Question

The code is:

import MeCab

m = MeCab.Tagger("-O wakati")
text = raw_input("Enter Japanese here: ")
print m.parse(text)

The problem is that after entering the string into the raw_input it gives an error in IDLE:

Traceback (most recent call last):
  File "C:\Users\---\Desktop\---\Python\japanesetest.py", line 5, in <module>
    print m.parse(text)
  File "C:\Users\---\Desktop\---\Python\lib\site-packages\MeCab.py", line 220...
    def parse(self, *args): return _MeCab.Tagger_parse(self, *args)
TypeError: in method 'Tagger_parse', argument 2 of type 'char const *'

If I do this however:

import MeCab

m = MeCab.Tagger("-O wakati")
print m.parse('なるほど、マルコフ辞書のキーはタプルにしたほうがスッキリしますね。')

I get the proper result:

なるほど 、 マルコフ 辞書 の キー は タプル に し た ほう が スッキリ し ます ね 。

Things I have tried are unicode tags at the beginning, writing to a textfile in unicode and parsing the text, and a few other million things. I’m running Python 2.7 and MeCab 0.98. If this can’t be answer, even a little light shed on the error would be appreciated.

Answer 1

I am able to run your snippet successfully using Python 2.7 and MeCab 0.98 in both IDLE and IPython command line.

import MeCab
m = MeCab.Tagger("-O wakati")
text = raw_input("Enter Japanese here: ")
Enter Japanese here: 私の車はとても高いです。
print m.parse(text)
私 の 車 は とても 高い です 。 

However, when reading from a UTF file I will get errors when trying to parse the text. For those cases I explicitly encode the text to shift-jis. You might try this technique. Below is an example.

rawtext = open("UTF.file", "rb").read()
tagger = MeCab.Tagger()
encoded_text = rawtext.encode('shift-jis', errors='ignore')
print tagger.parse(encoded_text).decode('shift-jis', errors='ignore')