Problem with CFDICT

[dor...@gmail.com]

Hi,

I use cjklib quite frequently and it's a very useful tool for me.

I used until now CEDICT and didn't get any problem but recently I made some tests with CFDICT and observed a problem:

the word 哪儿 gives the following error here below (I tested with a few words and it worked, this is the first one with that problem):

File "C:\Python27\lib\site-packages\cjklib-0.3.2-py2.7.egg\cjklib\dictionary\format.py", line 151, in format

    for idx, entity in enumerate(reading.split(' ')):

AttributeError: 'NoneType' object has no attribute 'split'

.

Here is the code to reproduce it:

from cjklib.dictionary import *

d = CFDICT()

result = d.getFor(u'哪儿')

Any idea of the cause and do you have any temporary workaround ?

      Thank you,

           Xavier

[Christoph Burgmer]

Hi Xavier,

a bit late but, is it possible that the entry is just missing a Pinyin entry? Or maybe an esoteric one (read "invalid")?

-Christoph

[Xavier Baele]

Hi,

the pinyin is "na3 r" in the french dictionary while in the english one it's "na3 r5", could it be the problem?

        THX,

          Xavier

[Christoph Burgmer]

Sounds reasonable.

I checked the settings for CFDICT and they share the same ones with CEDICT. However there seems to be a difference here. If this entry you mention is the only one that doesn't have the fifth tone marked with a number 5, then that should probably be corrected with CFDICT. If that is the general format of CFDCT that should probably be fixed then with cjklib.

-Christoph

[Xavier Baele]

Indeed.

I will check that and tell you.

      Xavier