On Mar 21 at 18:12 +0000, Spencer Andrew wrote: > You mentioned it may be better breaking the boundaries by character instead of > word (I'm guessing the program sees this sentence as just one word then?). Do > you know how I can do that?
Thanks for the details. I was able to replicate the problem.
I've just uploaded a new version of Gogorender (1.1.0) that includes a
very simple function for splitting text at character boundaries:
def list_split(text, category, config):
return list(text)
This function is enabled by editing config.py:
gogorender = { 'size' : 24, 'split_fn' : 'list_split' }
I'm afraid I don't know much about Japanese. If there is a proper
algorithm for breaking a text string up into words, it could be added
in a similar way, i.e. as a function that takes a unicode string and
returns a list of unicode strings.
Tim.
signature.asc
Description: Digital signature
