[Updates] ADD French, setup file, python3 support

[jacquesfize]

Hi Bermi,

Just found your module for pluralizing and singularizing. I needed both French and English support, so I changed your code ;) For the French support, I used the implementation made by @sblondon. Then, I added Python3 support and a setup file. In order to ease the module usage, I also changed the init.py file (see Getting Started section in the README).

The French implementation is a bit rough (not using regexp), so maybe a regexp transformation would be nice ;)

Best Regards,
Jacques.

[sblondon]

I think it's nice to have a generic library for several languages. :-)

• There are tests in pluralizefr so you can refactor it if you want without the fear of regression.
• There is a bug in the singularize() function:
  if the word parameter does not match a case, the function should return the word (for example the word riz). Currently, the returned value is None.

The current version of pluralizefr library cares only about lowercase. Do you want I modify the code to deal with uppercase?

I will add the singularize() feature to pluralizefr I received as PR: it was a WIP work but I think it's good enough. see sblondon/pluralizefr#1

I can upload the pluralizefr library to pypi.org if you want to use it quickly. Perhaps python-inflector could depends on it if the maintainer doesn't want to include the implementation?

[jacquesfize]

Sure, you can modify the code (for the uppercase support) ;) As for thesingularize bug, do you want me to modify the code? (and make another pull request ?)

Concerning the upload of pluralizefr to pipy.org, I don't think it's necessary. If pluralizefr was composed of many features, why not? But here, I think that aggregating pluralizefr to your code seems more relevant.

[sblondon]

I think there is a mistake: I wrote pluralizefr, not python-inflector. For the integration in python-inflector, you need a reply from bermi.

[isard]

@Jacobe2169 just saw rules/french.py ... wouldn't it be better to use u"" always?

I guess you used u"" if the word has at least one non ASCII character and "" otherwise; but it could be a problem if you mistype a word ("tete", for instance) and someone else corrects it to "tête" instead of u"tête" . :-)

[jacquesfize]

Sorry about the delay! @sblondon, sure you can integrate singularize() to pluralizefr, I think it's relevant to have both operations!

As for @isard question, in Python3, str variables are Unicode (source here). But, if you're working on python2.7, maybe adding the following import can help :

# Python 2 and 3
from __future__ import unicode_literals    # at top of module

s1 = 'The Zen of Python'
s2 = 'きたないのよりきれいな方がいい\n'

[sblondon]

@Jacobe2169 finally, I reused the moz4r's pull-request for the implementation of singularize() of pluralizefr. It catches more cases and I added some testcases. There are still some problems (for words finishing by 'aux' in plural which can finsih by 'al' or 'au' in singular).