add labels for lexical and synset nodes

[rhin0cer0s]

I made this branch for the node labels issue.

I'll take this opportunity to ask some questions :

• if I want to work on an other task should I branch again ( from node_labels ? master ? )
• I sometimes think about some improvements ( like possibility to add as much as labels as we need without touching directly writing functions, create a folder for each language in csv_files, etc ... ) should I open a issue for each or a big one for all ?

[moreymat]

My answer to (1) is quite trivial, but I will try it anyway: it depends on the level of coupling between the existing branch and your future modifications:

• no interaction or dependency: you should branch from master,
• interaction or dependency, but not exactly related: branch from the branch,
• interaction or dependency, really related: continue on the current branch.

Now that we have the foundations, we should be able to make incremental, well defined changes that impact only a small fraction of the code base. In these cases I can accept pull request several times a day, so you should be able to always branch from master.

If your PR is ready and I have not accepted it yet, you could do some incremental work on the report in the mean time :-)

About (2), I would advise you to open as many issues as needed so that each issue can be fixed by one "independent" PR. You can reference related issues ("#" followed by the number of the issue) when it is handful. Focused issues are easier to discuss and track, focused PR are easier to merge :-)

[moreymat]

@rhin0cer0s @zorgulle : every time you think a PR is ready for merge, please signal it in a comment on the PR.

[moreymat]

@rhin0cer0s : this is a good use of labels, however I would prefer "NonLexicalized" rather than "Fake", to adhere to the Wordnet terminology (see #17 ).

[rhin0cer0s]

Ok I think it is good for this one, we have labels on nodes.