Updates according to the newer version of the ORCID APIs

[vdmitriyev]

Please find below short description of the pull request:

• json mapping of the 'orcid-identifier' changed according to the new API of ORCID.
• json mapping of the 'work-citation' changed according to the new API of ORCID.
• Correcting issues with parsing keywords.
• Adding logging facilities to the script.
• Adding folder with example scripts and also adding example script that extracts all bibtex typed works and put them in one single bibtex file.

[yarikoptic]

could someone look into merging this PR? decided to give a try to this python module and some things don't work any longer so thought that may be this PR would address it:

*In [4]: john.publications
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-4-26ff28f7833f> in <module>()
----> 1 john.publications

/home/yoh/deb/perspect/orcid-python/orcid/rest.pyc in publications(self)
    105     def publications(self):
    106         if self._loaded_works is None:
--> 107             self._load_works()
    108         return self._loaded_works.publications
    109 

/home/yoh/deb/perspect/orcid-python/orcid/rest.pyc in _load_works(self)
     99     def _load_works(self):
    100         resp = requests.get(ORCID_PUBLIC_BASE_URL + self.orcid
--> 101                             + '/orcid-works', headers = BASE_HEADERS)
    102         self._loaded_works = Works(resp.json())
    103 

TypeError: cannot concatenate 'str' and 'NoneType' objects

*In [5]: john.researcher_urls
Out[5]: []

In [6]: john.orcid

[vdmitriyev]

I cloned this repo and applied changes. You can find updated version here - https://github.com/vdmitriyev/pyorcid . In folder with examples (in the same repo) you can also find full functioning examples demonstrating how to work with given package. Good luck!