Small python script to check and replace duplicated bib entries in your tex files

This reads your bibliographic entries and makes a python dictionary of duplicated entries. It then reads your *.tex files and replaces all those duplicated entries with one common entry. For eg:

...
'vihinen1994accuracy': ['Vihinen1994-sd', 'Vihinen1994-vq']
...

Here vihinen1994accuracy, Vihinen1994-sd, Vihinen1994-vq all refer to the same entry. This script reads your *.bib file and finds out such duplicates. Then it reads your chapter and replaces every instance of Vihinen1994-sd, Vihinen1994-vq with vihinen1994accuracy. Yor citations will then have no more duplicates!

Any changes made to your original chapter will be saved as a new file named _replaced_.tex. This will fix 95% of your problems (it certainly fixed mine!). However, you might still need to manually go through your references. For example, entries with title such as Multiple roles of the coding sequence 5' end... and Multiple roles of the coding sequence 5^{\prime} end... are the same but we cannot yet merge as one entry. Currently, we use \\emph and \textit as single entry because this is only so much that I can do.

See duplicate-bib-fix.ipynb for details. This is a Jupyter notebook file so you need to have Jupyter installed.

@article{breiman2001random,
  title={Random forests},
  author={Breiman, Leo},
  journal={Machine learning},
  volume={45},
  number={1},
  pages={5--32},
  year={2001},
  publisher={Springer}
}

@book{james2013introductiontostatlearning,
  title={An introduction to statistical learning},
  author={James, Gareth and Witten, Daniela and Hastie, Trevor and Tibshirani, Robert},
  volume={112},
  year={2013},
  publisher={Springer}
}

Example .bib file and a chapter from my own thesis are included.