dblptobib

Generate lists of papers in BibTeX or HTML format by gathering the necessary information from dblp. The goal here is to get a consistent and clean bibliography without the hassle of fiddling around with it manually.

Installation

Download the repository and run pip install . from inside the repository. This should download all dependencies (urllib3, xmltodict, appdirs) and give you excess to the command dblptobib from anywhere in your system.

If you want to make changes to the code and want them to apply immediately without rerunning the install command, you can run pip install -e . instead, to install the package using symlinks.

Usage

You have to specify two pieces of information when calling dblptobib: the papers that should be included and the output file. The output file is always the last argument; the format (BibTeX or HTML) is automatically deduced from the suffix (.bib or .html). For which papers to include there are different options.

List of Paper IDs

dblptobib list.txt out.bib

This reads a list of paper ids from the file list.txt, gets the information about the papers from dblp and writes the resulting BibTeX entries to out.bib. The paper ids are the ones used by dblp and list.txt should contain one id per line. Here is an example list.txt and the resulting out.bib:

conf/stoc/Cook71
conf/coco/Karp72

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% 1972

@inproceedings{Reduc_Among_Combi_Probl_CCC1972,
  title     = {Reducibility Among Combinatorial Problems},
  author    = {Richard M. Karp},
  pages     = {85--103},
  year      = {1972},
  booktitle = {Computational Complexity Conference (CCC)},
  doi       = {10.1007/978-1-4684-2001-2_9}
}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% 1971

@inproceedings{Compl_Theor_Proce_STOC1971,
  title     = {The Complexity of Theorem-Proving Procedures},
  author    = {Stephen A. Cook},
  pages     = {151--158},
  year      = {1971},
  booktitle = {Symposium on the Theory of Computing (STOC)},
  doi       = {10.1145/800157.805047}
}

All Papers of One Author

dblptobib author_id out.html

This gets all the papers of the author with the given id from dblp and outputs the result to HTML. As for the paper ids, the author ids are the ones used by dblp, e.g., my author id is 74/9125, so running dblptobib 74/9125 out.html outputs all my papers to out.html.

All Papers in the Local Database

dblptobib out.bib

This outputs all papers that are already in the local database, i.e., papers whose information have be gotten from dblp before.

The Local Database

Information retrieved from dblp is cached in a local database in the user data directory (e.g., ~/.local/share/dblptobib/ on Linux). To find this directory on your system, just run dblptobib without arguments.

To save calls to the dblp API (and time), an API call is only done if the information is not already in the local database. The only exception to this is when getting the list of papers of an author, as this list changes regularly for active researchers.

Manual Changes

In the rare occasion that the information from dblp is not to your liking, you can apply manual changes. This is done directly in the local database, which contains two folders, automatic/ and manual/. Do not touch automatic/, which contains all information gotten from dblp distributed over many json files. Instead, add your changes to manual/, which has the same directory structure as automatic. Every information in manual/ will overwrite the corresponding information in automatic/.

So if you, e.g., have a paper where dblp got the title wrong, then find the filename for this paper in the automatic/ folder and create a file with the same name in the manual/ folder with the following content.

{
    "title": "New Corrected Title."
}

Changing the Formatting and Other Features

If you want to adjust the formatting of the BibTeX or HTML output, you have to adjust the output_bib.py or output_html.py accordingly. They should be somewhat self-explanatory.

The python code supports other features like sorting/grouping by other things than the year of publication or filtering of, e.g., preprints. They are not accessible by just calling dblptobib, so you have to do some adjustments to the python code if you want to use them. Having a look at dblptobib.py and output.py is probably a good start for this.