Production code for PrePubMed
Python HTML
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.
arxiv update 20170203 Feb 4, 2017
biorxiv update 20170203 Feb 4, 2017
f1000research update 20170203 Feb 4, 2017
figshare 20160117 Jan 17, 2017
nature preparing for database update Jan 6, 2017
osf preparing for database update Jan 6, 2017
peerj update 20170203 Feb 4, 2017
preprints update 20170203 Feb 4, 2017
static ready for production Apr 19, 2016
templates help page Jan 21, 2017
wellcome update 20170203 Feb 4, 2017
winnower update 20170203 Feb 4, 2017
.gitignore update 20161214 Dec 14, 2016
LICENSE Initial commit Apr 4, 2016 Update Jul 29, 2016 nature preceding and wellcome research Jan 17, 2017 nature preceding and wellcome research Jan 17, 2017 started search function Apr 10, 2016 added Aug 11, 2016 author order Apr 23, 2016
todolist.txt bug fixes Jun 7, 2016 update fix Jan 18, 2017 figshare update Jul 29, 2016

This is the production code for

An article discussing how PrePubMed works is available at

PrePubMed runs on Django 1.8, and needs BeautifulSoup and requests.

The GRIM test plot function needs NumPy and Matplotlib.

I am currently using sqlite3 as the database engine. If I need to I could switch to either MySQL or PostreSQL, but query speed seems fine for now.

Despite the fact that I have a static folder I'm just relying on AWS S3 right now to serve the static files. The main site is in the "mysite" directory. Yes, I know the SECRET_KEY is visible, but nothing currently is getting encrypted. A new key will be made and stored in a file when needed. The only app is the "papers" directory, which contains the models logic along with some needed files.

The file runs once a day on its own and updates the database.

Each preprint server has its own folder. In each folder is code for the initial indexing (these might be outdated), and a text file that contains the articles that get indexed. There are also update and error logs for

If you want to get this running locally you'll need to change the settings file, namely the host settings. GitHub doesn't want the sqlite3 database in the repository so you will have to generate the database by making migrations and running "". This could take time. If you want it to go faster only make the database for a fraction of the articles.

This is the current to do list:

Improve indexing code (I'll do this)

Saved searches and email notifications (I can do this, need to figure out best way to authenticate email addresses)

If you want to contribute to PrePubMed fork the repository and try to limit your changes to what is necessary.