Python tools for aggregating physics literature

The pheed (PHysics fEED) package collects physics literature meta data from various online sources and normalizes the resulting data into a standard format. The package provides three principal data structures. A simplified summary, a Source produces and Article which has an Author. It is possible to search for new articles from a source given some temporal and categorial information, such as "new this week" or "quantum gravity", as well as search for new articles by particular authors. Where official APIs exist, we use them, else we resort to web-scraping. As a consequence, the pdf form of the articles may not always be available. We aim to add these features where possible while respecting the Terms and Conditions of the sources involved. The package hasn't added explicit support for non-peer-reviewed sources, such as Blogs, though we realize that these can be quite popular in the community and may add support in the future.

Principal Structure

The below outlines the key components of the pheed package, and defines some of the relevant attributes or purposes of each. Examples below will explain usage of each of these components.

• Source: An online origin for physics articles
• Article: A peer-reviewed publication, typically consisting of a pdf and some meta-data
• Author: Contributor to an Article, with the expected attributes (name, etc.)

Wrapped Sources

Though the pheed package contains all the tools necessary to wrap additonal sources, it comes with several sources already wrapped out of the box:

• APS Journals
• IOP Journals
• Arxiv