BibJSON is a convention for representing bibliographic metadata in JSON; it makes it easy to share and use bibliographic metadata online.

It is a form of JSON - a simple, useful and common way of representing data on the web; we use it to shift information around between our apps.

• A BibJSON record is a JSON object
• A BibJSON collection is a JSON object containing "metadata" followed by "records"
• The "records" key in a collection points to a list of BibJSON records (JSON objects)
• The collection and the records both have the "collection" key, and their value should be the same
• Each record should have a "cid" - an identifier unique within the parent collection
• Each record should have a "type" - such as "article", "book", or even "author"
• Record type places no constraint on what can be placed in the record
• The default set of keys are based on the bibtex keys
• BibJSON keys are lowercase, no spaces, and usually singular
• The keys can point to strings, lists, or objects
• Any thing that is a simple string should remain so
• Where object complexity is required, make it an object
• Where additional keys are namespaced, include a "namespace" declaration in the collection "metadata"
• BibJSON APIs may return other metadata relevant to the parent app; developers can identify such metadata by prefixing the key with "_"; just ignore what is not useful to you

BibtexParser is a python library to parse bibtex files. The code is tested with unittests.