Mangrove: A C++ Object-Document Mapper for MongoDB

For the next meetup, Adam Chelminski and Raphael Kargon will be presenting Mangrove, a C++ object-document mapper for MongoDB.

Abstract

Properly mapping between your program’s memory model and a database’s internal data representation is challenging and often involves large amounts of boilerplate code for the simplest tasks. MongoDB’s schema-less style, which provides much-needed flexibility for many applications, can also make this mapping particularly challenging. We present Mangrove, a MongoDB Object Document Mapper (ODM) for C++. Mangrove allows for automatic serialization of rich C++ objects to and from BSON, as well as “active record” style access to the database. Mangrove also provides query and update builders that allows compile-time type checking. We will discuss the design and implementation of Mangrove, which makes heavy use of C++11 and C++14 features.

Bios

Raphael Kargon is a rising junior at Brown University studying computer science. He is interning this summer at MongoDB, developing Mangrove with Adam Chelminski.

Adam Chelminski is a rising senior at Columbia University studying computer science. This is his second summer interning at MongoDB. Last year he worked on a new version of the MongoDB text index which improved the support for searching Unicode text.