Accelerate pySBOL3 loading and searching

[jakebeal]

Background

The pySBOL3 library is used for many SBOL3 projects, but certain operations currently do not scale well. In particular, with large documents:

• The document can take a long time to load
• Searching the document for objects involves a walk through every object checking for matching IDs.
• Calling lookup() to resolve object pointers searches the entire document via find (See issue: https://github.com/SynBioDex/pySBOL3/issues/237)

All of these should be able to be greatly improved by changing data structures to accelerate these operations. This will also require additional book-keeping, error-checking, and testing.

Goal

The goal of this project is to provide at least a 10x improvement in the speed of loading and searching large SBOL3 documents with pySBOL3.

Difficulty Level 2

The definition and focus of the project is clear, but it will require some sophistication in understanding Python.

Skills

Essential skills: Python Nice to have skills: SBOL

Public Repository

https://github.com/SynBioDex/pySBOL3

Potential Mentors

@jakebeal @bbartley @tcmitchell

[Aakash-02]

I would like to work on this project. How to become a project member?

[jakebeal]

@Aakash-02 I'm afraid that you have missed this year's deadline for applications. There are related projects on SBOL available at the NRNB Google Summer of Code project list, however.