This project looked really interesting until I read the words "mongodb". This is a blocker for me, but there is a greater issue that is actually worth reporting so here so I am:
The code is lacking a separation of concerns around data storage and usage. Looking at the code, this does seem to be the case.
The current interface is all static methods with no DB context or abstraction layer to allow other databases be used for data storage... so we cant even test if it actually works - as it a common need with oauth libs - without first setting up a MongoDB instance, and since MongoDB instances are a poison (its the wrong tool for the job 100% of the time due to its eventual consistency model, the a=inability to fail and recover gracefully, and lack of pure ACID support) nobody wants to do that.
Please support other databases.
This project looked really interesting until I read the words "mongodb". This is a blocker for me, but there is a greater issue that is actually worth reporting so here so I am:
The code is lacking a separation of concerns around data storage and usage. Looking at the code, this does seem to be the case.
The current interface is all static methods with no DB context or abstraction layer to allow other databases be used for data storage... so we cant even test if it actually works - as it a common need with oauth libs - without first setting up a MongoDB instance, and since MongoDB instances are a poison (its the wrong tool for the job 100% of the time due to its eventual consistency model, the a=inability to fail and recover gracefully, and lack of pure ACID support) nobody wants to do that.
Please support other databases.