Fresh off of the back of announcing the library formerly known as Steem Stream has been moved over to Hive and called Hive Stream are some nice new updates to the library to make it more useful.
One of the biggest updates is more efficient streaming of the chain, removing some unnecessary code paths and fixing a number parsing bug in the block behind count which now properly works. Another is the library now has tests, quite a few of them (and more to come).
Perhaps the biggest improvement is fallback client nodes. The previous version didn't support falling back to different API nodes, given we live in the real world and things are not perfect, having the streamer not stop or crash is important. So, you can supply multiple hive API endpoints and it will fall back if they fail.
While I do not believe in 100% code coverage, going forward all code will have accompanying tests to ensure nothing breaks to keep it stable. If you're developing a Hive dApp whether it be a bot or fully-blown application, Hive Stream will make it easier to parse the blockchain and react to events.
The code is publicly available here on GitHub: https://github.com/Vheissu/hive-stream -- As always, contributions are always welcomed.