Stream Function

Introduction

To process continuous high-frequency data in real-time, YoMo adopts the Functional Reactive Programming (FRP) Paradigm and makes stream computing easier than ever. YoMo uses the QUIC transport protocol, which largely improves the speed of data transfer, and maps QUIC streams into Rx streams. The user can then use the stream operators to process the stream as they like:

func Handler(rxstream rx.Stream) rx.Stream {
    stream := rxstream.
        Debounce(50).
        Map(computePeek).
        SlidingWindowWithTime(SlidingWindowInMS, SlidingTimeInMS, slidingAvg)

    return stream
}

Besides translating the raw data into usable information, a stream function can also display data in real-time on a web page, save the stream to a time series database, or print the nicely formatted data to StdOut:

func Handler(rxstream rx.Stream) rx.Stream {
    stream := rxstream.
        Map(store)
    return stream
}

A few things to note:

• The original stream sent by zipper is immutable. The Handler function operates on a stream by adding new items to the existing stream.

Stream Operators

Most operators take a stream as input and return a new stream as output, so they can be chained together. Some of the operators are implemented by ReactiveX:

Map - applies a function to each item emitted by an Observable.

TakeLast - takes the last n items emitted by an Observable.

Some are unique to YoMo:

• AuditTime
• DefaultIfEmptyWithTime
• RawBytes
• SlidingWindowWithCount
• SlidingWindowWithTime
• StdOut
• ZipMultiObservers

Examples

1. A simple real-time noise monitoring system. link
2. Working with multiple stream functions. link