CentralBufferManager.swift

Purpose

CentralBufferManager.swift is responsible for aggregating and managing streaming data received from wearable sensors, and handling the logic for flushing this data to a persistent backend or storage. It is used specifically for buffering data strings before export.

---

Class: CentralBufferManager

class CentralBufferManager {
    var dataBuffer: [String] = []
    let bufferLock = DispatchQueue(label: "com.sensoriahandler.bufferLock")

    func addData(data: String)
    private func processBuffer()
    private func sendDataToServer(_ data: [String])
}

---

Properties

• dataBuffer: An array of raw string data representing encoded telemetry readings (such as CSV-style lines).

• bufferLock: A DispatchQueue used to synchronize access to the buffer (though the locking logic is currently commented out).

---

Method: addData(data:)

func addData(data: String)

• Appends a new data string to the dataBuffer.

• If the buffer exceeds a certain threshold (currently 1000 entries), it triggers processBuffer().

Note: Locking logic using bufferLock.async is commented out and could be re-enabled for thread safety.

---

Method: processBuffer()

private func processBuffer()

• Optionally sorts the buffer (sorting is commented out).

• Sends the buffered data to the server using sendDataToServer(...).

---

Method: sendDataToServer(_:)

private func sendDataToServer(_ data: [String])

• Sets a data type ("S" likely stands for Sensoria).

• Delegates to LocationDataManager.shared.fetchLocationAndSaveData(...), which presumably attaches GPS metadata before saving.

• Clears the buffer only upon successful transmission.

---

Integration

CentralBufferManager plays a key role in ensuring:

• No data loss during high-frequency sampling.

• Buffered, batch-style transmission for performance and power efficiency.

• Potential location-aware export (if GPS is enabled in LocationDataManager).

---

Suggestions for Improvement

• Re-enable bufferLock.async to protect buffer from concurrent modification.

• Parameterize the threshold (currently hardcoded at 1000).

• Optionally add retry logic if the data fails to send multiple times.