nanots 0.18.0

Python bindings for nanots embedded time series database

NanoTS

NanoTS is a super fast, embedded time series database (so, no server). It supports 1 writer per stream + an unlimited number of readers.

NanoTS chooses to pre-allocate its entire storage file, this has a number of advantages:

• You cannot fill up your servers disk with data. Set it up and forget about it
• We don't waste time & cpu creating and deleting files forever
• If you configure your writer with auto_reclaim=true (the default) you will always have the newest data

Check out: https://github.com/dicroce/nanots

Building

python -m build --wheel

Basics Example

import os
import tempfile
import time
import json
import nanots

def nanots_basics_example():
    try:
        # 1. CREATE DATABASE - 64KB blocks, 100 blocks = ~6MB
        nanots.allocate_file("test.nts", 64*1024, 100)
        
        # 2. WRITE DATA
        writer = nanots.Writer(db_file)
        write_context = writer.create_context("sensors", "Temperature readings")
        
        # Write 10 temperature readings
        base_time = int(time.time() * 1000)
        for i in range(10):
            timestamp = base_time + (i * 5000)  # Every 5 seconds
            temperature = 20.0 + i + (i * 0.5)  # Increasing temp
            
            data = json.dumps({
                "temp_c": temperature,
                "sensor": "temp_01"
            }).encode('utf-8')
            
            print("Writing data:", data.decode('utf-8'), "at", timestamp)
            writer.write(write_context, data, timestamp, 0)
        
        # 3. READ DATA
        reader = nanots.Reader("test.nts")
        
        # Read all data
        frames = reader.read("sensors", base_time, base_time + 50000)
        
        # Show first 3 records
        for i, frame in enumerate(frames[:3]):
            data = json.loads(frame['data'].decode('utf-8'))
            print(f"      {i+1}. {data['temp_c']}°C from {data['sensor']}")
        
        # 4. ITERATE THROUGH DATA
        iterator = nanots.Iterator("test.nts", "sensors")

        # position the iterator...
        iterator.find(base_time + 10000)

        print("Iterating through data:")
        for i, frame in enumerate(iterator):
            data = json.loads(frame['data'].decode('utf-8'))
            print(f"      {i+1}. {data['temp_c']}°C from {data['sensor']} at {frame['timestamp']}")
                
        # 5. DISCOVER STREAMS
        streams = reader.query_stream_tags(base_time, base_time + 50000)
        
    except Exception as e:
        print(f"❌ Error: {e}")

if __name__ == "__main__":
    nanots_basics_example()