sage-data-client 0.6.0

Official Sage data client.

Sage Data API Client

This is the official Sage Python data API client. Its main goal is to make writing queries and working with the results easy. It does this by:

1. Providing a simple query function which talks to the data API.
2. Providing the results in an easy to use Pandas data frame.

Installation

pip install sage-data-client

Usage Examples

Query API

import sage_data_client

# query and load data into pandas data frame
df = sage_data_client.query(
    start="-1h",
    filter={
        "name": "env.temperature",
    }
)

# print results in data frame
print(df)

# meta columns are expanded into meta.fieldname. for example, here we print the unique nodes
print(df["meta.vsn"].unique())

# print stats of the temperature data grouped by node + sensor.
print(df.groupby(["meta.vsn", "meta.sensor"]).value.agg(["size", "min", "max", "mean"]))

import sage_data_client

# query and load data into pandas data frame
df = sage_data_client.query(
    start="-1h",
    filter={
        "name": "env.raingauge.*",
    }
)

# print number of results of each name
print(df.groupby(["meta.vsn", "name"]).size())

Load results from file

If we have saved the results of a query to a file data.json, we can also load using the load function as follows:

import sage_data_client

# load results from local file
df = sage_data_client.load("data.json")

# print number of results of each name
print(df.groupby(["meta.vsn", "name"]).size())

Integration with Notebooks

Since we leverage the fantastic work provided by the Pandas library, performing things like looking at dataframes or creating plots is easy.

A basic example of querying and plotting data can be found here.

Additional Examples

Additional code examples can be found in the examples directory.

If you're interested in contributing your own examples, feel free to add them to examples/contrib and open a PR!

Reference

The query function accepts the following arguments:

• start. Absolute or relative start timestamp. (required)
• end. Absolute or relative end timestamp.
• head. Limit results to head earliest values per series. (Only one of head or tail can be provided.)
• tail. Limit results to tail latest values per series. (Only one of head or tail can be provided.)
• filter. Key-value patterns to filter data on.