datacollector-api-client 0.1.4

Thin client for Azure Monitor DataCollector API.

A thin client to send log data to Azure Monitor by using the HTTP Data Collector API.

Installation

$ pip install datacollector-api-client

Usage

from datacollector_api_client.client import DataCollectorWrapper

wrapper = DataCollectorWrapper(LOGANALYTICS_WORKSPACE_ID, LOGANALYTICS_WORKSPACE_KEY)

data = [{
   "application": "my_app",
   "message": "my log message"
}]

response = wrapper.log_info(structured_log_message=data)
print(response)

To log a message as error, use: wrapper.log_error(structured_log_message=data)

You can also override the default name of the tables by setting parameter: log_type=”your_preferred_table_name”

Enriching structured logging with Databricks information

Pass your dbutils instance from your Databricks session to the library

wrapper = DataCollectorWrapper(LOGANALYTICS_WORKSPACE_ID, LOGANALYTICS_WORKSPACE_KEY, dbutils)

data = [{
   "application": "Notebook",
   "message": f'Number of rows in dataframe: {df.count()}'
}]

response = wrapper.log_info(structured_log_message=data)
print(response)

The following data is also collected and appended to your log:

• dbutils.notebook.entry_point.getDbutils().notebook().getContext().notebookPath().get()

• dbutils.notebook.entry_point.getDbutils().notebook().getContext().tags().apply(‘clusterId’)

• dbutils.notebook.entry_point.getDbutils().notebook().getContext().tags().apply(‘sessionId’)

• dbutils.notebook.entry_point.getDbutils().notebook().getContext().tags().apply(‘user’)