python-api

python-api

The Python API example shows how to do Honeycomb instrumentation with a Flask app.

Events are created per-HTTP-request (following the Honeycomb one event per unit of work model) using the @app.before_request decorator and sent after the request using the @app.after_request decorator. Because the events are stored on g, Flask's thread local storage, the events contain a variety of default properties describing the request as well as custom fields that can be added by any handler using g.ev.add_field.

Run Locally

Set your Honeycomb write key to HONEYCOMB_WRITEKEY.

MySQL and Flask must be installed (pip install flask).

First, create the database used by the app.

$ mysql -uroot -e 'create database example-python-api;'

Then run the Python app on localhost:

$ FLASK_APP=app.py flask run

A basic REST API for todos is exposed on port 5000.

$ curl \
    -H 'Content-Type: application/json' \
    -X POST -d '{"description": "Walk the dog", "due": 1518816723}' \
    localhost:5000/todos/
...

$ curl localhost:5000/todos/
[
  {
    "completed": false,
    "description": "Walk the dog",
    "due": "Fri, 16 Feb 2018 21:32:03 GMT",
    "id": 1
  }
]

$ curl -X PUT \
    -H 'Content-Type: application/json' \
    -d '{"description": "Walk the cat"}' \
    localhost:5000/todos/1/
{
  "completed": false,
  "description": "Walk the cat",
  "due": "Fri, 16 Feb 2018 21:32:03 GMT",
  "id": 1
}

$ curl -X DELETE localhost:5000/todos/1/
{
  "id": 1,
  "success": true
}

$ curl localhost:5000/todos/
[]

Run in Docker

This example can be run in Docker (Compose).

$ docker-compose build && docker-compose up -d

Event Fields

Name	Description	Example Value
errors.message	Message in the error encountered, if applicable	undefined
request.endpoint	Endpoint requested	/todos/
request.method	HTTP method	POST
request.path	Request path	/todos/
request.python_function	Python function serving the request	index
request.url_pattern	` Underlying routing pattern of the URL	/todos/<int:todo_id>/
request.user_agent	User agent for the request	curl/7.54.0
request.user_agent.browser	Web browser the request was served to	chrome
request.user_agent.platform	OS of the user agent	macos
request.user_agent.string	Literal user agent string	curl/7.54.0
request.user_agent.version	Version of the user agent	64.0.3282.186
response.status_code	HTTP status code of the response	404
timers.db.delete_todo	Time in milliseconds for DB call to delete a todo	23
timers.db.insert_todo_ms	Time in milliseconds for DB call to insert a todo	50
timers.db.select_all_todos	Time in millisconds for DB call to select all todos	11
timers.db.select_todo	Time in milliseconds for DB call to select a todo	4
timers.db.update_todo	Time in milliseconds for DB call to update a todo	50
timers.flask_time_ms	Total time in milliseconds Flask spent serving the request	75
todo.id	ID of the associated TODO	1

Example Queries