Migas Client

A Python package to communicate with a migas server.

About

migas (mee-gahs) is a Python client to facilitate communication with a migas server.

Usage

To start communicating with a migas server, the client must first be setup.

import migas; migas.setup()

By default, migas-py will communicate with the official hosted migas server. However it can easily be configured to communicate with any hosted migas server.

import migas; migas.setup(endpoint='your-endpoint')

setup() will populate the internal configuration, which is done at the process level.

API

migas includes the following functions to communicate with the telemetry server:

[migas.add_breadcrumb] (#add-breadcrumb)

---

Send a breadcrumb with usage information to the server.

Mandatory

• project - application name
• project_version - application version

Optional

• language (auto-detected)
• language_version (auto-detected)
• process:
  • status
  • status_desc
  • error_type
  • error_desc
• context:
  • user_id (auto-generated)
  • session_id
  • user_type
  • platform (auto-detected)
  • container (auto-detected)
  • is_ci (auto-detected)

add_breadcrumb example

>>> add_breadcrumb('nipreps/migas-py', '0.0.1', status='R', status_desc='Finished long step')
>>>

[migas.check_project] (#check-project)

---

Check a project version against later developments.

Mandatory

• project
• project_version

check_project example

>>> check_project('nipreps/migas-py', '0.0.1')
{'success': True, 'flagged': False, 'latest': '0.4.0', 'message': ''}

[migas.get_usage] (#get-usage)

---

Check number of uses a project has received from a start date, and optionally an end date. If no end date is specified, the current datetime is used.

get_usage example

>>> get_usage('nipreps/migas-py', '2022-07-01')
{'hits': 7, 'message': '', 'unique': False, 'success': True}

[migas.track_exit] (#track-exit)

---

Register an exit function to send a final ping upon termination of the Python interpreter. Useful when monitoring a process that may preemptively error. The inputs are equivalent to add_breadcrumb()

User Control

migas can controlled by the following environmental variables:

Envvar	Description	Value	Default
MIGAS_OPTOUT	Disable telemetry collection	Any	None
MIGAS_TIMEOUT	Seconds to wait for server response	Number >= 0	5
MIGAS_LOG_LEVEL	Logger level	Logging levels	WARNING

Configuration

The internal configuration stores the following telemetry information:

• language and language version
• operating system
• run within a container
• run from continuous integration