Skip to content

A python package that communicates with 3rd party taxonomy vendors

License

Notifications You must be signed in to change notification settings

openedx/taxonomy-connector

Taxonomy

PyPI Codecov CI

The taxonomy service is a library that can be installed in other edX components that provides access to third party taxonomy vendors. EMSI is currently the only vendor available, but others may be integrated as necessary. This service can communicate with the vendor to get job, skill, and salary data. This service can also be used to submit data (course descriptions, etc.) to the vendor to produce potential matches for a skill or job.

After submitting a pull request, please use the Github "Reviewers" widget to add relevant reviewers and track review process.

Getting Started

To install taxonomy-connector, for example, in Course Discovery, follow these steps:

  1. It is recommended that you clone this repo into a sub-folder of your working directory. Create a sub-folder with name src if it doesn't already exist.
  2. Clone this repository into the src folder.
  3. Go to the shell of the host environment where you want to install this package and run this pip install -e /edx/src/taxonomy-connector
  4. Changes made into the taxonomy repository will now be picked up by your host environment.

Notes:

  • In order to communicate with EMSI service, you need to set the values of client_id and client_secret. These values are picked up from the host environment so you need to pass them in .yaml file of the host environment.
  • Also, to make taxonomy work, the host platform must add an implementation of data providers written in ./taxonomy/providers
  • Taxonomy APIs use throttle rate set in DEFAULT_THROTTLE_RATES settings by default. Custom Throttle rate can by set by adding ScopedRateThrottle class in DEFAULT_THROTTLE_CLASSES settings and taxonomy-api-throttle-scope key in DEFAULT_THROTTLE_RATES
  • For the skill tags to be verified, the management command finalize_xblockskill_tags needs to be run periodically.
  • Also, You can configure the skill tags verification by setting the values of SKILLS_VERIFICATION_THRESHOLD, SKILLS_VERIFICATION_RATIO_THRESHOLD, SKILLS_IGNORED_THRESHOLD and SKILLS_IGNORED_RATIO_THRESHOLD in the host platform or by passing the values to the command using the args --min-verified-votes, --ratio-verified-threshold, --min-ignored-votes and --ratio-ignored-threshold.
REST_FRAMEWORK = {
    'DEFAULT_THROTTLE_CLASSES': (
        'rest_framework.throttling.UserRateThrottle',
        'rest_framework.throttling.ScopedRateThrottle'
    ),
    'DEFAULT_THROTTLE_RATES': {
        'user': '100/hour',
        'taxonomy-api-throttle-scope': '60/min',  # custom throttle rate for taxonomy api
    },
}

Developer Notes

  • To run unit tests, create a virtualenv, install the requirements with make requirements and then run make test
  • To update the requirements, run make upgrade
  • To run quality checks, run make quality
  • Please do not import models directly in course discovery. e:g if you want to import CourseSkills in Discovery, use the utility get_whitelisted_course_skills instead of directly importing it.

Reporting Security Issues

Please do not report security issues in public. Please email security@openedx.org.

Getting Help

Have a question about this repository, or about Open edX in general? Please refer to this list of resources if you need any assistance.