Fixing issue #6892: collect and report aggregate stats about client browsers through telemetry

[paulbusse]

Summary

The change collects the HTTP user agent in (kolibri/core/auth/api.py)SessionViewset.get_session_response and passes it on to UserSessionLog.update_log. The latter function uses the ua_parser module to parse the HTTP user agent string into the requested string. and saves it to the database.

The data is summarized in extract_facility_statistics.

Other changes:

• ua_parser 0.10.0 was added to the dependencies
• logger/migrations/0008_usersessionlog_device_info.py was generated
• test_extract_facility_statistic was adapted to add the new field

Reviewer guidance

• the update_log definition has been adapted. I could only find one call to the function. So I did not default the new argument.
• ua_parser is very fault resistant, I used that in the code. I don't check for None on the families
• I used a multiline statement to generate the device_info string, that may be against coding guidelines.

I tested this using all available browsers (chrome, edge, firefox, comodo dragon) on all available os's (windows 10, ubuntu). The pingback is not tested.

References

• collect and report aggregate stats about client browsers through telemetry #6892

---

Contributor Checklist

PR process:

• PR has the correct target branch and milestone
• PR has 'needs review' or 'work-in-progress' label
• If PR is ready for review, a reviewer has been added. (Don't use 'Assignees')
• If this is an important user-facing change, PR or related issue has a 'changelog' label
• If this includes an internal dependency change, a link to the diff is provided

Testing:

• Contributor has fully tested the PR manually
• If there are any front-end changes, before/after screenshots are included
• Critical user journeys are covered by Gherkin stories
• Critical and brittle code paths are covered by unit tests

Reviewer Checklist

• Automated test coverage is satisfactory
• PR is fully functional
• PR has been tested for accessibility regressions
• External dependency files were updated if necessary (yarn and pip)
• Documentation is updated
• Contributor is in AUTHORS.md

[rtibbles]

Everything looks good - one place where the code cold be clearer (in the place with the multi-line string concatenation as you suspected).

Tests only failed because of linting - the easiest way to manage this is to run:

pre-commit install
pre-commit run --all-files

This will run our linting pass on all the files

[paulbusse]

Thanks for the tip.

[rtibbles]

Thanks for the PR! I pushed the linting fixes so we can get this merged straight away. Great stuff!

[rtibbles]

Test failure appears to be erroneous, but Travis is not letting me restart the job. Nothing in the PR is touching the code paths that the failing test covers. Merging. Will fix on the release branch if it continues to fail.