2020-08-16: OCL Squad call - MVP Release update and QA update

Friendly reminder about our OCL for OpenMRS Squad call this coming week. I’m posting this a couple days ahead as I will be away; however, @michaelbontyes has graciously agreed to take on the Product Manager role again this week and will shepherd the call (thank you Michael!). :pray:

Notes : https://notes.openmrs.org/oclsquad2020

When : Wednesday 7:30pm IST | 5pm Nairobi | 4pm Cape Town | 2pm UTC | 10am Boston | 7am Seattle

Where : Join from computer or mobile: https://iu.zoom.us/j/750230470 Meeting ID: 750 230 470 One tap mobile +12532158782,750230470# US +13462487799,750230470# US

Agenda @michaelbontyes and I were thinking of the following; apologies if some of this was recently covered, and both of us missed it:

  • MVP Released on Friday :tada:

  • Booking time to discuss next steps for local-editing workflow: Please do this doodle and Grace will set it up when she’s back: https://doodle.com/poll/m85vuf9k6s5gzmzv

  • Technical QA Update - @k.joseph

    • Recent Automated Tests update from Kaweesi:
      • The Login/Logout feature is covered partially, there’s not a test for a failed login
      • Dictionaries create, view and edit are covered
      • Concept search, create or import, view, edit are all covered. Retrieving, sorting, filtering are broken
      • Source and progress notification functionalities aren’t covered.
      • The tests ran for at-least 13 minutes on my Mac(Processor:2.7 GHz Intel Core i5, Memory:16 GB 1867 MHz DDR3) and 25 minutes on CI
    • QA-related questions to review together:
      • Any new updates from Kaweesi, and progress on bringing together Cucumber and Cypress?
      • Who can get to know test suite with Kaweesi to share the knowledge?
      • Where are automated tests currently running - QA, Staging, or Production environments? Which ones do we want to use? Ideal = some key tests for all, in case of issues when in future, for example, if they’re referencing a different backend api version
      • How much automated test coverage do we currently have, overall?
      • Performance Check, e.g. for CSV import - did we check if the overall performance is quick enough?
  • Onboarding Status - @jwnasambu @tendomart @insookwa

    • Any blockers? Experience thus far?
    • Unit Tests progress
    • Tickets Update
  • Experience using MVP for Bangladesh - @muhima08

    • Any support needed? Issues found so far?
  • Other Ongoing Work - @kirity @swedhan

    • Any OCL API dependencies that still need to be resolved?

    • Multiple Entries of Custom Concept bug

      • Background: Turned out to be a reference deletion issue, and not a new bug, just newly discovered (i.e., not a new bug, not a regression). OCL API v2 (latest production release of OCL API) doesn’t have this bug, so was not a blocker for MVP release.
      • Cause: failure to delete a previously published reference in a collection when a previously published custom concept is edited. Our client issues the proper DELETE request and API acts like it was done, but doesn’t actually do the deletion.
      • Any update from the OCL Team on bug fix in OCL API v1? Can we start editing custom concepts?
    • Are these issues resolved?

      • Editing a private organization dictionary - unblocked?
      • Need to verify concept Numeric is current, not using default value unexpectedly
      • CSV import - unblocked, right?

@paynejd @dkayiwa @burke @ball @michaelbontyes @muhima08 @mogoodrich @akanter @tendomart @kirity @swedhan @insookwa @jwnasambu @tendomart

2 Likes

Written out a sample Login/logout test at: OCLOMRS-884: Cucumber with cypress by kaweesi · Pull Request #595 · openmrs/openmrs-ocl-client · GitHub

travis-ci which loads its own local instance

i think above 60% broken tests inclusive but very minimal unit test coverage

Not really except for my computer specific response above, i consider it worth our time, IPV; cypress performed not as i expected, i think puppeteer would perform a lot better but i think it’s worth testing out csv import with a small and large dataset and compare

2 Likes

Thanks for asking.

1 Like

Thank you for the update, @jwnasambu! Keep up the good work.

Thank you for this clear update @k.joseph! In your opinion, what do you think we should focus on next as a quad, so that the next release and future releases don’t require so much manual testing?

@grace did you also say

How about the fact that no failure message is returned to the user on failled login.

Actually just seen this https://issues.openmrs.org/browse/OCLOMRS-855 for MVP+.

1 Like

Yep! You’re welcome to pick that one up :slight_smile: