[Epic] Datahub QA tests

[zelima]

As an admin, I want to have test spec written and executed twice by different people so that I know where I'm good or where I should improve.

As an admin, I want to monitor test outcome, so that I know the quality of the product is improving or no.

As a product owner, I want to test the product under predefined criteria (spec), so that showcase page is perfected and we have a perfect data push flow that only has minor bugs - the product is attractive to users!

Acceptance Criteria

• [x] test spec covers the most of possible scenarios and is documented in spreadsheet
• [x] tests are executed regularly (daily) by responsible person(s)
• [x] Outcome of the tests is written in spreadsheet

Tasks

• [x] come up with specs and fill the spreadsheet - https://docs.google.com/spreadsheets/d/1ssAH2LmHMWyc38H6I85GP9OZLcJOK0klw-RZmKuEpZk/edit#gid=0
• [x] decide eabout example packages to use from https://github.com/frictionlessdata/test-data
  • [x] identify what's missing and add if necessary
• Implement tests - automate as much test cases as possible (moved to #120 )
• [x] follow basic loop and report outcome in spreadsheet

analysis

Basic loop that should be good for the launch (can be used in the sign up process):

• Install data command line tool
• data get sample-data (Or use your own data) [not a focus]
• data push
  • (init behind the scenes will add a graph for them)
• Visit the showcase page
  • My data appears quickly
  • I.e. Instantaneously after cli says it’s published
  • Tweet the result ;-) with a good title (e.g. 'US GDP - datahub.io')
  • Bonus: "twitter cards" / facebook cards
  • Get the JSON …
  • Validation report
• Can use the data
  • in my code
  • data get xxx

[zelima]

@AcckiyGerman think we can close this one, WDYT?

[AcckiyGerman]

@zelima yes, the issues pipeline is working now.

On Feb 16, 2018 14:30, "Irakli Mchedlishvili" notifications@github.com wrote:

@AcckiyGerman https://github.com/acckiygerman think we can close this one, WDYT?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/datahq/pm/issues/109#issuecomment-366223481, or mute the thread https://github.com/notifications/unsubscribe-auth/ALKlIsXCClb5dM-HlzXxBSgazOklQ23oks5tVXT6gaJpZM4RwInB .

[AcckiyGerman]

Status:

• all key user scenarios are working and covered by tests
• advanced user scenarios are covered by tests (not all scenarios are working yet, but we have documented bugs and we are actively fixing this issues)
• test-data is created

Not ready:

• [ ] automate as much tests as possible

@zelima @anuveyatsu Automating tests are important. I suppose to create separate issue for tracking its progress, so we could close this EPIC

[zelima]

@AcckiyGerman +1

[anuveyatsu]

@AcckiyGerman @zelima I've opened a new issue for automation of these tests #120 so closing this one.