Clone Tools
  • last updated a few seconds ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
Nikodem Graczewski – this looks good I resolved the last issue I had, as it looks like the tests were simplified

Nikodem Graczewski – this looks good

I resolved the last issue I had, as it looks like the tests were simplified

OLMIS-2921: Removed redundant logic tests from html specs

  1. … 4 more files in changeset.
Nikodem Graczewski Thanks for breaking down your concerns a little – You are right that there is a need for automated tests with a scope that is larger than a unit test We definitely don't want to...

Nikodem Graczewski Thanks for breaking down your concerns a little – You are right that there is a need for automated tests with a scope that is larger than a unit test

We definitely don't want to have two unit tests that test the same thing – that violates DRY and orthogonal code

I feel that HTML and HTML unit tests should check for the appearance of specific strings/elements, and that methods/events defined in the form call the right function // I'd be concerned if they had a larger scope than that

We do need to build requirements around UI integration tests, as there is weird behavior there – and the only way we can prove we are fixing are integration tests

anyways – Yes, please remove duplicative tests

cool — I feel like positive configurations are easier to reason about.... but I'm not worried about this too much ie "trackable:false"

cool — I feel like positive configurations are easier to reason about.... but I'm not worried about this too much

ie "trackable:false"

https://openlmis.atlassian.net/browse/?focusedCommentId=44322&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#OLMIS-2921comment-44322

https://openlmis.atlassian.net/browse/?focusedCommentId=44322&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#OLMIS-2921comment-44322

Nick Reid I guess a large chunk of the code tested here (and in other html tests) is in the controller: *clearing reason selection after status change *clearing decommission date after status ch...

Nick Reid

I guess a large chunk of the code tested here (and in other html tests) is in the controller:

  • clearing reason selection after status change
  • clearing decommission date after status change
  • cancel button opening confirmService (basically all cancel button tests)
  • calling inventoryItemService after submitting form


Should I remove those too? I always though about those tests as a way of testing the whole page integrity.

Nick Reid It's our custom property for (not) tracking the state with stateTracker.

Nick Reid It's our custom property for (not) tracking the state with stateTracker.

Yea... but the code that actually does that work is the controller.... If we were setting that logic in the html (like a form action parameter) then this would make sense here If the HTML doesn't ...

Yea... but the code that actually does that work is the controller.... If we were setting that logic in the html (like a form action parameter) then this would make sense here

If the HTML doesn't know about it, then the HTML shouldn't test it

Looks good Going to make a different ticket about moving notifications to be above modal layers

Looks good

Going to make a different ticket about moving notifications to be above modal layers

What does this do? I can't find a mention of it in the code or ui-router docs

What does this do? I can't find a mention of it in the code or ui-router docs

Nick Reid This should be resolved by now.

Nick Reid This should be resolved by now.

Nick Reid I feel being brought back to the previous state is a desired behavior when successfully submitting the form, at least this is the behavior described in the ticket.

Nick Reid I feel being brought back to the previous state is a desired behavior when successfully submitting the form, at least this is the behavior described in the ticket.

Mateusz Kwiatkowski This was causing the details modal to keep changes made here(without saving them on the backend).

Mateusz Kwiatkowski This was causing the details modal to keep changes made here(without saving them on the backend).

Nikodem Graczewski I don't get why we droped vm here.... was it causing going to wrong state?

Nikodem Graczewski I don't get why we droped vm here.... was it causing going to wrong state?

I like this filter – I do wonder if its over engineering — but I do trust your judgement

I like this filter – I do wonder if its over engineering — but I do trust your judgement

I think its really weird that we need to test this in this HTML file – shouldn't it just be something that matters to the HTML? Ideally each piece of code is independent, unless the dependencies a...

I think its really weird that we need to test this in this HTML file – shouldn't it just be something that matters to the HTML?

Ideally each piece of code is independent, unless the dependencies are explicitly injected — so we should test for the submission, NOT the redirect

Maybe add a test here for two different states? I feel like we should really be putting the test in the stateTrackerService, but... its not working?

Maybe add a test here for two different states? I feel like we should really be putting the test in the stateTrackerService, but... its not working?

Nikodem Graczewski I'm testing this code in test, and am still seeing this issue – any thoughts about what's wrong?

Nikodem Graczewski I'm testing this code in test, and am still seeing this issue – any thoughts about what's wrong?

This has a side effect.... if I open the view from the inventory list, I end up at the inventory page use the StateTackerService to figure out where to go back to

This has a side effect.... if I open the view from the inventory list, I end up at the inventory page

use the StateTackerService to figure out where to go back to

LGTM

LGTM

There is some weird problem with modal template being rendered before firing $onInit method... Without this we're getting some weird errors in the console.

There is some weird problem with modal template being rendered before firing $onInit method... Without this we're getting some weird errors in the console.

OLMIS-2921: Status Update modal buttons will now take users to the correct pages
OLMIS-2921: Status Update modal buttons will now take users to the correct pages
I really like the idea of a parent state, didn't think of that. I think we can try it with facility screen I'm currently working on?

I really like the idea of a parent state, didn't think of that. I think we can try it with facility screen I'm currently working on?

Nick Reid It's for both when passing new item or editing an existing one so we don't have to download it again, cause we've most likely did that a couple of seconds ago.

Nick Reid It's for both when passing new item or editing an existing one so we don't have to download it again, cause we've most likely did that a couple of seconds ago.