fix(list view): Introduce parent details in the list view. #2291
Conversation
nimishamukherjee
force-pushed
the
display_parent
branch
from
83c0cdb to
7c58d4f
Compare
nimishamukherjee
force-pushed
the
display_parent
branch
3 times, most recently
from
c19af81 to
0ebc850
Compare
|
@nimishamukherjee fabric8/fabric8-ui:SNAPSHOT-PR-2291-9 is deployed and available for testing at https://fabric8-ui-fabric8-ui-pr-2291-fabric8-planner.badger.fabric8.io |
|
@nimishamukherjee fabric8/fabric8-ui:SNAPSHOT-PR-2291-10 is deployed and available for testing at https://fabric8-ui-fabric8-ui-pr-2291-fabric8-planner.badger.fabric8.io |
|
@nimishamukherjee Take a look at this. |
|
@nimishamukherjee This looks like a bug |
|
@nimishamukherjee and this too |
|
The tool-tip for the workitem's type of the parent is showing like this: |
|
@nimishamukherjee Can we show the quick-preview of the immediate parent instead of the detail view, when the user clicks on the immediate parent ID? What does UX has to say about this? |
|
@nimishamukherjee We need a UX review too. |
|
One more issue. The Workitem type icon of immediate parent goes missing in the iteration context:
|
|
@nimishamukherjee This looks broken to me. I see
|
|
Any WI cannot have two parents. Its missing a check. |
|
@nimishamukherjee fabric8/fabric8-ui:SNAPSHOT-PR-2291-11 is deployed and available for testing at https://fabric8-ui-fabric8-ui-pr-2291-fabric8-planner.badger.fabric8.io |
nimishamukherjee
force-pushed
the
display_parent
branch
from
6d6cbff to
f07384a
Compare
…rk item is created.
…ck add context when work item has been de-selected.
1. Display work item type tooltip for parents 2. Handle special characters for parent title
nimishamukherjee
force-pushed
the
display_parent
branch
from
f07384a to
9b5ec30
Compare
|
@nimishamukherjee fabric8/fabric8-ui:SNAPSHOT-PR-2291-14 is deployed and available for testing at https://fabric8-ui-fabric8-ui-pr-2291-fabric8-planner.badger.fabric8.io |
@rgarg1 checked with UX on this. The parent needs to open in detailed view. |
@jarifibrahim This is a backend issue and @kwk is looking in to this. We could create an issue for the UI also, disable parent linking if the work item already has a parent. |
Based on [this discussion](https://chat.openshift.io/developers/pl/gqwxx3mj13fejfdwqqc61mrwba) we decided to make the parenting link type have a tree topology. See also fabric8-ui/fabric8-planner#2291 (comment)
|
Issues I see:
|
|
@nimishamukherjee Reviewed! |
to fabric8-services/fabric8-wit@c6fc618 As you can see there are quite a lot of changes (**83 files changed, 2914 insertions(+), 728 deletions(-)**) and the delta is quite huge. We might want to deploy more often to reduce this delta... ## Quiet CI (fabric8-services/fabric8-wit#1704) Installs RPM more quietly, builds docker containers more quietly, and pulls docker images more quietly. ## `make dev` fails to start because of container connection issue on macOS (fabric8-services/fabric8-wit#1705) remove the `network_mode` option in the containers specify the host/port for the `auth` service to connect to its `db`, using the internal Postgres port (5432) and using the container's hostname (the container name itself) same applies for the `core` service to connect to its `db` Fixes fabric8-services/fabric8-wit#1705 ## Random test failures because of work item creation error (fabric8-services/fabric8-wit#1676) Using the `Last-Modified` response header when possible Also, making sure that the list of spaces is always ordered by `updated_at DESC` to get the most recents spaces first. Move the `work item number sequence` in its own subpackage. Refactored the test to use the TestFixture to initialise the work items from the data set. Using an `UPSERT` statement to avoid creating then updating the work item number sequence, which also avoids putting a `LOCK` on the row (using the `... FOR UPDATE` option in the `SELECT` statement) Remove the `optionalKeywords` which contained the work item number but failed when the 3rd work item was retrieved because its number ('3') would match the description. This worked in the past when the number did not exist and the search was based on the ID (UUID). Added some weight on the work item number (`*A`) when the URL matches a known one and the number can be extracted. Also, add a test to search by number using the `number:%d` query. Adding a test to create work items concurrently Refactor identities creation in test: using UUIDs of created identities instead of usernames also, move a comment to `init()` where it belongs also, comment on use of `:*A` in fulltext search Also: move some log statements to `DEBUG` instead of `INFO` to reduce the noise during test executions Fixes fabric8-services/fabric8-wit#1676 ## WIP sirupsen-case for fabric8-wit (fabric8-services/fabric8-wit#1672) Goa 1.3 now seems to check length requirements in from the design. * Update golden files by running these commands: $ go test -v github.com/fabric8-services/fabric8-wit/controller -run \ TestSuiteWorkItemType -testify.m TestCreate -update $ go test -v github.com/fabric8-services/fabric8-wit/controller -run \ TestSuiteWorkItemType -testify.m TestShow -update $ go test -v github.com/fabric8-services/fabric8-wit/controller -run \ TestSuiteWorkItemType -testify.m TestValidate -update * Switch postgres to registry.centos.org/postgresql/postgresql:9.6 in CI tests * Forgot to pass F8_RESOURCE_REMOTE=1 when running remote tests WITH coverage Fixes fabric8-services/fabric8-wit#1395 ## Make link topology a type based on string (fabric8-services/fabric8-wit#1710) * Make link topology a type based on string * Rename CheckValidTopology to CheckValid and make it a member function of a Topology ## Simplify usage of test fixture customization functions (fabric8-services/fabric8-wit#1711) ## Added "not found" cases for search by number (fabric8-services/fabric8-wit#1712) * Update search_repository_whitebox_test.go Update TestSearch - By Number Add "not found" case for when searching by space and not by space. Added case with single and multiple results for when searching by number without a space. ## 'make dev' fails to start because of container connection issue on macOS (fabric8-services/fabric8-wit#1705) (fabric8-services/fabric8-wit#1713) keep the previous changes in `docker-compose.macos.yml` for now revert the changes as they break on Linux Fixes fabric8-services/fabric8-wit#1705 ## Improve the field type system and add tests (fabric8-services/fabric8-wit#1130) We now support a "bool" type for fields in a work item type. I've added a lot of tests for valid and invalid field types on both, repo- and controller-level. The field type `KindWorkitemReference` has been removed. ## Do not import pkg 'notification' in pkg 'test' (fabric8-services/fabric8-wit#1716) Move the 'test/notification.go' file into a subpackage Rename the 'NotificationChannel' into 'FakeNotificationChannel' to enforce its testing purpose. Fixes fabric8-services/fabric8-wit#1716 ## Work items disappear after unlinking (fabric8-services/fabric8-wit#1708) * Work items disappear after unlinking (fabric8-services/fabric8-wit#1708) Fix the SQL filter query by making sure only undeleted links are taken into account when the `parentexists` filter is activated. Also, reduce the number of subselects and add index on table and improve query Also, refactor the `search/search_repository_blackbox_test.go` by using sub tests. Also, make sure that tests in `TestSearchParentExists` include the `userSpaceID` in the search expression. Fixes fabric8-services/fabric8-wit#1708 ## Creator filter (fabric8-services/fabric8-wit#1715) Filter for work item creator ## Use just `state` instead of `workitemstate` (fabric8-services/fabric8-wit#1721) ## Only space collaborator and WI creators should be able to update/delete links (fabric8-services/fabric8-wit#1718) ## Enable Authorization by default (fabric8-services/fabric8-wit#1723) Previously authorization had to be enabled explicitly in dev mode. With this change, WIT talks to AUTH which handles the authorization implementation details ( like talking to keycloak ). ## Copy links to relationships for `List space` endpoint (1/2) (fabric8-services/fabric8-wit#1687) This ensure that the response will be compliant with the JSON-API specification. Fixes fabric8-services/fabric8-wit#1687 ## Save description when new iteration created (fabric8-services/fabric8-wit#1726) Fixes: https://openshift.io/openshiftio/openshiftio/plan/detail/1555 ## Delete iteration API (fabric8-services/fabric8-wit#1683) Allow the user to delete the iteration (soft-delete) using API. ## WIP: Make the parenting link type have a tree-topology (fabric8-services/fabric8-wit#1729) Based on [this discussion](https://chat.openshift.io/developers/pl/gqwxx3mj13fejfdwqqc61mrwba) we decided to make the parenting link type have a tree topology. See also fabric8-ui/fabric8-planner#2291 (comment)
@SMahil will fix this in a separate PR as we got the designs today from UX and she is on PTO tomorrow.
Create a separate issue for this as this is not related to this PR.
Please refer to our Cabal discussion - https://docs.google.com/document/d/17LdWVYXoR4CXZUsRMq5qzpicQ49Yr0nR79lMYAumYIk/edit Clicking on A or C would take the user in to exploration mode and would display B again |
|
@nimishamukherjee fabric8/fabric8-ui:SNAPSHOT-PR-2291-15 is deployed and available for testing at https://fabric8-ui-fabric8-ui-pr-2291-fabric8-planner.badger.fabric8.io |
|
@nimishamukherjee Do we already have a defect to track the indentation issue or should I log one? |
|
Work items created to track issues on this PR
|
|
@rgarg1 the behaviour has not changed. |
|
@nimishamukherjee fabric8/fabric8-ui:SNAPSHOT-PR-2291-16 is deployed and available for testing at https://fabric8-ui-fabric8-ui-pr-2291-fabric8-planner.badger.fabric8.io |
The following link types exist in the current production database but they are not known to the code and therefore we re-assign all links associated to those link types with their appropriate *known* link type and later remove these unknown link types. aad2a4ad-d601-4104-9804-2c977ca2e0c1 355b647b-adc5-46b3-b297-cc54bc0554e6 7479a9b9-8607-46fa-9535-d448fa8768ab There also exist two unknown link categories [see here](https://api.openshift.io/api/workitemlinkcategories): 04b89525-84ff-406c-8e18-d990936bdb74 6329fbb3-399b-4a78-ae3b-6a6faa8b1084 Those will be removed as well. After this PR is merged, we should only have three link types to choose from in the UI and the duplicates should be gone as well. Without the duplicate entries of `parent of` and `child of` you then can no longer link a WI to two parents. Without this PR you still can link a child WI to two parent WIs when you choose from two different link types. To double check, what link types exist, try this command: ```sh curl --silent https://api.openshift.io/api/spaces/020f756e-b51a-4b43-b113-45cec16b9ce9/workitemlinktypes \ | jq ".data[] | .id, .attributes.name" ``` It should output the id and name of each link type in the current production database: ```json "aad2a4ad-d601-4104-9804-2c977ca2e0c1" "Bug blocker" "355b647b-adc5-46b3-b297-cc54bc0554e6" "Related planner item" "7479a9b9-8607-46fa-9535-d448fa8768ab" "Parent child item" "2cea3c79-3b79-423b-90f4-1e59174c8f43" "Bug blocker" "9b631885-83b1-4abb-a340-3a9ede8493fa" "Related planner item" "25c326a7-6d03-4f5a-b23b-86a9ee4171e9" "Parent child item" ``` In 90c595e I have introduced fixed IDs for link types and link categories and that must have been the time when the old ones weren't deleted. As part of this PR I've replaced many of the `panic` calls within the migration tests with normal test failures. I've also wrapped errors and added filenames of executed SQL files to the error output to increase *debuggability* (is that a word?). This relates to #1729 and fabric8-ui/fabric8-planner#2291 (comment)
|
@nimishamukherjee and @rgarg1 FYI we should no longer have multiple parent links in production and we should no longer be able to produce one of them. |
The following link types exist in the current production database but they are not known to the code and therefore we re-assign all links associated to those link types with their appropriate *known* link type and later remove these unknown link types. aad2a4ad-d601-4104-9804-2c977ca2e0c1 355b647b-adc5-46b3-b297-cc54bc0554e6 7479a9b9-8607-46fa-9535-d448fa8768ab There also exist two unknown link categories [see here](https://api.openshift.io/api/workitemlinkcategories): 04b89525-84ff-406c-8e18-d990936bdb74 6329fbb3-399b-4a78-ae3b-6a6faa8b1084 Those will be removed as well. After this PR is merged, we should only have three link types to choose from in the UI and the duplicates should be gone as well. Without the duplicate entries of `parent of` and `child of` you then can no longer link a WI to two parents. Without this PR you still can link a child WI to two parent WIs when you choose from two different link types. To double check, what link types exist, try this command: ```sh curl --silent https://api.openshift.io/api/spaces/020f756e-b51a-4b43-b113-45cec16b9ce9/workitemlinktypes \ | jq ".data[] | .id, .attributes.name" ``` It should output the id and name of each link type in the current production database: ```json "aad2a4ad-d601-4104-9804-2c977ca2e0c1" "Bug blocker" "355b647b-adc5-46b3-b297-cc54bc0554e6" "Related planner item" "7479a9b9-8607-46fa-9535-d448fa8768ab" "Parent child item" "2cea3c79-3b79-423b-90f4-1e59174c8f43" "Bug blocker" "9b631885-83b1-4abb-a340-3a9ede8493fa" "Related planner item" "25c326a7-6d03-4f5a-b23b-86a9ee4171e9" "Parent child item" ``` In 90c595ea I have introduced fixed IDs for link types and link categories and that must have been the time when the old ones weren't deleted. As part of this PR I've replaced many of the `panic` calls within the migration tests with normal test failures. I've also wrapped errors and added filenames of executed SQL files to the error output to increase *debuggability* (is that a word?). This relates to #1729 and fabric8-ui/fabric8-planner#2291 (comment)
Introduce parent information on the list view.
Visual design - https://redhat.invisionapp.com/share/YZDOM6S2R#/screens
Visual design with specs - https://redhat.invisionapp.com/share/YZDOM6S2R#/screens/255484840
Visual design for indentation - https://redhat.invisionapp.com/share/XQDLTAHCY#/screens/260232204