Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

workflow UI #99

Merged
merged 3 commits into from
Jul 25, 2024
Merged
Show file tree
Hide file tree
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
tons of improvements to the workflow UI
  • Loading branch information
bgentry committed Jul 25, 2024
commit 663ac755f46bde27cabae97903ce4e66ec0f99b9
159 changes: 158 additions & 1 deletion api_handler.go
Original file line number Diff line number Diff line change
Expand Up @@ -603,6 +603,82 @@ func (a *workflowGetEndpoint) Execute(ctx context.Context, req *workflowGetReque
}, nil
}

//
// workflowListEndpoint
//

type workflowListEndpoint struct {
apiBundle
apiendpoint.Endpoint[jobCancelRequest, listResponse[workflowListItem]]
}

func (*workflowListEndpoint) Meta() *apiendpoint.EndpointMeta {
return &apiendpoint.EndpointMeta{
Pattern: "GET /api/workflows",
StatusCode: http.StatusOK,
}
}

type workflowListRequest struct {
After *string `json:"-" validate:"omitempty"` // from ExtractRaw
Limit *int `json:"-" validate:"omitempty,min=0,max=1000"` // from ExtractRaw
State string `json:"-" validate:"omitempty,oneof=active inactive"` // from ExtractRaw
}

func (req *workflowListRequest) ExtractRaw(r *http.Request) error {
if after := r.URL.Query().Get("after"); after != "" {
req.After = (&after)
}

if limitStr := r.URL.Query().Get("limit"); limitStr != "" {
limit, err := strconv.Atoi(limitStr)
if err != nil {
return apierror.NewBadRequest("Couldn't convert `limit` to integer: %s.", err)
}

req.Limit = &limit
}

if state := r.URL.Query().Get("state"); state != "" {
req.State = (state)
}

return nil
}

func (a *workflowListEndpoint) Execute(ctx context.Context, req *workflowListRequest) (*listResponse[workflowListItem], error) {
switch req.State {
case "active":
workflows, err := dbsqlc.New().WorkflowListActive(ctx, a.dbPool, &dbsqlc.WorkflowListActiveParams{
After: ptrutil.ValOrDefault(req.After, ""),
PaginationLimit: int32(ptrutil.ValOrDefault(req.Limit, 100)),
})
if err != nil {
return nil, fmt.Errorf("error listing workflows: %w", err)
}

return listResponseFrom(sliceutil.Map(workflows, internalWorkflowListActiveToSerializableWorkflow)), nil
case "inactive":
workflows, err := dbsqlc.New().WorkflowListInactive(ctx, a.dbPool, &dbsqlc.WorkflowListInactiveParams{
After: ptrutil.ValOrDefault(req.After, ""),
PaginationLimit: int32(ptrutil.ValOrDefault(req.Limit, 100)),
})
if err != nil {
return nil, fmt.Errorf("error listing workflows: %w", err)
}
return listResponseFrom(sliceutil.Map(workflows, internalWorkflowListInactiveToSerializableWorkflow)), nil
default:
workflows, err := dbsqlc.New().WorkflowListAll(ctx, a.dbPool, &dbsqlc.WorkflowListAllParams{
After: ptrutil.ValOrDefault(req.After, ""),
PaginationLimit: int32(ptrutil.ValOrDefault(req.Limit, 100)),
})
if err != nil {
return nil, fmt.Errorf("error listing workflows: %w", err)
}
return listResponseFrom(sliceutil.Map(workflows, internalWorkflowListAllToSerializableWorkflow)), nil
}
}

type RiverJob struct {
ID int64 `json:"id"`
Args json.RawMessage `json:"args"`
Expand Down Expand Up @@ -642,7 +718,7 @@ func internalJobToSerializableJob(internal *dbsqlc.RiverJob) *RiverJob {
Attempt: int(internal.Attempt),
AttemptedAt: timePtr(internal.AttemptedAt),
AttemptedBy: attemptedBy,
CreatedAt: internal.CreatedAt.Time,
CreatedAt: internal.CreatedAt.Time.UTC(),
Errors: errs,
FinalizedAt: timePtr(internal.FinalizedAt),
Kind: internal.Kind,
Expand Down Expand Up @@ -721,6 +797,87 @@ func riverQueuesToSerializableQueues(internal []*rivertype.Queue, counts []*dbsq
return listResponseFrom(queues)
}

type workflowListItem struct {
CountAvailable int `json:"count_available"`
CountCancelled int `json:"count_cancelled"`
CountCompleted int `json:"count_completed"`
CountDiscarded int `json:"count_discarded"`
CountFailedDeps int `json:"count_failed_deps"`
CountPending int `json:"count_pending"`
CountRetryable int `json:"count_retryable"`
CountRunning int `json:"count_running"`
CountScheduled int `json:"count_scheduled"`
CreatedAt time.Time `json:"created_at"`
ID string `json:"id"`
Name *string `json:"name"`
}

func internalWorkflowListActiveToSerializableWorkflow(internal *dbsqlc.WorkflowListActiveRow) *workflowListItem {
var name *string
if internal.WorkflowName != "" {
name = &internal.WorkflowName
}

return &workflowListItem{
CountAvailable: int(internal.CountAvailable),
CountCancelled: int(internal.CountCancelled),
CountCompleted: int(internal.CountCompleted),
CountDiscarded: int(internal.CountDiscarded),
CountFailedDeps: int(internal.CountFailedDeps),
CountPending: int(internal.CountPending),
CountRetryable: int(internal.CountRetryable),
CountRunning: int(internal.CountRunning),
CountScheduled: int(internal.CountScheduled),
CreatedAt: internal.EarliestCreatedAt.Time.UTC(),
ID: internal.WorkflowID,
Name: name,
}
}

func internalWorkflowListAllToSerializableWorkflow(internal *dbsqlc.WorkflowListAllRow) *workflowListItem {
var name *string
if internal.WorkflowName != "" {
name = &internal.WorkflowName
}

return &workflowListItem{
CountAvailable: int(internal.CountAvailable),
CountCancelled: int(internal.CountCancelled),
CountCompleted: int(internal.CountCompleted),
CountDiscarded: int(internal.CountDiscarded),
CountFailedDeps: int(internal.CountFailedDeps),
CountPending: int(internal.CountPending),
CountRetryable: int(internal.CountRetryable),
CountRunning: int(internal.CountRunning),
CountScheduled: int(internal.CountScheduled),
CreatedAt: internal.EarliestCreatedAt.Time.UTC(),
ID: internal.WorkflowID,
Name: name,
}
}

func internalWorkflowListInactiveToSerializableWorkflow(internal *dbsqlc.WorkflowListInactiveRow) *workflowListItem {
var name *string
if internal.WorkflowName != "" {
name = &internal.WorkflowName
}

return &workflowListItem{
CountAvailable: int(internal.CountAvailable),
CountCancelled: int(internal.CountCancelled),
CountCompleted: int(internal.CountCompleted),
CountDiscarded: int(internal.CountDiscarded),
CountFailedDeps: int(internal.CountFailedDeps),
CountPending: int(internal.CountPending),
CountRetryable: int(internal.CountRetryable),
CountRunning: int(internal.CountRunning),
CountScheduled: int(internal.CountScheduled),
CreatedAt: internal.EarliestCreatedAt.Time.UTC(),
ID: internal.WorkflowID,
Name: name,
}
}

func timePtr(t pgtype.Timestamptz) *time.Time {
if !t.Valid {
return nil
Expand Down
97 changes: 97 additions & 0 deletions api_handler_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -531,3 +531,100 @@ func TestAPIHandlerWorkflowGet(t *testing.T) {
requireAPIError(t, apierror.NewNotFoundWorkflow(workflowID.String()), err)
})
}

func TestAPIHandlerWorkflowList(t *testing.T) {
t.Parallel()

ctx := context.Background()

t.Run("Success", func(t *testing.T) {
t.Parallel()

endpoint, bundle := setupEndpoint[workflowListEndpoint](ctx, t)

job1 := testfactory.Job(ctx, t, bundle.exec, &testfactory.JobOpts{
Metadata: []byte(`{"workflow_id":"1", "workflow_name":"first_wf", "task":"a"}`),
State: ptrutil.Ptr(rivertype.JobStatePending),
})
job2 := testfactory.Job(ctx, t, bundle.exec, &testfactory.JobOpts{
FinalizedAt: ptrutil.Ptr(time.Now()),
Metadata: []byte(`{"workflow_id":"2", "task":"b"}`),
State: ptrutil.Ptr(rivertype.JobStateCompleted),
})
_ = testfactory.Job(ctx, t, bundle.exec, &testfactory.JobOpts{
FinalizedAt: ptrutil.Ptr(time.Now()),
Metadata: []byte(`{"workflow_id":"2", "task":"c", "workflow_deps_failed_at":"2024-01-01T00:00:00Z"}`),
State: ptrutil.Ptr(rivertype.JobStateCancelled),
})

t.Run("All", func(t *testing.T) {
resp, err := endpoint.Execute(ctx, &workflowListRequest{})
require.NoError(t, err)
require.Len(t, resp.Data, 2)
require.Equal(t, 1, resp.Data[0].CountCancelled)
require.Equal(t, 1, resp.Data[0].CountCompleted)
t.Logf("resp0: %+v", resp.Data[0])
require.Equal(t, 1, resp.Data[0].CountFailedDeps)
require.Nil(t, resp.Data[0].Name)

require.Equal(t, 0, resp.Data[1].CountAvailable)
require.Equal(t, 0, resp.Data[1].CountCancelled)
require.Equal(t, 0, resp.Data[1].CountCompleted)
require.Equal(t, 0, resp.Data[1].CountDiscarded)
require.Equal(t, 0, resp.Data[1].CountFailedDeps)
require.Equal(t, 1, resp.Data[1].CountPending)
require.Equal(t, 0, resp.Data[1].CountRetryable)
require.Equal(t, 0, resp.Data[1].CountRunning)
require.Equal(t, 0, resp.Data[1].CountScheduled)
require.Equal(t, "first_wf", *resp.Data[1].Name)
})

t.Run("Active", func(t *testing.T) {
resp, err := endpoint.Execute(ctx, &workflowListRequest{State: "active"})
require.NoError(t, err)
require.Len(t, resp.Data, 1)
require.Equal(t, 0, resp.Data[0].CountAvailable)
require.Equal(t, 0, resp.Data[0].CountCancelled)
require.Equal(t, 0, resp.Data[0].CountCompleted)
require.Equal(t, 0, resp.Data[0].CountDiscarded)
require.Equal(t, 0, resp.Data[0].CountFailedDeps)
require.Equal(t, 1, resp.Data[0].CountPending)
require.Equal(t, 0, resp.Data[0].CountRetryable)
require.Equal(t, 0, resp.Data[0].CountRunning)
require.Equal(t, 0, resp.Data[0].CountScheduled)
require.Equal(t, "first_wf", *resp.Data[0].Name)
require.Equal(t, job1.CreatedAt.UTC(), resp.Data[0].CreatedAt)
})

t.Run("Inactive", func(t *testing.T) {
resp, err := endpoint.Execute(ctx, &workflowListRequest{State: "inactive"})
require.NoError(t, err)
require.Len(t, resp.Data, 1)
require.Equal(t, 1, resp.Data[0].CountCompleted)
require.Equal(t, 1, resp.Data[0].CountFailedDeps)
require.Nil(t, resp.Data[0].Name)
require.Equal(t, job2.CreatedAt.UTC(), resp.Data[0].CreatedAt)
})
})

t.Run("Limit", func(t *testing.T) {
t.Parallel()

endpoint, bundle := setupEndpoint[workflowListEndpoint](ctx, t)

_ = testfactory.Job(ctx, t, bundle.exec, &testfactory.JobOpts{
Metadata: []byte(`{"workflow_id":"1", "workflow_name":"first_wf", "task":"a"}`),
State: ptrutil.Ptr(rivertype.JobStatePending),
})
_ = testfactory.Job(ctx, t, bundle.exec, &testfactory.JobOpts{
Metadata: []byte(`{"workflow_id":"2", "task":"b"}`),
ScheduledAt: ptrutil.Ptr(time.Now().Add(time.Hour)),
State: ptrutil.Ptr(rivertype.JobStateScheduled),
})

resp, err := endpoint.Execute(ctx, &workflowListRequest{Limit: ptrutil.Ptr(1)})
require.NoError(t, err)
require.Len(t, resp.Data, 1)
require.Equal(t, "2", resp.Data[0].ID) // DESC order means last one gets returned
})
}
1 change: 1 addition & 0 deletions handler.go
Original file line number Diff line number Diff line change
Expand Up @@ -100,6 +100,7 @@ func NewHandler(opts *HandlerOpts) (http.Handler, error) {
apiendpoint.Mount(mux, opts.Logger, &queueResumeEndpoint{apiBundle: apiBundle})
apiendpoint.Mount(mux, opts.Logger, &stateAndCountGetEndpoint{apiBundle: apiBundle})
apiendpoint.Mount(mux, opts.Logger, &workflowGetEndpoint{apiBundle: apiBundle})
apiendpoint.Mount(mux, opts.Logger, &workflowListEndpoint{apiBundle: apiBundle})
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you want to add an "integration level" (end-to-end HTTP stack) test for the new endpoint in handler_test.go? Should be close to just a one liner.


if err := mountStaticFiles(opts.Logger, mux); err != nil {
return nil, err
Expand Down
4 changes: 3 additions & 1 deletion internal/dbsqlc/sqlc.yaml
Original file line number Diff line number Diff line change
@@ -1,7 +1,9 @@
version: "2"
sql:
- engine: "postgresql"
queries: "query.sql"
queries:
- "query.sql"
- "workflow.sql"
schema: "migrations"
gen:
go:
Expand Down
85 changes: 85 additions & 0 deletions internal/dbsqlc/workflow.sql
Original file line number Diff line number Diff line change
@@ -0,0 +1,85 @@
-- name: WorkflowListActive :many
WITH workflow_ids AS (
SELECT DISTINCT ON (workflow_id) metadata->>'workflow_id' AS workflow_id
FROM river_job
WHERE state IN ('available', 'pending', 'retryable', 'running', 'scheduled')
AND metadata ? 'workflow_id'
AND metadata->>'workflow_id' > @after::text
ORDER BY workflow_id DESC
LIMIT @pagination_limit::integer
)

SELECT
(river_job.metadata->>'workflow_id')::text AS workflow_id,
coalesce(river_job.metadata->>'workflow_name', '')::text AS workflow_name,
MIN(river_job.created_at)::timestamptz AS earliest_created_at,
COUNT(*) FILTER (WHERE river_job.metadata ? 'workflow_deps_failed_at') AS count_failed_deps,
COUNT(*) FILTER (WHERE river_job.state = 'available') AS count_available,
COUNT(*) FILTER (WHERE river_job.state = 'cancelled') AS count_cancelled,
COUNT(*) FILTER (WHERE river_job.state = 'completed') AS count_completed,
COUNT(*) FILTER (WHERE river_job.state = 'discarded') AS count_discarded,
COUNT(*) FILTER (WHERE river_job.state = 'pending') AS count_pending,
COUNT(*) FILTER (WHERE river_job.state = 'retryable') AS count_retryable,
COUNT(*) FILTER (WHERE river_job.state = 'running') AS count_running,
COUNT(*) FILTER (WHERE river_job.state = 'scheduled') AS count_scheduled
FROM river_job
INNER JOIN workflow_ids
ON river_job.metadata->>'workflow_id' = workflow_ids.workflow_id
WHERE
river_job.metadata ? 'workflow_id'
GROUP BY river_job.metadata->>'workflow_id', metadata->>'workflow_name'
ORDER BY river_job.metadata->>'workflow_id' DESC;

-- name: WorkflowListAll :many
SELECT
(river_job.metadata->>'workflow_id')::text AS workflow_id,
coalesce(river_job.metadata->>'workflow_name', '')::text AS workflow_name,
MIN(river_job.created_at)::timestamptz AS earliest_created_at,
COUNT(*) FILTER (WHERE river_job.metadata ? 'workflow_deps_failed_at') AS count_failed_deps,
COUNT(*) FILTER (WHERE river_job.state = 'available') AS count_available,
COUNT(*) FILTER (WHERE river_job.state = 'cancelled') AS count_cancelled,
COUNT(*) FILTER (WHERE river_job.state = 'completed') AS count_completed,
COUNT(*) FILTER (WHERE river_job.state = 'discarded') AS count_discarded,
COUNT(*) FILTER (WHERE river_job.state = 'pending') AS count_pending,
COUNT(*) FILTER (WHERE river_job.state = 'retryable') AS count_retryable,
COUNT(*) FILTER (WHERE river_job.state = 'running') AS count_running,
COUNT(*) FILTER (WHERE river_job.state = 'scheduled') AS count_scheduled
FROM river_job
WHERE
river_job.metadata ? 'workflow_id'
AND river_job.metadata->>'workflow_id' > @after::text
GROUP BY river_job.metadata->>'workflow_id', metadata->>'workflow_name'
ORDER BY river_job.metadata->>'workflow_id' DESC
LIMIT @pagination_limit::integer;

-- name: WorkflowListInactive :many
WITH active_workflow_ids AS (
SELECT DISTINCT ON (workflow_id) metadata->>'workflow_id' AS workflow_id
FROM river_job
WHERE state IN ('available', 'pending', 'retryable', 'running', 'scheduled')
AND metadata->>'workflow_id' > @after::text
AND metadata ? 'workflow_id'
ORDER BY workflow_id DESC
)

SELECT
(river_job.metadata->>'workflow_id')::text AS workflow_id,
coalesce(river_job.metadata->>'workflow_name', '')::text AS workflow_name,
MIN(river_job.created_at)::timestamptz AS earliest_created_at,
COUNT(*) FILTER (WHERE river_job.metadata ? 'workflow_deps_failed_at') AS count_failed_deps,
COUNT(*) FILTER (WHERE river_job.state = 'available') AS count_available,
COUNT(*) FILTER (WHERE river_job.state = 'cancelled') AS count_cancelled,
COUNT(*) FILTER (WHERE river_job.state = 'completed') AS count_completed,
COUNT(*) FILTER (WHERE river_job.state = 'discarded') AS count_discarded,
COUNT(*) FILTER (WHERE river_job.state = 'pending') AS count_pending,
COUNT(*) FILTER (WHERE river_job.state = 'retryable') AS count_retryable,
COUNT(*) FILTER (WHERE river_job.state = 'running') AS count_running,
COUNT(*) FILTER (WHERE river_job.state = 'scheduled') AS count_scheduled
FROM river_job
WHERE
state IN ('completed', 'cancelled', 'discarded')
AND metadata ? 'workflow_id'
AND metadata ->> 'workflow_id' NOT IN (SELECT workflow_id FROM active_workflow_ids)
GROUP BY metadata->>'workflow_id', metadata->>'workflow_name'
ORDER BY workflow_id DESC
LIMIT @pagination_limit::integer;
Loading