15s was too short for 25MB uploads on slow connections (~256KB/s takes
~100s). Both timeouts are raised to 120s to accommodate MAX_UPLOAD_SIZE_MB
at worst-case bandwidth with headroom.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
- Expand stubbedFileStorer with deletedKey tracking and deleteErr injection field
- Implement TestFileDownload (FILE-04): 302 redirect to presigned URL
- Implement TestFileDownload_NonOwner: non-owner gets 404
- Implement TestFileDelete (FILE-05): HTMX delete, S3+DB both deleted
- Implement TestFileDelete_S3Failure: S3 error does not abort DB delete, 200 returned
- Implement TestFileOwnership (FILE-06): non-owner gets 404 on all three routes
All 6 in-scope findings (CR-01, CR-02, WR-01, WR-02, WR-03, WR-04)
have been applied and verified. Updated status from issues_found to fixed.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
maxPos + 100 could silently overflow to a negative value when maxPos
approached MaxInt32. Added a maxAllowedPosition guard that returns a
validation error before the InsertTask call if the column position space
is exhausted.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
A description of spaces-only was being stored as a valid non-null DB value
because the empty-string check ran before trimming. Now consistent with how
other nullable text fields are handled.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
Without these headers, HTMX used the form's own hx-target="#column-{status}"
+ hx-swap="beforeend", appending the error form into the task column and
destroying all visible task cards. The error form now lands back in the
add-task slot where it belongs.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
Both the single-task branch and the main loop were using _, _ = to
discard UpdateTask errors. Now both log the error and return 500 so
the client is never shown a false success when DB writes fail.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
The raw fmt.Fprintf bypassed templ's auto-escaping pipeline and was
inconsistent with every other handler. Added TaskCardGone(taskID uuid.UUID)
to tasks.templ and updated TaskDeleteHandler to use it. Ran just generate.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
Both handlers were missing the mandatory ParseForm call before reading
PostFormValue. This caused gorilla/csrf (which reads the body for CSRF
token validation) to consume the body, leaving PostFormValue to return
empty strings. TaskReorderHandler was used as the correct reference.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
All 7 TASK requirements verified in browser. Two bugs found and fixed
during verification (badge count, drag-and-drop draggable selector).
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
Sortable.js draggable must match direct children of .sortable-column.
Using .task-card (grandchild) caused Sortable to detach it from its
.task-card-zone wrapper, breaking HTMX OOB swap targets and making
drag appear to do nothing. Changed to .task-card-zone so the full
wrapper moves, keeping id= attributes intact for HTMX round-trips.
Also removed redundant form.dispatchEvent() before htmx.trigger()
which could cause a double submit on reorder.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
Replace htmx.onLoad (requires htmx at parse time) with native
document.addEventListener('DOMContentLoaded') + 'htmx:afterSettle'
so Sortable.js is guaranteed loaded before init runs.
Add task-count-badge-{status} wrapper IDs and updateBadges() that
recounts .task-card elements on every HTMX settle so badge counts
stay in sync after create, delete, and reorder operations.
Co-Authored-By: Claude Sonnet 4.6 (1M context) <noreply@anthropic.com>
- 04-03-SUMMARY.md created with full execution details
- STATE.md updated: decisions, metrics, notes for plan 03
- ROADMAP.md updated with plan progress (3/4 summaries for phase 4)
- Remove t.Skip("handlers_tasks not yet implemented") from TestTaskOrderPersists
- Full test suite green: go test ./... exits 0, no FAIL lines
- All 9 TestTask* tests active (skip on missing TEST_DATABASE_URL per existing pattern)
- TaskReorderHandler: POST /tablos/{id}/tasks/reorder updates status+position
- Fetches existing task via GetTaskByID before UpdateTask (T-04-08 mass assignment guard)
- Supports both array form (task_id[]/task_col[]) and single-value form (task_id/status/position)
- Invalid UUIDs silently skipped (D-05); tasks from other tablos skipped (T-04-10)
- Returns updated KanbanBoard outerHTML for HTMX swap
- Remove t.Skip from TestTaskReorderCrossColumn and TestTaskReorderSameColumn
- Add handlers_tasks.go: TasksDeps, TaskNewFormHandler, TaskCancelNewHandler, TaskCreateHandler, TaskShowHandler, TaskDeleteConfirmHandler, TaskDeleteHandler, plus stub Edit/Update/Reorder handlers
- Add task routes to router.go (static before parametric per Pitfall 1)
- Add TasksDeps param to NewRouter; update main.go and all test callers
- Move TaskColumns/TaskColumnLabels to templates package to avoid import cycle
