Repository navigation
feat(tickets): preview a parent initiative chain in the dry run and TUI (DW-574, 3/3) - #1013
Conversation
WalkthroughParent-initiative runs now build child initiatives in separate runs, with worktree isolation and epic-boundary pauses. The CLI previews the chain, and the TUI shows child status and ticket boards. Tests cover previews, isolation checks, and resumed child runs. ChangesParent-initiative ticket chains
Priority: Estimated code review effort: 3 (Moderate) | ~25 minutes Change: Feature Sequence Diagram(s)sequenceDiagram
actor Operator
participant RunCLI
participant ParentInitiative
participant ChildRun
participant TicketTree
Operator->>RunCLI: start run for parent initiative
RunCLI->>ParentInitiative: select next eligible child
RunCLI->>ChildRun: start child run in isolated worktree
ChildRun->>TicketTree: build child tickets
ChildRun-->>RunCLI: pause at epic-boundary
Operator->>RunCLI: close epic and resume chain
RunCLI->>ParentInitiative: close drained child and select next child
RunCLI->>ChildRun: start next child run
Suggested reviewers: Merge Risk:
 |
| Check name | Status | Explanation |
|---|---|---|
| Description Check |  |
Check skipped - CodeRabbit’s high-level summary is enabled. |
| Title check | |
The title names the main change: a dry-run and TUI preview of a parent initiative chain in the tickets flow. The changeset adds chain previews to run --dry-run (cli.py, tests/test_cli.py), the TUI s… |
| Docstring Coverage | |
Docstring coverage is 91.30% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentati… |
| Linked Issues check | |
Check skipped because no linked issues were found for this pull request. |
| Out of Scope Changes check | |
Check skipped because no linked issues were found for this pull request. |
Full details: Title check
-
Explanation
The title names the main change: a dry-run and TUI preview of a parent initiative chain in the tickets flow. The changeset adds chain previews to
run --dry-run(cli.py, tests/test_cli.py), the TUI start-run preview (tui/screens/modals.py, tui/data.py), and matching docs. The ticket reference (DW-574, 3/3) and the feat(tickets) prefix are accurate and specific enough for a teammate to scan history.
Full details: Docstring Coverage
-
Explanation
Docstring coverage is 91.30% which is sufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 23 functions across 7 files. (8 skipped: 5 unsupported, 3 too large.)
Finishing Touches: 💡 1
-
🛠️ Fix failing CI checks: 💡
-
- Commit to this branch
- Create a new PR
📝 Generate docstrings
-
- Commit to this branch
- Create a new PR
🧪 Generate unit tests (beta)
-
- Commit to this branch
- Create a new PR
-
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.
A rabbit checks the children’s queue,
Then hops along the ticket trail.
At each epic, pauses to review,
Resumes when work can set sail.
The next child waits, ears alert,
While carrots mark each completed quest.
Comment @coderabbitai help to get the list of available commands.
pbean
changed the title
d017010 to
58be106
Compare
49db530 to
fa19a23
Compare
…DW-574) `run --dry-run` on a v7 parent initiative exited 1: it read the tree unpinned, which refuses a parent. It now lists the parent's children in build order (status, ticket counts, what each waits on, and a tag: next, ready, waiting, no folder, done/dropped, drained), prints the first child's tickets block through that child's folder, and ends with `predicted: chain starts with <child>; later children are chosen as each finishes` -- or, when no child can start, the end line `run` prints. --epic/--story on a parent are refused with `run`'s own text. The tickets block is extracted into `_dry_run_ticket_block`, which the one-initiative dry run prints unchanged. The dry-run banner now mirrors the tickets preflight's refusal of a parent initiative (`install.tickets_parent_problems`): before, a parent under `isolation = "none"` (or `gates.retrospective = "auto"`, or a refused child store) got no warning, though the real run refuses it. Decisions: - The dry run never writes. A real run first closes drained children (`_close_drained_children`: marks `status: done`, may commit); the dry run does not call it and does not simulate it. It tags such a child `drained (the run closes it first)`, picks as the read-only tree answers (`_chain_pick`, which only reads, now takes the already-read view), and the predicted line says the run closes those children before its first pick, which the preview does not simulate. - --max-stories caps the first child's listing as before. It is one cap across the chain, but the dry run predicts only the first child.
… through its child (DW-574) Start-run preview: `project_tickets_overview` detects a parent active initiative as `bmad-loop run` does (`install._initiative_children`; a read fault is the board's fault) and returns the chain: the parent, one `ChainChild` per child in build order (`next <parent>` pinned to the parent: status, closed/total tickets, what it waits on, a tag) and the board of the child a run builds first, read through that child's folder with the tree pinned to the parent. `TicketsBoard` grows defaulted `parent`, `children` and `chain_end` fields; a fault still carries no rows. The modal renders the children lines, then that board's counts and checkpoint flags (or why no child can start). "First" is the pure `initiativechain.next_child` with `drained` as the only pass-over probe, tagged "first ready": the CLI pick (`cli._chain_pick`) also passes over a ready child whose fresh pick dispatches nothing, which costs a `next`/`status` spawn per child, so in that corner the preview can name a different child than the run. A dim line points at `run --dry-run` for the exact chain prediction. Dashboard board: a chained child run pinned its child, but the live `[core] active_initiative` names the parent, so `tickets.py status` refused it with pin-mismatch. It now pins and reads as `bmad-loop status` does (`runs.recorded_tree_pin`'s main-checkout arm: `tree_pin` over `paths_for_state`), from the stat-gated tickets-mode paths rather than a `load_paths` per poll tick; a config that no longer names the parent is the board's fault. `initiative_board` takes `parent=` to pin the parent and read through the child folder; its cache key carries the parent. Unchained runs read as before.
…mux (DW-574) Row 15: a parent initiative with children A and B (B after A), one epic and one ticket each, store T1, `isolation = "worktree"`, retrospective `never`. `run` builds A's ticket in a unit mount and pauses at A's epic boundary; the test closes A's epic; `resume` finishes A's run, closes A (`status: done`, committed as `chore(tickets): <A> done`), and the chain launches B's run, which builds B's ticket and pauses at B's epic boundary. Asserts both run journals' sides of the switch (`initiative-chain-next`, `initiative-chain`), both plans done, each session in its unit mount, `tickets.parent_initiative` in B's status, the user's config.user.toml bytes unchanged, and a clean checkout. The fake build-auto is unchanged: its `tickets.py find` resolves the child through the mount's rewritten config. Row 16: the same parent under `isolation = "none"` exits 1 at preflight, creates no run dir, leaves the tree clean, and `validate` reports `tickets.initiative-parent`. `_scaffold` takes `children` (parent layout via `_write_parent`); new `_child`, `_run_journal`, `_runs_by_initiative`; `_status_doc` takes a run dir. The real-tmux E2E inventory counts 16 tickets rows. Ablations performed: no child in the mount config rewrite (WorktreeFlow) -> row 15's A run escalates; no isolation cause in `install._tickets_parent_findings` -> row 16's run composes.
…iew and epic pauses (DW-574)
fa19a23 to
57041cf
Compare
|
@codex review |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
Codex Review: Didn't find any major issues. Already looking forward to the next diff. Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. Codex can also answer questions or update the PR. Try commenting "@codex address that feedback". |
Part 6 of 8 of the deferred-work ledger stack (DW-574, 3 of 3).
What
Preview, TUI, E2E and docs for the parent initiative chain.
Why
Without this,
run --dry-runon a parent exits 1, the TUI start preview cannot show what a chain builds, and a chained run's board reads the wrong tree.How
run --dry-runon a parent lists its children in build order (status, ticket counts, what each waits on, a tag), prints the first child's tickets block, and ends with the predicted chain start or the end line.Testing
uv run pytest -q -n logicalat this PR's tip: 17607 passed, 124 skipped.uv run pyright(1.1.411): 0 errors.Changelog
## [Unreleased]/ Added: build a parent initiative's children one run each (DW-574).Summary by CodeRabbit