From 5c2df3df3b42e5f51f81c3d8f5b44417bac54041 Mon Sep 17 00:00:00 2001 From: Taylor Eernisse Date: Fri, 13 Feb 2026 10:53:33 -0500 Subject: [PATCH] chore(beads): sync issue tracker Export latest bead state to JSONL. Co-Authored-By: Claude Opus 4.6 --- .beads/issues.jsonl | 3 ++- .beads/last-touched | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/.beads/issues.jsonl b/.beads/issues.jsonl index 6afe48b..c2ae9ea 100644 --- a/.beads/issues.jsonl +++ b/.beads/issues.jsonl @@ -18,6 +18,7 @@ {"id":"bd-18qs","title":"Implement entity table + filter bar widgets","description":"## Background\nThe entity table and filter bar are shared widgets used by Issue List, MR List, and potentially Search results. The entity table supports sortable columns with responsive width allocation. The filter bar provides a typed DSL for filtering with inline diagnostics.\n\n## Approach\nEntity Table (view/common/entity_table.rs):\n- EntityTable widget: generic over row type\n- TableRow trait: fn cells(&self) -> Vec, fn sort_key(&self, col: usize) -> Ordering\n- Column definitions: name, min_width, flex_weight, alignment, sort_field\n- Responsive column fitting: hide low-priority columns as terminal narrows\n- Keyboard: j/k scroll, J/K page scroll, Tab cycle sort column, Enter select, g+g top, G bottom\n- Visual: alternating row colors, selected row highlight, sort indicator arrow\n\nFilter Bar (view/common/filter_bar.rs):\n- FilterBar widget wrapping ftui TextInput\n- DSL parsing (crate filter_dsl.rs): quoted values (\"in progress\"), negation prefix (-closed), field:value syntax (author:taylor, state:opened, label:bug), free-text search\n- Inline diagnostics: unknown field names highlighted, cursor position for error\n- Applied filter chips shown as tags below the input\n\nFilter DSL (filter_dsl.rs):\n- parse_filter_tokens(input: &str) -> Vec\n- FilterToken enum: FieldValue{field, value}, Negation{field, value}, FreeText(String), QuotedValue(String)\n- Validation: known fields per entity type (issues: state, author, assignee, label, milestone, status; MRs: state, author, reviewer, target_branch, source_branch, label, draft)\n\n## Acceptance Criteria\n- [ ] EntityTable renders with responsive column widths\n- [ ] Columns hide gracefully when terminal is too narrow\n- [ ] j/k scrolls, Enter selects, Tab cycles sort column\n- [ ] Sort indicator (arrow) shows on active sort column\n- [ ] FilterBar captures text input and parses DSL tokens\n- [ ] Quoted values preserved as single token\n- [ ] Negation prefix (-closed) creates exclusion filter\n- [ ] field:value syntax maps to typed filter fields\n- [ ] Unknown field names highlighted as error\n- [ ] Filter chips rendered below input bar\n\n## Files\n- CREATE: crates/lore-tui/src/view/common/entity_table.rs\n- CREATE: crates/lore-tui/src/view/common/filter_bar.rs\n- CREATE: crates/lore-tui/src/filter_dsl.rs\n\n## TDD Anchor\nRED: Write test_parse_filter_basic in filter_dsl.rs that parses \"state:opened author:taylor\" and asserts two FieldValue tokens.\nGREEN: Implement parse_filter_tokens with field:value splitting.\nVERIFY: cargo test --manifest-path crates/lore-tui/Cargo.toml test_parse_filter\n\nAdditional tests:\n- test_parse_quoted_value: \"in progress\" -> single QuotedValue token\n- test_parse_negation: -closed -> Negation token\n- test_parse_mixed: state:opened \"bug fix\" -wontfix -> 3 tokens of correct types\n- test_column_hiding: EntityTable with 5 columns hides lowest priority at 60 cols\n\n## Edge Cases\n- Filter DSL must handle Unicode in values (CJK issue titles)\n- Empty filter string should show all results (no-op)\n- Very long filter strings must not overflow the input area\n- Tab cycling sort must wrap around (last column -> first)\n- Column widths must respect min_width even when terminal is very narrow","status":"open","priority":2,"issue_type":"task","created_at":"2026-02-12T16:58:07.586225Z","created_by":"tayloreernisse","updated_at":"2026-02-12T18:11:28.085981Z","compaction_level":0,"original_size":0,"labels":["TUI"],"dependencies":[{"issue_id":"bd-18qs","depends_on_id":"bd-1cl9","type":"blocks","created_at":"2026-02-12T18:11:28.085954Z","created_by":"tayloreernisse"},{"issue_id":"bd-18qs","depends_on_id":"bd-6pmy","type":"blocks","created_at":"2026-02-12T17:09:48.569569Z","created_by":"tayloreernisse"}]} {"id":"bd-18t","title":"Implement discussion truncation logic","description":"## Background\nDiscussion threads can contain dozens of notes spanning thousands of characters. The truncation module ensures discussion documents stay within a 32k character limit (suitable for embedding chunking) by dropping middle notes while preserving first and last notes for context. A separate hard safety cap of 2MB applies to ALL document types for pathological content (pasted logs, base64 blobs). Issue/MR documents are NOT truncated by the discussion logic — only the hard cap applies.\n\n## Approach\nCreate `src/documents/truncation.rs` per PRD Section 2.3:\n\n```rust\npub const MAX_DISCUSSION_CHARS: usize = 32_000;\npub const MAX_DOCUMENT_CHARS_HARD: usize = 2_000_000;\n\npub struct NoteContent {\n pub author: String,\n pub date: String,\n pub body: String,\n}\n\npub struct TruncationResult {\n pub content: String,\n pub is_truncated: bool,\n pub reason: Option,\n}\n\npub enum TruncationReason {\n TokenLimitMiddleDrop,\n SingleNoteOversized,\n FirstLastOversized,\n HardCapOversized,\n}\n```\n\n**Core functions:**\n- `truncate_discussion(notes: &[NoteContent], max_chars: usize) -> TruncationResult`\n- `truncate_utf8(s: &str, max_bytes: usize) -> &str` (shared with fts.rs)\n- `truncate_hard_cap(content: &str) -> TruncationResult` (for any doc type)\n\n**Algorithm for truncate_discussion:**\n1. Format all notes as `@author (date):\\nbody\\n\\n`\n2. If total <= max_chars: return as-is\n3. If single note: truncate at UTF-8 boundary, append `[truncated]`, reason = SingleNoteOversized\n4. Binary search: find max N where first N notes + last 1 note + marker fit within max_chars\n5. If first + last > max_chars: keep only first (truncated), reason = FirstLastOversized\n6. Otherwise: first N + marker + last M, reason = TokenLimitMiddleDrop\n\n**Marker format:** `\\n\\n[... N notes omitted for length ...]\\n\\n`\n\n## Acceptance Criteria\n- [ ] Discussion with total < 32k chars returns untruncated\n- [ ] Discussion > 32k chars: middle notes dropped, first + last preserved\n- [ ] Truncation marker shows correct count of omitted notes\n- [ ] Single note > 32k chars: truncated at UTF-8-safe boundary with `[truncated]` appended\n- [ ] First + last note > 32k: only first note kept (truncated if needed)\n- [ ] Hard cap (2MB) truncates any document type at UTF-8-safe boundary\n- [ ] `truncate_utf8` never panics on multi-byte codepoints (emoji, CJK, accented chars)\n- [ ] `TruncationReason::as_str()` returns DB-compatible strings matching CHECK constraint\n\n## Files\n- `src/documents/truncation.rs` — new file\n- `src/documents/mod.rs` — add `pub use truncation::{truncate_discussion, truncate_hard_cap, TruncationResult, NoteContent};`\n\n## TDD Loop\nRED: Tests in `#[cfg(test)] mod tests`:\n- `test_no_truncation_under_limit` — 3 short notes, all fit\n- `test_middle_notes_dropped` — 10 notes totaling > 32k, first+last preserved\n- `test_single_note_oversized` — one note of 50k chars, truncated safely\n- `test_first_last_oversized` — first=20k, last=20k, only first kept\n- `test_one_note_total` — single note under limit: no truncation\n- `test_utf8_boundary_safety` — content with emoji/CJK at truncation point\n- `test_hard_cap` — 3MB content truncated to 2MB\n- `test_marker_count_correct` — marker says \"[... 5 notes omitted ...]\" when 5 dropped\nGREEN: Implement truncation logic\nVERIFY: `cargo test truncation`\n\n## Edge Cases\n- Empty notes list: return empty content, not truncated\n- All notes are empty strings: total = 0, no truncation\n- Note body contains only multi-byte characters: truncate_utf8 walks backward to find safe boundary\n- Note body with trailing newlines: formatted output should not have excessive blank lines","status":"closed","priority":3,"issue_type":"task","created_at":"2026-01-30T15:25:45.597167Z","created_by":"tayloreernisse","updated_at":"2026-01-30T17:21:32.256569Z","closed_at":"2026-01-30T17:21:32.256507Z","close_reason":"Completed: truncate_discussion, truncate_hard_cap, truncate_utf8, TruncationReason with as_str(), 12 tests pass","compaction_level":0,"original_size":0,"dependencies":[{"issue_id":"bd-18t","depends_on_id":"bd-36p","type":"blocks","created_at":"2026-01-30T15:29:15.947679Z","created_by":"tayloreernisse"}]} {"id":"bd-18yh","title":"NOTE-2C: Note document extractor function","description":"## Background\nEach non-system note becomes a searchable document in the FTS/embedding pipeline. Follows the pattern of extract_issue_document() (line 85), extract_mr_document() (line 186), extract_discussion_document() (line 302) in src/documents/extractor.rs.\n\n## Approach\nAdd pub fn extract_note_document(conn: &Connection, note_id: i64) -> Result> to src/documents/extractor.rs:\n\n1. Fetch note with JOIN to discussions and projects:\n SELECT n.id, n.gitlab_id, n.author_username, n.body, n.note_type, n.is_system, n.created_at, n.updated_at, n.position_new_path, n.position_new_line, n.position_old_path, n.position_old_line, n.resolvable, n.resolved, n.resolved_by, d.noteable_type, d.issue_id, d.merge_request_id, p.path_with_namespace, p.id as project_id\n FROM notes n\n JOIN discussions d ON n.discussion_id = d.id\n JOIN projects p ON n.project_id = p.id\n WHERE n.id = ?\n\n2. Return None for: system notes (is_system = 1), not found, orphaned discussions (no parent issue/MR)\n\n3. Fetch parent entity (Issue or MR) — get iid, title, web_url, labels:\n For issues: SELECT iid, title, web_url FROM issues WHERE id = ?\n For MRs: SELECT iid, title, web_url FROM merge_requests WHERE id = ?\n Labels: SELECT label_name FROM issue_labels/mr_labels WHERE issue_id/mr_id = ?\n (Same pattern as extract_discussion_document lines 332-401)\n\n4. Build paths: BTreeSet from position_old_path + position_new_path (filter None values)\n\n5. Build URL: parent_web_url + \"#note_{gitlab_id}\"\n\n6. Format content with structured key-value header:\n [[Note]]\n source_type: note\n note_gitlab_id: {gitlab_id}\n project: {path_with_namespace}\n parent_type: {Issue|MergeRequest}\n parent_iid: {iid}\n parent_title: {title}\n note_type: {DiffNote|DiscussionNote|...}\n author: @{author}\n created_at: {iso8601}\n resolved: {true|false} (only if resolvable)\n path: {position_new_path}:{line} (only if DiffNote with path)\n labels: {comma-separated parent labels}\n url: {url}\n\n --- Body ---\n\n {body}\n\n7. Title: \"Note by @{author} on {Issue|MR} #{iid}: {parent_title}\"\n\n8. Compute hashes: content_hash via compute_content_hash() (line 66), labels_hash via compute_list_hash(), paths_hash via compute_list_hash(). Apply truncate_hard_cap() (imported from truncation.rs at line 9).\n\n9. Return DocumentData (struct defined at line 47) with: source_type: SourceType::Note, source_id: note_id, project_id, author_username, labels, paths (as Vec), labels_hash, paths_hash, created_at, updated_at, url, title, content_text (from hard_cap), content_hash, is_truncated, truncated_reason.\n\n## Files\n- MODIFY: src/documents/extractor.rs (add extract_note_document after extract_discussion_document, ~line 500)\n- MODIFY: src/documents/mod.rs (add extract_note_document to pub use exports, line 12 area)\n\n## TDD Anchor\nRED: test_note_document_basic_format — insert project, issue, discussion, note; extract; assert content contains [[Note]], author, parent reference.\nGREEN: Implement extract_note_document with SQL JOIN and content formatting.\nVERIFY: cargo test note_document_basic_format -- --nocapture\nTests: test_note_document_diffnote_with_path, test_note_document_inherits_parent_labels, test_note_document_mr_parent, test_note_document_system_note_returns_none, test_note_document_not_found, test_note_document_orphaned_discussion, test_note_document_hash_deterministic, test_note_document_empty_body, test_note_document_null_body\n\n## Acceptance Criteria\n- [ ] extract_note_document returns Some(DocumentData) for non-system notes\n- [ ] Returns None for system notes, not-found, orphaned discussions\n- [ ] Content includes structured [[Note]] header with all parent context fields\n- [ ] DiffNote includes file path and line info in content header\n- [ ] Labels inherited from parent issue/MR\n- [ ] URL format: parent_url#note_{gitlab_id}\n- [ ] Title format: \"Note by @{author} on {Issue|MR} #{iid}: {parent_title}\"\n- [ ] Hash is deterministic across calls (same input = same hash)\n- [ ] Empty/null body handled gracefully (use empty string)\n- [ ] truncate_hard_cap applied to content\n- [ ] All 10 tests pass\n\n## Dependency Context\n- Depends on NOTE-2B (bd-ef0u): SourceType::Note variant must exist to construct DocumentData\n\n## Edge Cases\n- NULL body: use empty string \"\" — not all notes have body text\n- Orphaned discussion: parent issue/MR deleted but discussion remains — return None\n- Very long note body: truncate_hard_cap handles this (2MB limit)\n- Note with no position data: skip path line in content header\n- Note on MR vs Issue: different label table (mr_labels vs issue_labels)","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-12T17:02:01.802842Z","created_by":"tayloreernisse","updated_at":"2026-02-12T18:13:23.928224Z","closed_at":"2026-02-12T18:13:23.928173Z","close_reason":"Implemented by agent swarm","compaction_level":0,"original_size":0,"labels":["per-note","search"],"dependencies":[{"issue_id":"bd-18yh","depends_on_id":"bd-2ezb","type":"blocks","created_at":"2026-02-12T17:04:49.598730Z","created_by":"tayloreernisse"},{"issue_id":"bd-18yh","depends_on_id":"bd-3cjp","type":"blocks","created_at":"2026-02-12T17:04:50.015759Z","created_by":"tayloreernisse"}]} +{"id":"bd-1au9","title":"Audit and improve test coverage across ingestion module","description":"During code reorganization, discovered that ingestion/issues.rs has only 4 tests covering passes_cursor_filter, while 10 production functions (~400 lines) are untested:\n\nUNTESTED FUNCTIONS in ingestion/issues.rs:\n- ingest_issues() - main async pipeline with cursor-based pagination, shutdown handling\n- process_single_issue() - transforms GitLab issue, wraps in transaction\n- process_issue_in_transaction() - DB upsert with ON CONFLICT, label/assignee/milestone association, dirty tracking\n- upsert_label_tx() - label upsert with INSERT OR IGNORE + created count tracking\n- link_issue_label_tx() - issue-label junction table insert\n- upsert_milestone_tx() - milestone upsert with RETURNING id\n- get_sync_cursor() - reads sync_cursors table for incremental sync\n- update_sync_cursor() - writes sync cursor with tie-breaker ID\n- get_issues_needing_discussion_sync() - identifies issues needing discussion refresh\n- parse_timestamp() - RFC3339 parsing with error wrapping\n\nLIKELY SIMILAR GAPS in sibling files:\n- ingestion/merge_requests.rs (479 lines) - parallel structure to issues.rs\n- ingestion/discussions.rs (469 lines prod code) - discussion upsert pipeline\n- ingestion/mr_discussions.rs (738 lines prod before tests) - MR discussion pipeline\n- ingestion/orchestrator.rs (1703 lines) - full pipeline orchestration\n\nThe ingestion module handles the most critical data path (GitLab API -> SQLite) yet relies primarily on integration-level orchestrator tests rather than unit tests for individual functions.\n\nPRIORITY AREAS:\n1. DB upsert logic with ON CONFLICT handling (data correctness)\n2. Cursor-based pagination (incremental sync correctness)\n3. Label/milestone/assignee association (relational integrity)\n4. Dirty tracker marking after upserts (document pipeline triggering)\n5. Discussion sync queue population (cascading sync correctness)\n6. Error handling paths (invalid timestamps, missing data)\n\nAPPROACH: Use in-memory SQLite (create_connection(Path::new(\":memory:\")) + run_migrations) for unit tests. See existing patterns in core/db_tests.rs and documents/regenerator_tests.rs.","status":"open","priority":2,"issue_type":"task","created_at":"2026-02-13T00:53:15.302370Z","created_by":"tayloreernisse","updated_at":"2026-02-13T00:53:15.305167Z","compaction_level":0,"original_size":0,"labels":["testing"]} {"id":"bd-1b0n","title":"OBSERV: Print human-readable timing summary after interactive sync","description":"## Background\nInteractive users want a quick timing summary after sync completes. This is the human-readable equivalent of meta.stages in robot JSON. Gated behind IngestDisplay::show_text so it doesn't appear in -q, robot, or progress_only modes.\n\n## Approach\nAdd a function to format and print the timing summary, called from run_sync() after the pipeline completes:\n\n```rust\nfn print_timing_summary(stages: &[StageTiming], total_elapsed: Duration) {\n eprintln!();\n eprintln!(\"Sync complete in {:.1}s\", total_elapsed.as_secs_f64());\n for stage in stages {\n let dots = \".\".repeat(20_usize.saturating_sub(stage.name.len()));\n eprintln!(\n \" {} {} {:.1}s ({} items{})\",\n stage.name,\n dots,\n stage.elapsed_ms as f64 / 1000.0,\n stage.items_processed,\n if stage.errors > 0 { format!(\", {} errors\", stage.errors) } else { String::new() },\n );\n }\n}\n```\n\nCall in run_sync() (src/cli/commands/sync.rs), after pipeline and before return:\n```rust\nif display.show_text {\n let stages = metrics_handle.extract_timings();\n print_timing_summary(&stages, start.elapsed());\n}\n```\n\nOutput format per PRD Section 4.6.4:\n```\nSync complete in 45.2s\n Ingest issues .... 12.3s (150 items, 42 discussions)\n Ingest MRs ....... 18.9s (85 items, 1 error)\n Generate docs .... 8.5s (235 documents)\n Embed ............ 5.5s (1024 chunks)\n```\n\n## Acceptance Criteria\n- [ ] Interactive lore sync prints timing summary to stderr after completion\n- [ ] Summary shows total time and per-stage breakdown\n- [ ] lore -q sync does NOT print timing summary\n- [ ] Robot mode does NOT print timing summary (only JSON)\n- [ ] Error counts shown when non-zero\n- [ ] cargo clippy --all-targets -- -D warnings passes\n\n## Files\n- src/cli/commands/sync.rs (add print_timing_summary function, call after pipeline)\n\n## TDD Loop\nRED: test_timing_summary_format (capture stderr, verify format matches PRD example pattern)\nGREEN: Implement print_timing_summary, gate behind display.show_text\nVERIFY: cargo test && cargo clippy --all-targets -- -D warnings\n\n## Edge Cases\n- Empty stages (e.g., sync with no projects configured): print \"Sync complete in 0.0s\" with no stage lines\n- Very fast stages (<1ms): show \"0.0s\" not scientific notation\n- Stage names with varying lengths: dot padding keeps alignment readable","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-04T15:54:32.109882Z","created_by":"tayloreernisse","updated_at":"2026-02-04T17:32:52.558314Z","closed_at":"2026-02-04T17:32:52.558264Z","close_reason":"Added print_timing_summary with per-stage breakdown (name, elapsed, items, errors, rate limits), nested sub-stage support, gated behind metrics Option","compaction_level":0,"original_size":0,"labels":["observability"],"dependencies":[{"issue_id":"bd-1b0n","depends_on_id":"bd-1zj6","type":"blocks","created_at":"2026-02-04T15:55:20.162069Z","created_by":"tayloreernisse"},{"issue_id":"bd-1b0n","depends_on_id":"bd-3er","type":"parent-child","created_at":"2026-02-04T15:54:32.110706Z","created_by":"tayloreernisse"}]} {"id":"bd-1b50","title":"Update existing tests for new ScoringConfig fields","description":"## Background\nThe existing test test_expert_scoring_weights_are_configurable (who.rs:3551-3574) constructs a ScoringConfig with only the original 3 fields. After bd-2w1p adds 8 new fields, this test will not compile without ..Default::default().\n\n## Approach\nFind the test at who.rs:3551-3574. The flipped config construction at line 3567:\n```rust\nlet flipped = ScoringConfig {\n author_weight: 5,\n reviewer_weight: 30,\n note_bonus: 1,\n};\n```\nChange to:\n```rust\nlet flipped = ScoringConfig {\n author_weight: 5,\n reviewer_weight: 30,\n note_bonus: 1,\n ..Default::default()\n};\n```\n\nAlso check default_scoring() helper at line 2451 — it calls ScoringConfig::default() which already works.\n\n### Important: Scope boundary\nThis bead ONLY handles ScoringConfig struct literal changes. The query_expert() function signature change (7 params -> 10 params) happens in bd-13q8 (Layer 3), which is responsible for updating all test callsites at that time.\n\n### Why existing assertions do not break:\nAll test data is inserted with now_ms(). With as_of_ms also at ~now_ms(), elapsed ~0ms, decay ~1.0. So integer-rounded scores are identical to the flat-weight model.\n\n## Acceptance Criteria\n- [ ] cargo test passes with zero assertion changes to existing test values\n- [ ] test_expert_scoring_weights_are_configurable compiles and passes\n- [ ] All other existing who tests pass unchanged\n- [ ] No new test code needed — only ..Default::default() additions\n- [ ] cargo check --all-targets clean\n\n## Files\n- MODIFY: src/cli/commands/who.rs (ScoringConfig literal at line 3567)\n\n## TDD Loop\nN/A — mechanical change, no new tests.\nVERIFY: cargo check --all-targets && cargo test -p lore -- test_expert_scoring_weights_are_configurable\n\n## Edge Cases\n- Search for ALL ScoringConfig { ... } literals in test module — there may be more than the one at line 3567\n- The default_scoring() helper at line 2451 uses ScoringConfig::default() — no change needed","status":"closed","priority":3,"issue_type":"task","created_at":"2026-02-09T17:00:45.084472Z","created_by":"tayloreernisse","updated_at":"2026-02-12T20:43:04.409277Z","closed_at":"2026-02-12T20:43:04.409239Z","close_reason":"Implemented by time-decay swarm: 3 agents, 12 tasks, 621 tests passing, all quality gates green","compaction_level":0,"original_size":0,"labels":["scoring","test"],"dependencies":[{"issue_id":"bd-1b50","depends_on_id":"bd-2w1p","type":"blocks","created_at":"2026-02-09T17:01:11.362893Z","created_by":"tayloreernisse"}]} {"id":"bd-1b6k","title":"Epic: TUI Phase 5.5 — Reliability Test Pack","description":"## Background\nPhase 5.5 is a comprehensive reliability test suite covering race conditions, stress tests, property-based testing, and deterministic clock verification. These tests ensure the TUI is robust under adverse conditions (rapid input, concurrent writes, resize storms, backpressure).\n\n## Acceptance Criteria\n- [ ] Stale response drop tests pass\n- [ ] Sync cancel/resume tests pass\n- [ ] SQLITE_BUSY retry tests pass\n- [ ] Resize storm + rapid keypress tests pass without panic\n- [ ] Property tests for navigation invariants pass\n- [ ] Performance benchmark fixtures (S/M/L tiers) pass SLOs\n- [ ] Event fuzz tests: 10k traces with zero invariant violations\n- [ ] Deterministic clock/render tests produce identical output\n- [ ] 30-minute soak test: no panic, no deadlock, memory growth < 5%\n- [ ] Concurrent pagination/write race tests: no duplicate/skipped rows\n- [ ] Query cancellation race tests: no cross-task bleed, no stuck loading","status":"open","priority":1,"issue_type":"epic","created_at":"2026-02-12T17:04:04.486702Z","created_by":"tayloreernisse","updated_at":"2026-02-12T18:11:51.508682Z","compaction_level":0,"original_size":0,"labels":["TUI"],"dependencies":[{"issue_id":"bd-1b6k","depends_on_id":"bd-3t6r","type":"blocks","created_at":"2026-02-12T18:11:51.508655Z","created_by":"tayloreernisse"}]} @@ -85,7 +86,7 @@ {"id":"bd-1y7q","title":"Write invariant tests for ranking system","description":"## Background\nInvariant tests catch subtle ranking regressions that example-based tests miss. These test properties that must hold for ANY input, not specific values.\n\n## Approach\n\n### test_score_monotonicity_by_age:\nGenerate 50 random (age_ms, half_life_days) pairs using a simple LCG PRNG (deterministic seed for reproducibility). Assert decay(older) <= decay(newer) for all pairs where older > newer. Tests the pure half_life_decay() function only.\n\n### test_row_order_independence:\nInsert the same 5 signals in two orderings (forward and reverse). Run query_expert on both -> assert identical username ordering and identical scores (f64 bit-equal). Use a deterministic dataset with varied timestamps.\n\n### test_reviewer_split_is_exhaustive:\nSet up 3 reviewers on the same MR:\n1. Reviewer with substantive DiffNotes (>= 20 chars) -> must appear in participated ONLY\n2. Reviewer with no DiffNotes -> must appear in assigned-only ONLY\n3. Reviewer with trivial note (< 20 chars) -> must appear in assigned-only ONLY\nUse --explain-score to verify each reviewer's components: participated reviewer has reviewer_participated > 0 and reviewer_assigned == 0; others have reviewer_assigned > 0 and reviewer_participated == 0.\n\n### test_deterministic_accumulation_order:\nInsert signals for one user with 15 MRs at varied timestamps. Run query_expert 100 times in a loop. Assert all 100 runs produce the exact same f64 score (use == not approx, to verify bit-identical results from sorted accumulation).\n\n## Acceptance Criteria\n- [ ] All 4 tests pass\n- [ ] No flakiness across 10 consecutive cargo test runs\n- [ ] test_score_monotonicity covers at least 50 random pairs\n- [ ] test_deterministic_accumulation runs at least 100 iterations\n\n## Files\n- src/cli/commands/who.rs (test module)\n\n## Edge Cases\n- LCG PRNG for monotonicity test: use fixed seed, not rand crate (avoid dependency)\n- Bit-identical f64: use assert_eq!(a, b) not approx — the deterministic ordering guarantees this\n- Row order test: must insert in genuinely different orders, not just shuffled within same transaction","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-09T17:00:35.774542Z","created_by":"tayloreernisse","updated_at":"2026-02-09T17:17:18.920235Z","closed_at":"2026-02-09T17:17:18.920188Z","close_reason":"Tests distributed to implementation beads: monotonicity->bd-1soz, row_order+split+deterministic->bd-13q8","compaction_level":0,"original_size":0,"labels":["scoring","test"]} {"id":"bd-1y8","title":"Implement chunk ID encoding module","description":"## Background\nsqlite-vec uses a single integer rowid for embeddings. To store multiple chunks per document, we encode (document_id, chunk_index) into a single rowid using a multiplier. This module is shared between the embedding pipeline (encode on write) and vector search (decode on read). The encoding scheme supports up to 1000 chunks per document.\n\n## Approach\nCreate `src/embedding/chunk_ids.rs`:\n\n```rust\n/// Multiplier for encoding (document_id, chunk_index) into a single rowid.\n/// Supports up to 1000 chunks per document (32M chars at 32k/chunk).\npub const CHUNK_ROWID_MULTIPLIER: i64 = 1000;\n\n/// Encode (document_id, chunk_index) into a sqlite-vec rowid.\n///\n/// rowid = document_id * CHUNK_ROWID_MULTIPLIER + chunk_index\npub fn encode_rowid(document_id: i64, chunk_index: i64) -> i64 {\n document_id * CHUNK_ROWID_MULTIPLIER + chunk_index\n}\n\n/// Decode a sqlite-vec rowid back into (document_id, chunk_index).\npub fn decode_rowid(rowid: i64) -> (i64, i64) {\n let document_id = rowid / CHUNK_ROWID_MULTIPLIER;\n let chunk_index = rowid % CHUNK_ROWID_MULTIPLIER;\n (document_id, chunk_index)\n}\n```\n\nAlso create the parent module `src/embedding/mod.rs`:\n```rust\npub mod chunk_ids;\n// Later beads add: pub mod ollama; pub mod pipeline;\n```\n\nUpdate `src/lib.rs`: add `pub mod embedding;`\n\n## Acceptance Criteria\n- [ ] `encode_rowid(42, 0)` == 42000\n- [ ] `encode_rowid(42, 5)` == 42005\n- [ ] `decode_rowid(42005)` == (42, 5)\n- [ ] Roundtrip: decode(encode(doc_id, chunk_idx)) == (doc_id, chunk_idx) for all valid inputs\n- [ ] CHUNK_ROWID_MULTIPLIER is 1000\n- [ ] `cargo test chunk_ids` passes\n\n## Files\n- `src/embedding/chunk_ids.rs` — new file\n- `src/embedding/mod.rs` — new file (module root)\n- `src/lib.rs` — add `pub mod embedding;`\n\n## TDD Loop\nRED: Tests in `#[cfg(test)] mod tests`:\n- `test_encode_single_chunk` — encode(1, 0) == 1000\n- `test_encode_multi_chunk` — encode(1, 5) == 1005\n- `test_decode_roundtrip` — property test over range of doc_ids and chunk_indices\n- `test_decode_zero_chunk` — decode(42000) == (42, 0)\n- `test_multiplier_value` — assert CHUNK_ROWID_MULTIPLIER == 1000\nGREEN: Implement encode_rowid, decode_rowid\nVERIFY: `cargo test chunk_ids`\n\n## Edge Cases\n- chunk_index >= 1000: not expected (documents that large would be pathological), but no runtime panic — just incorrect decode. The embedding pipeline caps chunks well below this.\n- document_id = 0: valid (encode returns chunk_index directly)","status":"closed","priority":3,"issue_type":"task","created_at":"2026-01-30T15:26:34.060769Z","created_by":"tayloreernisse","updated_at":"2026-01-30T16:51:59.048910Z","closed_at":"2026-01-30T16:51:59.048843Z","close_reason":"Completed: chunk_ids module with encode_rowid/decode_rowid, CHUNK_ROWID_MULTIPLIER=1000, 6 tests pass","compaction_level":0,"original_size":0} {"id":"bd-1yu","title":"[CP1] GitLab types for issues, discussions, notes","description":"Add TypeScript interfaces for GitLab API responses.\n\nTypes to add to src/gitlab/types.ts:\n- GitLabIssue: id, iid, project_id, title, description, state, timestamps, author, labels[], labels_details?, web_url\n- GitLabDiscussion: id (string), individual_note, notes[]\n- GitLabNote: id, type, body, author, timestamps, system, resolvable, resolved, resolved_by, resolved_at, position?\n\nFiles: src/gitlab/types.ts\nDone when: Types compile and match GitLab API documentation","status":"tombstone","priority":2,"issue_type":"task","created_at":"2026-01-25T15:19:00.558718Z","created_by":"tayloreernisse","updated_at":"2026-01-25T15:21:35.153996Z","deleted_at":"2026-01-25T15:21:35.153993Z","deleted_by":"tayloreernisse","delete_reason":"delete","original_type":"task","compaction_level":0,"original_size":0} -{"id":"bd-1yx","title":"Implement rename chain resolution for file-history","description":"## Background\n\nRename chain resolution is the core algorithm for Gate 4. When querying history of src/auth.rs, it finds MRs that touched the file when it was previously named src/authentication.rs. This is reused by Gate 5 (trace) as well.\n\n**Spec reference:** `docs/phase-b-temporal-intelligence.md` Section 4.6 (Rename Handling).\n\n## Codebase Context\n\n- mr_file_changes table (migration 016, bd-1oo): merge_request_id, project_id, old_path, new_path, change_type\n- change_type='renamed' rows have both old_path and new_path populated\n- Partial index `idx_mfc_renamed` on (project_id, change_type) WHERE change_type='renamed' optimizes BFS queries\n- Also `idx_mfc_project_path` on (project_id, new_path) and `idx_mfc_project_old_path` partial index\n- No timeline/trace/file_history modules exist yet in src/core/\n\n## Approach\n\nCreate `src/core/file_history.rs`:\n\n```rust\nuse std::collections::HashSet;\nuse std::collections::VecDeque;\nuse rusqlite::Connection;\nuse crate::core::error::Result;\n\n/// Resolves a file path through its rename history.\n/// Returns all equivalent paths (original + renames) for use in queries.\n/// BFS in both directions: forward (old_path -> new_path) and backward (new_path -> old_path).\npub fn resolve_rename_chain(\n conn: &Connection,\n project_id: i64,\n path: &str,\n max_hops: usize, // default 10 from CLI\n) -> Result> {\n let mut visited: HashSet = HashSet::new();\n let mut queue: VecDeque = VecDeque::new();\n\n visited.insert(path.to_string());\n queue.push_back(path.to_string());\n\n let forward_sql = \"SELECT mfc.new_path FROM mr_file_changes mfc \\\n WHERE mfc.project_id = ?1 AND mfc.old_path = ?2 AND mfc.change_type = 'renamed'\";\n let backward_sql = \"SELECT mfc.old_path FROM mr_file_changes mfc \\\n WHERE mfc.project_id = ?1 AND mfc.new_path = ?2 AND mfc.change_type = 'renamed'\";\n\n while let Some(current) = queue.pop_front() {\n if visited.len() > max_hops + 1 { break; }\n\n // Forward: current was the old name -> discover new names\n let mut stmt = conn.prepare(forward_sql)?;\n let forward: Vec = stmt.query_map(\n rusqlite::params\\![project_id, current],\n |row| row.get(0),\n )?.filter_map(|r| r.ok()).collect();\n\n // Backward: current was the new name -> discover old names\n let mut stmt = conn.prepare(backward_sql)?;\n let backward: Vec = stmt.query_map(\n rusqlite::params\\![project_id, current],\n |row| row.get(0),\n )?.filter_map(|r| r.ok()).collect();\n\n for discovered in forward.into_iter().chain(backward) {\n if visited.insert(discovered.clone()) {\n queue.push_back(discovered);\n }\n }\n }\n\n Ok(visited.into_iter().collect())\n}\n```\n\nRegister in `src/core/mod.rs`: add `pub mod file_history;`\n\n## Acceptance Criteria\n\n- [ ] `resolve_rename_chain()` follows renames in both directions (forward + backward)\n- [ ] Cycles detected via HashSet (same path never visited twice)\n- [ ] Bounded at max_hops (default 10)\n- [ ] No renames found: returns vec with just the original path\n- [ ] max_hops=0: returns just original path without querying DB\n- [ ] Module registered in src/core/mod.rs as `pub mod file_history;`\n- [ ] `cargo check --all-targets` passes\n- [ ] `cargo clippy --all-targets -- -D warnings` passes\n\n## Files\n\n- `src/core/file_history.rs` (NEW)\n- `src/core/mod.rs` (add `pub mod file_history;`)\n\n## TDD Loop\n\nRED:\n- `test_rename_chain_no_renames` — returns just original path\n- `test_rename_chain_forward` — a.rs -> b.rs -> c.rs: starting from a.rs finds all three\n- `test_rename_chain_backward` — starting from c.rs finds a.rs and b.rs\n- `test_rename_chain_cycle_detection` — a->b->a terminates without infinite loop\n- `test_rename_chain_max_hops_zero` — returns just original path\n- `test_rename_chain_max_hops_bounded` — chain longer than max is truncated\n\nTests need in-memory DB with migrations applied through 016 + mr_file_changes test data with change_type='renamed'.\n\nGREEN: Implement BFS with visited set.\n\nVERIFY: `cargo test --lib -- file_history`\n\n## Edge Cases\n\n- File never renamed: single-element vec\n- Circular rename (a->b->a): visited set prevents infinite loop\n- max_hops=0: return just original path, no queries executed\n- Case sensitivity: paths are case-sensitive (Linux default, matches GitLab behavior)\n- Multiple renames from same old_path: BFS discovers all branches\n","status":"open","priority":2,"issue_type":"task","created_at":"2026-02-02T21:34:08.985345Z","created_by":"tayloreernisse","updated_at":"2026-02-05T20:54:52.423441Z","compaction_level":0,"original_size":0,"labels":["gate-4","phase-b","query"],"dependencies":[{"issue_id":"bd-1yx","depends_on_id":"bd-14q","type":"parent-child","created_at":"2026-02-02T21:34:08.986730Z","created_by":"tayloreernisse"},{"issue_id":"bd-1yx","depends_on_id":"bd-1oo","type":"blocks","created_at":"2026-02-02T21:34:16.698782Z","created_by":"tayloreernisse"}]} +{"id":"bd-1yx","title":"Implement rename chain resolution for file-history","description":"## Background\n\nRename chain resolution is the core algorithm for Gate 4. When querying history of src/auth.rs, it finds MRs that touched the file when it was previously named src/authentication.rs. This is reused by Gate 5 (trace) as well.\n\n**Spec reference:** `docs/phase-b-temporal-intelligence.md` Section 4.6 (Rename Handling).\n\n## Codebase Context\n\n- mr_file_changes table (migration 016, bd-1oo): merge_request_id, project_id, old_path, new_path, change_type\n- change_type='renamed' rows have both old_path and new_path populated\n- Partial index `idx_mfc_renamed` on (project_id, change_type) WHERE change_type='renamed' optimizes BFS queries\n- Also `idx_mfc_project_path` on (project_id, new_path) and `idx_mfc_project_old_path` partial index\n- No timeline/trace/file_history modules exist yet in src/core/\n\n## Approach\n\nCreate `src/core/file_history.rs`:\n\n```rust\nuse std::collections::HashSet;\nuse std::collections::VecDeque;\nuse rusqlite::Connection;\nuse crate::core::error::Result;\n\n/// Resolves a file path through its rename history.\n/// Returns all equivalent paths (original + renames) for use in queries.\n/// BFS in both directions: forward (old_path -> new_path) and backward (new_path -> old_path).\npub fn resolve_rename_chain(\n conn: &Connection,\n project_id: i64,\n path: &str,\n max_hops: usize, // default 10 from CLI\n) -> Result> {\n let mut visited: HashSet = HashSet::new();\n let mut queue: VecDeque = VecDeque::new();\n\n visited.insert(path.to_string());\n queue.push_back(path.to_string());\n\n let forward_sql = \"SELECT mfc.new_path FROM mr_file_changes mfc \\\n WHERE mfc.project_id = ?1 AND mfc.old_path = ?2 AND mfc.change_type = 'renamed'\";\n let backward_sql = \"SELECT mfc.old_path FROM mr_file_changes mfc \\\n WHERE mfc.project_id = ?1 AND mfc.new_path = ?2 AND mfc.change_type = 'renamed'\";\n\n while let Some(current) = queue.pop_front() {\n if visited.len() > max_hops + 1 { break; }\n\n // Forward: current was the old name -> discover new names\n let mut stmt = conn.prepare(forward_sql)?;\n let forward: Vec = stmt.query_map(\n rusqlite::params\\![project_id, current],\n |row| row.get(0),\n )?.filter_map(|r| r.ok()).collect();\n\n // Backward: current was the new name -> discover old names\n let mut stmt = conn.prepare(backward_sql)?;\n let backward: Vec = stmt.query_map(\n rusqlite::params\\![project_id, current],\n |row| row.get(0),\n )?.filter_map(|r| r.ok()).collect();\n\n for discovered in forward.into_iter().chain(backward) {\n if visited.insert(discovered.clone()) {\n queue.push_back(discovered);\n }\n }\n }\n\n Ok(visited.into_iter().collect())\n}\n```\n\nRegister in `src/core/mod.rs`: add `pub mod file_history;`\n\n## Acceptance Criteria\n\n- [ ] `resolve_rename_chain()` follows renames in both directions (forward + backward)\n- [ ] Cycles detected via HashSet (same path never visited twice)\n- [ ] Bounded at max_hops (default 10)\n- [ ] No renames found: returns vec with just the original path\n- [ ] max_hops=0: returns just original path without querying DB\n- [ ] Module registered in src/core/mod.rs as `pub mod file_history;`\n- [ ] `cargo check --all-targets` passes\n- [ ] `cargo clippy --all-targets -- -D warnings` passes\n\n## Files\n\n- `src/core/file_history.rs` (NEW)\n- `src/core/mod.rs` (add `pub mod file_history;`)\n\n## TDD Loop\n\nRED:\n- `test_rename_chain_no_renames` — returns just original path\n- `test_rename_chain_forward` — a.rs -> b.rs -> c.rs: starting from a.rs finds all three\n- `test_rename_chain_backward` — starting from c.rs finds a.rs and b.rs\n- `test_rename_chain_cycle_detection` — a->b->a terminates without infinite loop\n- `test_rename_chain_max_hops_zero` — returns just original path\n- `test_rename_chain_max_hops_bounded` — chain longer than max is truncated\n\nTests need in-memory DB with migrations applied through 016 + mr_file_changes test data with change_type='renamed'.\n\nGREEN: Implement BFS with visited set.\n\nVERIFY: `cargo test --lib -- file_history`\n\n## Edge Cases\n\n- File never renamed: single-element vec\n- Circular rename (a->b->a): visited set prevents infinite loop\n- max_hops=0: return just original path, no queries executed\n- Case sensitivity: paths are case-sensitive (Linux default, matches GitLab behavior)\n- Multiple renames from same old_path: BFS discovers all branches\n","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-02T21:34:08.985345Z","created_by":"tayloreernisse","updated_at":"2026-02-13T14:00:46.354253Z","closed_at":"2026-02-13T14:00:46.354201Z","close_reason":"Implemented resolve_rename_chain() BFS in src/core/file_history.rs with 8 tests covering: no renames, forward chain, backward chain, cycle detection, max_hops=0, max_hops bounded, branching renames, project isolation. All 765 tests pass, clippy+fmt clean.","compaction_level":0,"original_size":0,"labels":["gate-4","phase-b","query"],"dependencies":[{"issue_id":"bd-1yx","depends_on_id":"bd-14q","type":"parent-child","created_at":"2026-02-02T21:34:08.986730Z","created_by":"tayloreernisse"},{"issue_id":"bd-1yx","depends_on_id":"bd-1oo","type":"blocks","created_at":"2026-02-02T21:34:16.698782Z","created_by":"tayloreernisse"}]} {"id":"bd-1yz","title":"Implement MR document extraction","description":"## Background\nMR documents are similar to issue documents but include source/target branch information in the header. The extractor queries merge_requests and mr_labels tables. Like issue extraction, it produces a DocumentData struct for the regeneration pipeline.\n\n## Approach\nImplement `extract_mr_document()` in `src/documents/extractor.rs`:\n\n```rust\n/// Extract a searchable document from a merge request.\n/// Returns None if the MR has been deleted from the DB.\npub fn extract_mr_document(conn: &Connection, mr_id: i64) -> Result>\n```\n\n**SQL queries (from PRD Section 2.2):**\n```sql\n-- Main entity\nSELECT m.id, m.iid, m.title, m.description, m.state, m.author_username,\n m.source_branch, m.target_branch,\n m.created_at, m.updated_at, m.web_url,\n p.path_with_namespace, p.id AS project_id\nFROM merge_requests m\nJOIN projects p ON p.id = m.project_id\nWHERE m.id = ?\n\n-- Labels\nSELECT l.name FROM mr_labels ml\nJOIN labels l ON l.id = ml.label_id\nWHERE ml.merge_request_id = ?\nORDER BY l.name\n```\n\n**Document format:**\n```\n[[MergeRequest]] !456: Implement JWT authentication\nProject: group/project-one\nURL: https://gitlab.example.com/group/project-one/-/merge_requests/456\nLabels: [\"feature\", \"auth\"]\nState: opened\nAuthor: @johndoe\nSource: feature/jwt-auth -> main\n\n--- Description ---\n\nThis MR implements JWT-based authentication...\n```\n\n**Key difference from issues:** The `Source:` line with `source_branch -> target_branch`.\n\n## Acceptance Criteria\n- [ ] Deleted MR returns Ok(None)\n- [ ] MR document has `[[MergeRequest]]` prefix with `!` before iid (not `#`)\n- [ ] Source line shows `source_branch -> target_branch`\n- [ ] Labels sorted alphabetically in JSON array\n- [ ] content_hash computed from full content_text\n- [ ] labels_hash computed from sorted labels\n- [ ] paths is empty (MR-level docs don't have DiffNote paths; those are on discussion docs)\n- [ ] `cargo test extract_mr` passes\n\n## Files\n- `src/documents/extractor.rs` — implement `extract_mr_document()`\n\n## TDD Loop\nRED: Tests in `#[cfg(test)] mod tests`:\n- `test_mr_document_format` — verify header matches PRD template with Source line\n- `test_mr_not_found` — returns Ok(None)\n- `test_mr_no_description` — header only\n- `test_mr_branch_info` — Source line correct\nGREEN: Implement extract_mr_document with SQL queries\nVERIFY: `cargo test extract_mr`\n\n## Edge Cases\n- MR with NULL description: skip \"--- Description ---\" section\n- MR with NULL source_branch or target_branch: omit Source line (shouldn't happen in practice)\n- Draft MRs: state field captures this, no special handling needed","status":"closed","priority":3,"issue_type":"task","created_at":"2026-01-30T15:25:45.521703Z","created_by":"tayloreernisse","updated_at":"2026-01-30T17:30:04.308781Z","closed_at":"2026-01-30T17:30:04.308598Z","close_reason":"Implemented extract_mr_document() with Source line, PRD format, and 5 tests","compaction_level":0,"original_size":0,"dependencies":[{"issue_id":"bd-1yz","depends_on_id":"bd-36p","type":"blocks","created_at":"2026-01-30T15:29:15.749264Z","created_by":"tayloreernisse"},{"issue_id":"bd-1yz","depends_on_id":"bd-hrs","type":"blocks","created_at":"2026-01-30T15:29:15.814729Z","created_by":"tayloreernisse"}]} {"id":"bd-1zj6","title":"OBSERV: Enrich robot JSON meta with run_id and stages","description":"## Background\nRobot JSON currently has a flat meta.elapsed_ms. This enriches it with run_id and a stages array, making every lore --robot sync output a complete performance profile.\n\n## Approach\nThe robot JSON output is built in src/cli/commands/sync.rs. The current SyncResult (line 15-25) is serialized into the data field. The meta field is built alongside it.\n\n1. Find or create the SyncMeta struct (likely near SyncResult). Add fields:\n```rust\n#[derive(Debug, Serialize)]\nstruct SyncMeta {\n run_id: String,\n elapsed_ms: u64,\n stages: Vec,\n}\n```\n\n2. After run_sync() completes, extract timings from MetricsLayer:\n```rust\nlet stages = metrics_handle.extract_timings();\nlet meta = SyncMeta {\n run_id: run_id.to_string(),\n elapsed_ms: start.elapsed().as_millis() as u64,\n stages,\n};\n```\n\n3. Build the JSON envelope:\n```rust\nlet output = serde_json::json!({\n \"ok\": true,\n \"data\": result,\n \"meta\": meta,\n});\n```\n\nThe metrics_handle (Arc) must be passed from main.rs to the command handler. This requires adding a parameter to handle_sync_cmd() and run_sync(), or using a global. Prefer parameter passing.\n\nSame pattern for standalone ingest: add stages to IngestMeta.\n\n## Acceptance Criteria\n- [ ] lore --robot sync output includes meta.run_id (string, 8 hex chars)\n- [ ] lore --robot sync output includes meta.stages (array of StageTiming)\n- [ ] meta.elapsed_ms still present (total wall clock time)\n- [ ] Each stage has name, elapsed_ms, items_processed at minimum\n- [ ] Top-level stages have sub_stages when applicable\n- [ ] lore --robot ingest also includes run_id and stages\n- [ ] cargo clippy --all-targets -- -D warnings passes\n\n## Files\n- src/cli/commands/sync.rs (add SyncMeta struct, wire extract_timings)\n- src/cli/commands/ingest.rs (same for standalone ingest)\n- src/main.rs (pass metrics_handle to command handlers)\n\n## TDD Loop\nRED: test_sync_meta_includes_stages (run robot-mode sync, parse JSON, assert meta.stages is array)\nGREEN: Add SyncMeta, extract timings, include in JSON output\nVERIFY: cargo test && cargo clippy --all-targets -- -D warnings\n\n## Edge Cases\n- Empty stages: if sync runs with --no-docs --no-embed, some stages won't exist. stages array is shorter, not padded.\n- extract_timings() called before root span closes: returns incomplete tree. Must call AFTER run_sync returns (span is dropped on function exit).\n- metrics_handle clone: MetricsLayer uses Arc internally, clone is cheap (reference count increment).","status":"closed","priority":2,"issue_type":"task","created_at":"2026-02-04T15:54:32.062410Z","created_by":"tayloreernisse","updated_at":"2026-02-04T17:31:11.073580Z","closed_at":"2026-02-04T17:31:11.073534Z","close_reason":"Wired MetricsLayer into subscriber stack (all 4 branches), added run_id to SyncResult, enriched SyncMeta with run_id + stages Vec, updated print_sync_json to accept MetricsLayer and extract timings","compaction_level":0,"original_size":0,"labels":["observability"],"dependencies":[{"issue_id":"bd-1zj6","depends_on_id":"bd-34ek","type":"blocks","created_at":"2026-02-04T15:55:20.085372Z","created_by":"tayloreernisse"},{"issue_id":"bd-1zj6","depends_on_id":"bd-3er","type":"parent-child","created_at":"2026-02-04T15:54:32.063354Z","created_by":"tayloreernisse"}]} {"id":"bd-1zow","title":"Implement Search screen (state + action + view)","description":"## Background\nThe Search screen provides full-text and semantic search across all indexed documents. It supports 3 modes (lexical FTS5, hybrid FTS+vector, semantic vector-only), a split-pane layout with results on the left and preview on the right, and capability-aware mode selection based on available indexes.\n\n## Approach\nState (state/search.rs):\n- SearchState: query (String), query_input (TextInput), query_focused (bool), mode (SearchMode), results (Vec), selected_index (usize), preview (Option), capabilities (SearchCapabilities), generation (u64)\n- SearchMode: Lexical, Hybrid, Semantic\n- SearchCapabilities: has_fts (bool), has_embeddings (bool), embedding_coverage_pct (f32)\n- SearchResult: doc_id, entity_type, entity_iid, project_path, title, snippet, score, mode_used\n- SearchPreview: full document text or entity detail\n\n**Capability detection** (on screen entry):\n- Probe documents_fts table: SELECT COUNT(*) FROM documents_fts_docsize (uses fast B-tree count, not FTS5 virtual table scan — see MEMORY.md perf audit)\n- Probe embeddings: SELECT COUNT(*) FROM embeddings / SELECT COUNT(*) FROM documents to compute coverage pct\n- If has_fts=false: disable Lexical and Hybrid modes, only Semantic available\n- If has_embeddings=false: disable Semantic and Hybrid modes, only Lexical available\n- If both false: show \"No search indexes found. Run lore generate-docs and lore embed first.\"\n\n**Score explanation (e key):**\n- Press e on a selected result to toggle a score breakdown panel\n- For Lexical: show FTS5 bm25 raw score\n- For Hybrid: show FTS score, vector score, and RRF combined score with weights\n- For Semantic: show cosine similarity score\n- Panel appears below the selected result row, Esc or e dismisses\n\n**Debounced input (200ms):**\n- Uses Msg::SearchDebounceArmed and Msg::SearchDebounced timer pattern\n- On keystroke in query input: arm debounce timer via Cmd::timer(200ms, Msg::SearchDebounced)\n- On SearchDebounced: execute search with current query text\n- This prevents flooding the search backend on rapid typing\n\nAction (action.rs):\n- fetch_search_capabilities(conn) -> SearchCapabilities: probe FTS and embedding tables\n- execute_search(conn, query, mode, limit) -> Vec: dispatches to correct search backend. Uses existing crate::search module functions.\n- fetch_search_preview(conn, result) -> SearchPreview: loads full entity detail for selected result\n\nView (view/search.rs):\n- Split pane: results list (60%) | preview (40%)\n- Query bar at top with mode indicator (L/H/S)\n- Mode switching: Tab cycles modes (only available modes based on capabilities)\n- Score column shows numeric score; e key expands explanation\n- Empty query shows recent entities instead of empty state\n- Narrow terminal (<100 cols): hide preview pane\n\n## Acceptance Criteria\n- [ ] 3 search modes: Lexical, Hybrid, Semantic\n- [ ] Mode switching via Tab, only available modes selectable based on capability detection\n- [ ] Capability detection probes FTS and embedding tables on screen entry\n- [ ] Graceful degradation: unavailable modes shown as greyed out with reason\n- [ ] \"No search indexes\" message when both FTS and embeddings are empty\n- [ ] 200ms debounce on search input (timer-driven via Msg::SearchDebounceArmed/Fired)\n- [ ] Split pane: results | preview\n- [ ] Enter on result navigates to entity detail\n- [ ] Score shown next to each result\n- [ ] e key toggles score explanation panel for selected result\n- [ ] Empty query shows recent entities instead of empty state\n- [ ] Narrow terminal (<100 cols): hide preview pane\n\n## Files\n- MODIFY: crates/lore-tui/src/state/search.rs (expand from stub)\n- MODIFY: crates/lore-tui/src/action.rs (add search functions)\n- CREATE: crates/lore-tui/src/view/search.rs\n\n## TDD Anchor\nRED: Write test_search_capability_detection that creates DB with FTS but no embeddings, asserts has_fts=true, has_embeddings=false, Semantic mode disabled.\nGREEN: Implement fetch_search_capabilities.\nVERIFY: cargo test --manifest-path crates/lore-tui/Cargo.toml test_search_capability\n\nAdditional tests:\n- test_debounce_prevents_rapid_search: simulate 5 keystrokes in 100ms, assert only 1 search executed\n- test_score_explanation_lexical: verify bm25 score shown for Lexical mode result\n- test_empty_query_shows_recent: assert recent entities returned when query is empty\n\n## Edge Cases\n- Search query < 2 chars: don't execute search (debounce filter)\n- FTS5 special characters (*, \", -): escape or pass through based on mode\n- Hybrid mode: uses existing RRF implementation from crate::search module\n- Very large result sets: limit to 100 results, show \"more results available\" hint\n- Preview pane on narrow terminal (<100 cols): hide preview, full-width results only\n- FTS count performance: use documents_fts_docsize shadow table for COUNT (19x faster)\n\n## Dependency Context\nUses existing search infrastructure from lore core (crate::search::{FtsQueryMode, to_fts_query} — note private submodules, import via crate::search).\nUses SearchDebounceArmed/SearchDebounced Msg variants from \"Implement core types\" (bd-c9gk).\nUses TaskSupervisor debounce management from \"Implement TaskSupervisor\" (bd-3le2).\nUses AppState composition from \"Implement AppState composition\" (bd-1v9m).","status":"open","priority":2,"issue_type":"task","created_at":"2026-02-12T17:00:48.862621Z","created_by":"tayloreernisse","updated_at":"2026-02-12T18:11:33.891935Z","compaction_level":0,"original_size":0,"labels":["TUI"],"dependencies":[{"issue_id":"bd-1zow","depends_on_id":"bd-1mju","type":"blocks","created_at":"2026-02-12T17:10:02.823681Z","created_by":"tayloreernisse"},{"issue_id":"bd-1zow","depends_on_id":"bd-nwux","type":"blocks","created_at":"2026-02-12T18:11:33.891908Z","created_by":"tayloreernisse"}]} diff --git a/.beads/last-touched b/.beads/last-touched index ea9f7be..ad6ea2a 100644 --- a/.beads/last-touched +++ b/.beads/last-touched @@ -1 +1 @@ -bd-226s +bd-1yx