feat(viz): Phase 2 — particles, ripple/pulse, ownership mode, time scrubber, guided tour#1
Open
feat(viz): Phase 2 — particles, ripple/pulse, ownership mode, time scrubber, guided tour#1
Conversation
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…k score Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds colors.js with DIR_COLORS palette and riskColor interpolation, layout.js with computeLayout() that maps data.tree into 1000x1000 world-space fileNodes and regions, and wires both into main.js. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
After a window resize, camera bounds are updated but camInitial was stale, causing incorrect zoom computation. Fix by registering a resize callback from interaction.js that refreshes camInitial and resets pan/zoom. Also removes unused imports: getInstancedMesh, dirColor, filePositions. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replaces D3/SVG force-directed viz with Three.js orthographic treemap. - Data pipeline: SQLite → data.json with risk scores and co-change data - Hierarchical treemap layout via d3-hierarchy - Instanced WebGL rendering with bloom post-processing - Architecture and Risk display modes - GPU color-buffer node picking - BFS blast radius on selection - Search with autocomplete - Stats panel, hotspot list, selection detail panel
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
… doc Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…vity, owner rows Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Bug 1 (search): Add Enter key handler to navigate to first result when the dropdown is visible. Previously pressing Enter did nothing, so users had to click results manually. Bug 2 (scrubber): Hide the scrubber bar when commits.first === commits.last (zero temporal range) instead of showing a bar that doesn't respond to dragging. Also fix the root cause in the Rust indexer: co-change edges now store the actual git commit timestamp (CommitInfo.timestamp) via the new add_cochange_edge method, rather than the indexing time (Utc::now). This produces meaningful temporal spread in data.json after re-indexing. Bug 3 (co-change sparsity): Lower the default cochange threshold in viz.py from 0.25 to 0.10 so more pairs survive the filter on repositories with limited commit history. Users can still pass --cochange-threshold to override. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Re-indexed cartograph itself with the fixed add_cochange_edge method. Co-change edges now store actual git commit timestamps rather than indexing time. 153 co-change pairs at threshold 0.10 (vs 30 at 0.25). Scrubber stays hidden (first === last) because this repo's entire history falls within one calendar month (March 2026). On a production repo with years of history, the scrubber will show real temporal spread. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ory) Indexed openclaw: 6566 TypeScript files, 10k commits, 563 co-change pairs at threshold 0.10. - commits: 2 buckets (Feb 2026, Mar 2026) — scrubber now functional - cochange_by_node: 399 nodes with partners (vs 30 on cartograph itself) - owners: 298 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The hardcoded 10k cap was cutting off history for large repos. OpenClaw has 19k commits — all fell in Feb-Mar 2026 with the old cap, giving the scrubber only 2 months. --max-commits defaults to 50000 (0 = no limit). The internal safety cap in commits.rs is raised to 100k since the CLI now owns the policy. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…its) Full 19,558-commit history mined (vs previous 10k cap). - commits: 3 buckets Jan–Mar 2026 (vs 2 before) — scrubber spans 3 months - cochange_by_node: 222 nodes (vs 399 at 10k cap — normalized confidence shifts with more commits, fewer pairs clear the 0.10 threshold) - 302 co-change edges at threshold 0.10 Older commits (Nov–Dec 2025) touch files since renamed/deleted; those paths have no entity match so they don't produce co-change edges. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Previously, only files matching current paths produced co-change edges.
Files renamed or moved had their entire pre-rename history ignored,
limiting the scrubber to Jan-Mar 2026 for openclaw despite commits
going back to Nov 2025.
Changes:
- commits.rs: add ChangeKind::Renamed { old_path } and parse R<score>
lines properly (was silently discarding old_path)
- cochange.rs: build_rename_map() walks commits newest-first to build
old_path → current_path resolution, with chained rename support.
analyze_cochanges() resolves all historical paths before creating
pairs, and skips deleted files.
- main.rs: --max-commits flag already in place from previous commit
Tests: test_rename_tracking_merges_history, test_deleted_files_excluded,
test_parse_rename_lines all pass.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…ltered build_commits() was receiving only the high-confidence edges (>=0.10), which were all from Jan-Mar 2026. The scrubber never showed older months even when the DB had edges going back to Nov 2025. Fix: load_entities_and_edges() now returns temporal_cochange_edges (threshold=0) alongside the normal cochange_edges. build_data() passes temporal_cochange_edges to build_commits() so the scrubber reflects the full activity history of the codebase. Result for openclaw: 5 months (Nov 2025 - Mar 2026) vs 3 before. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Phase 2 — Visualization Enhancements
Unified Filters
Test Plan
≥ N%×on chip → that filter clears, others unaffected🤖 Generated with Claude Code