feat: TUI pipeline builder (Phase 1 MVP) + auto-versioning by benbernard · Pull Request #92 · benbernard/RecordStream

benbernard · 2026-02-23T05:33:52Z

Summary

TUI Pipeline Builder (Phase 1 MVP)

Interactive TUI for building RecordStream pipelines, built with OpenTUI + React.

Vertical split-pane layout: Pipeline stage list (left) + data inspector (right)
Operation picker: Categorized (input/transform/output) with fuzzy search and preview pane
Live data inspection: Auto-updates when navigating between stages. Table, prettyprint, and JSON views.
Pipeline executor: Reuses all existing RecordStream operations via createOperation(). Handles line-oriented, bulk-content, and self-contained input ops.
Export: x copies pipeline as shell pipe script to clipboard. X opens format picker (pipe script / chain command / save to file).
Undo/redo: Full undo/redo on all structural pipeline edits (add/delete/edit/toggle/reorder stages). 200-entry stack.
Smart caching: Cache invalidation cascades downstream when a stage is edited. Record count ratios shown in inspector header.
Error handling: Inline error indicators on failed stages, downstream stages greyed out, error details in inspector.
Keyboard-first: vim-style navigation (j/k), context-sensitive key hints in status bar, help panel (?).
Entry point: recs tui [file] or recs tui --help

New files: 32 source files under src/tui/ + bin/recs-tui.ts

Auto-Versioning

Release workflow auto-increments patch version from git tags on every merge to master
No manual version bumps needed — git tags are the source of truth
package.json version stamped into build artifacts but not committed back

Advisory Benchmarks

Benchmark CI step never fails the build (advisory-only)
PR comment includes note about GitHub Actions runner variance
Short summary at top with collapsible details section

Test plan

920 tests pass (0 failures)
TypeScript type-check clean
Lint clean (only pre-existing warning)
88 TUI-specific tests: reducer (42), executor (18), serialization (17), integration (11)
Fuzzy search tests (13+)
UX review completed — all P0/P1 issues resolved
QA verification pass — all imports resolve, CLI entry points work

🤖 Generated with Claude Code

Implements the core data model for the TUI pipeline builder: - types.ts: All TypeScript types (Stage, Fork, PipelineState, PipelineAction union, etc.) - reducer.ts: pipelineReducer with undo/redo middleware, all action handlers - selectors.ts: getActivePath, isDownstreamOfError, getStageOutput, etc. - undo.ts: extractSnapshot, describeAction, UNDOABLE_ACTIONS set Also fixes OpenTUI prop usage (color -> fg, backgroundColor -> bg) in modal components to resolve type errors. Includes 42 unit tests covering add/delete/move stages, undo/redo, toggle enabled, error propagation, cursor movement, and selectors. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

- Wire `recs tui` subcommand in bin/recs.ts (spawns bin/recs-tui.ts) - Add bin/recs-tui.ts CLI entry with --session, --pipeline, --list, --clean - Add src/tui/index.tsx with launchTui() using OpenTUI React renderer - Add src/tui/components/App.tsx root component with welcome/main scaffolding - Add src/tui/utils/fuzzy-match.ts for AddStageModal search (13 tests) - Configure tsconfig.json for JSX (react-jsx, @opentui/react, .tsx includes) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

- serialization.ts: exportAsPipeScript() for multi-line shell scripts with shebang and proper shellEscape(), exportAsChainCommand() for single-line recs chain format, copyToClipboard() via pbcopy/xclip - ExportPicker.tsx: format picker modal (pipe script / chain / save) - 17 tests for shell escaping, pipe script output, and chain format Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

11 integration tests exercising the real reducer, executor, and operations together: - Full pipeline lifecycle (grep → sort, grep → sort → collate) - Undo/redo with execution verification - Cache invalidation after stage modification - Error propagation and isDownstreamOfError - Export round-trip (pipe script and chain command) - Toggle stage enabled/disabled during execution - Large pipeline (10 stages) with full path execution - Input op integration (fromcsv with file input) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Fix P0 bug where UPDATE_STAGE_ARGS and TOGGLE_STAGE did not invalidate cache entries, causing the inspector to show stale results after editing a stage's arguments or toggling enabled state. Cache invalidation now cascades to all downstream stages in the fork. Wire missing keybindings: x (export pipe script), X (ExportPicker), r (re-run from cursor), t (cycle inspector view), J/K (reorder stages), v (export records to temp file), Ctrl+C (quit). Remove Phase 2+ shortcuts (f, b, i, p) from HelpPanel and StatusBar. Additional UX fixes: auto-calculate RecordTable column widths from data, show "N of M records (X%)" ratio in InspectorHeader, increase pipeline panel width from 28 to 32. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

github-actions · 2026-02-23T05:35:03Z

Performance Benchmark Results

⚠️ 12 regressions detected out of 103 benchmarks (threshold: 25%)

Benchmark	Median	Baseline	Delta
InputStream.fromString — 100 records	376.0µs	216.7µs	+73.5% 🔴
Compiled KeySpec.resolveValue — nested (address/zip)	302.8µs	169.1µs	+79.1% 🔴
Compiled KeySpec.resolveValue — deep (address/coords/lat)	230.9µs	122.8µs	+87.9% 🔴
Compiled KeySpec.resolveValue — array (tags/#0)	261.0µs	155.4µs	+68.0% 🔴
grep — 10K records (string match)	821.8µs	452.2µs	+81.8% 🔴
collate — 100 records (count by city)	438.4µs	315.6µs	+38.9% 🔴
chain — 5 ops (grep	eval	grep	eval
new Record() — 10K objects	153.3µs	97.0µs	+58.1% 🔴
Record.dataRef — 10K records (zero-copy)	86.2µs	35.1µs	+145.7% 🔴
chain — 3 ops (grep	eval	grep), 100 records	162.4µs
chain — 3 ops (grep	eval	grep), 10K records	1.36ms
manual buffer (isolated) — 10K lines	6.67ms	5.32ms	+25.3% 🔴

103 benchmarks: 7 faster, 19 slower, 77 within noise (10%)

ℹ️ Note: Benchmarks are advisory-only. GitHub Actions shared runners have variable performance, so results may fluctuate ±25% between runs. For reliable benchmarking, run locally with bun run bench.

Full benchmark results

JSON Parsing

Benchmark	Median	Baseline	Delta	Throughput
Record.fromJSON — 100 lines	158.5µs	161.2µs	-1.7%	630.73K rec/s
Record.fromJSON — 10K lines	13.88ms	13.16ms	+5.5%	720.54K rec/s, 213.4 MB/s
InputStream.fromString — 100 records	376.0µs	216.7µs	+73.5% 🔴	265.98K rec/s
InputStream.fromString — 10K records	18.25ms	18.50ms	-1.3%	548.03K rec/s, 162.3 MB/s
JSON.parse baseline — 10K lines (no Record)	12.76ms	13.23ms	-3.5%	783.52K rec/s, 232.0 MB/s
JSON.parse single array — 10K records	12.31ms	12.91ms	-4.6%	812.35K rec/s, 240.6 MB/s

JSON Serialization

Benchmark	Median	Baseline	Delta	Throughput
Record.toString — 100 records	82.6µs	90.4µs	-8.6%	1.21M rec/s
Record.toString — 10K records	8.38ms	7.07ms	+18.5% 🔴	1.19M rec/s, 353.6 MB/s
Record.toJSON — 10K records	282.3µs	285.1µs	-1.0%	35.43M rec/s
JSON.stringify baseline — 10K objects (no Record)	7.82ms	8.06ms	-3.0%	1.28M rec/s, 378.8 MB/s
Batch join — 10K records (map+join)	8.24ms	8.71ms	-5.4%	1.21M rec/s, 359.5 MB/s

KeySpec Access

Benchmark	Median	Baseline	Delta	Throughput
KeySpec — simple key (name)	223.6µs	212.5µs	+5.2%	44.73M rec/s
KeySpec — nested key (address/zip)	560.9µs	556.2µs	+0.8%	17.83M rec/s
KeySpec — deep nested (address/coords/lat)	1.07ms	1.09ms	-1.5%	9.35M rec/s
KeySpec — array index (tags/#0)	924.6µs	922.7µs	+0.2%	10.82M rec/s
Direct property access baseline (rec['name'])	80.5µs	90.0µs	-10.6% 🟢	124.25M rec/s
Direct nested access baseline (rec.address.coords.lat)	93.8µs	96.8µs	-3.1%	106.63M rec/s
KeySpec construction — cached (same spec 10K times)	343.3µs	279.6µs	+22.8% 🔴	29.13M rec/s
KeySpec construction — unique specs (10K different)	2.27ms	2.51ms	-9.5%	4.40M rec/s
Compiled KeySpec.resolveValue — nested (address/zip)	302.8µs	169.1µs	+79.1% 🔴	33.03M rec/s
Compiled KeySpec.resolveValue — deep (address/coords/lat)	230.9µs	122.8µs	+87.9% 🔴	43.31M rec/s
Compiled KeySpec.resolveValue — array (tags/#0)	261.0µs	155.4µs	+68.0% 🔴	38.31M rec/s
Compiled KeySpec.setValue — nested (address/zip)	151.9µs	147.6µs	+2.9%	65.84M rec/s

Core Operations

Benchmark	Median	Baseline	Delta	Throughput
grep — 10K records (r.age > 50)	449.1µs	456.3µs	-1.6%	22.27M rec/s
grep — 10K records (string match)	821.8µs	452.2µs	+81.8% 🔴	12.17M rec/s
eval — 10K records (add computed field)	2.09ms	2.46ms	-15.2% 🟢	4.79M rec/s
xform — 10K records (push each record)	1.83ms	1.75ms	+4.4%	5.46M rec/s
sort — 100 records (by score, numeric)	167.6µs	141.0µs	+18.8% 🔴	596.77K rec/s
sort — 10K records (by score, numeric)	17.86ms	17.98ms	-0.6%	559.83K rec/s
sort — 10K records (by name, lexical)	12.99ms	11.85ms	+9.7%	769.54K rec/s
collate — 100 records (count by city)	438.4µs	315.6µs	+38.9% 🔴	228.11K rec/s
collate — 10K records (count by city)	11.86ms	13.12ms	-9.6%	842.97K rec/s
fromcsv — 10K rows (parse CSV to records)	16.22ms	15.32ms	+5.9%	616.59K rec/s, 40.5 MB/s

Pipeline Overhead

Benchmark	Median	Baseline	Delta	Throughput
chain — single op (grep), 10K records	7.72ms	7.37ms	+4.7%	1.30M rec/s
chain — 3 ops (grep	eval	grep), 10K records	7.41ms	7.50ms
chain — 5 ops (grep	eval	grep	eval	grep), 10K records
passthrough baseline — 10K records (direct collector)	6.36ms	6.56ms	-3.0%	1.57M rec/s

Record Creation & Serialization

Benchmark	Median	Baseline	Delta	Throughput
new Record() — 10K objects	153.3µs	97.0µs	+58.1% 🔴	65.25M rec/s
new Record() empty — 10K	145.1µs	141.1µs	+2.8%	68.94M rec/s
Record.get — 10K records × 3 fields	61.3µs	51.1µs	+19.9% 🔴	489.33M rec/s
Record.set — 10K records × 1 field	61.1µs	61.6µs	-0.8%	163.68M rec/s
Record.toJSON — 10K records	281.9µs	285.1µs	-1.2%	35.48M rec/s
Record.toString — 10K records	8.52ms	7.07ms	+20.6% 🔴	1.17M rec/s
Record.clone — 10K records	58.72ms	60.76ms	-3.3%	170.29K rec/s
Record.fromJSON — 10K lines	13.37ms	13.16ms	+1.6%	747.91K rec/s, 221.5 MB/s
Record.dataRef — 10K records (zero-copy)	86.2µs	35.1µs	+145.7% 🔴	116.07M rec/s
Record.sort — 10K records (numeric field)	11.38ms	11.66ms	-2.4%	878.39K rec/s
Record.sort — 10K records (lexical field)	6.03ms	6.05ms	-0.4%	1.66M rec/s
Record.cmp — 1M comparisons (single field)	104.96ms	113.88ms	-7.8%	9.53M rec/s
Record.sort — 10K records (nested field numeric)	15.94ms	15.26ms	+4.5%	627.17K rec/s
Record.cmp — 1M comparisons (multi-field cached)	87.99ms	88.86ms	-1.0%	11.36M rec/s
Record.sort — 10K records (cached comparator reuse)	11.59ms	11.64ms	-0.4%	862.47K rec/s

Chain vs Pipe

Benchmark	Median	Baseline	Delta	Throughput
chain — 2 ops (grep	eval), 100 records	166.1µs	155.6µs	+6.8%
pipe — 2 ops (grep	eval), 100 records	359.93ms	373.10ms	-3.5%
implicit — 2 ops (grep	eval), 100 records	142.5µs	136.4µs	+4.4%
chain — 2 ops (grep	eval), 1K records	290.7µs	314.7µs	-7.6%
pipe — 2 ops (grep	eval), 1K records	374.31ms	426.88ms	-12.3% 🟢
implicit — 2 ops (grep	eval), 1K records	208.2µs	191.2µs	+8.9%
chain — 2 ops (grep	eval), 10K records	1.06ms	1.06ms	+0.0%
pipe — 2 ops (grep	eval), 10K records	383.41ms	366.77ms	+4.5%
implicit — 2 ops (grep	eval), 10K records	1.07ms	1.08ms	-0.8%
chain — 3 ops (grep	eval	grep), 100 records	162.4µs	102.0µs
pipe — 3 ops (grep	eval	grep), 100 records	549.05ms	549.01ms
implicit — 3 ops (grep	eval	grep), 100 records	178.6µs	192.2µs
chain — 3 ops (grep	eval	grep), 1K records	300.3µs	297.2µs
pipe — 3 ops (grep	eval	grep), 1K records	562.57ms	558.11ms
implicit — 3 ops (grep	eval	grep), 1K records	230.6µs	214.8µs
chain — 3 ops (grep	eval	grep), 10K records	1.36ms	1.07ms
pipe — 3 ops (grep	eval	grep), 10K records	554.31ms	549.33ms
implicit — 3 ops (grep	eval	grep), 10K records	1.08ms	1.08ms
chain — 5 ops (grep	eval	grep	eval	grep), 100 records
pipe — 5 ops (grep	eval	grep	eval	grep), 100 records
implicit — 5 ops (grep	eval	grep	eval	grep), 100 records
chain — 5 ops (grep	eval	grep	eval	grep), 1K records
pipe — 5 ops (grep	eval	grep	eval	grep), 1K records
implicit — 5 ops (grep	eval	grep	eval	grep), 1K records
chain — 5 ops (grep	eval	grep	eval	grep), 10K records
pipe — 5 ops (grep	eval	grep	eval	grep), 10K records
implicit — 5 ops (grep	eval	grep	eval	grep), 10K records

Line Reading

Benchmark	Median	Baseline	Delta	Throughput
InputStream.fromFile — 100 lines	476.8µs	487.4µs	-2.2%	209.73K rec/s, 61.9 MB/s
InputStream.fromString — 100 lines	182.6µs	197.6µs	-7.6%	547.56K rec/s, 161.7 MB/s
manual buffer (isolated) — 100 lines	263.3µs	253.3µs	+3.9%	379.81K rec/s, 112.2 MB/s
bulk text + split — 100 lines	70.3µs	75.9µs	-7.4%	1.42M rec/s, 420.1 MB/s
node readline — 100 lines	545.0µs	449.6µs	+21.2% 🔴	183.48K rec/s, 54.2 MB/s
TextDecoderStream — 100 lines	247.4µs	273.9µs	-9.7%	404.21K rec/s, 119.4 MB/s
binary newline scan — 100 lines	274.5µs	371.8µs	-26.2% 🟢	364.28K rec/s, 107.6 MB/s
bun native stdin — 100 lines	24.62ms	24.44ms	+0.8%	4.06K rec/s, 1.2 MB/s
InputStream.fromFile — 10K lines	23.22ms	22.54ms	+3.0%	430.68K rec/s, 127.5 MB/s
InputStream.fromString — 10K lines	18.09ms	17.63ms	+2.6%	552.84K rec/s, 163.7 MB/s
manual buffer (isolated) — 10K lines	6.67ms	5.32ms	+25.3% 🔴	1.50M rec/s, 443.7 MB/s
bulk text + split — 10K lines	2.38ms	2.50ms	-4.8%	4.20M rec/s, 1243.6 MB/s
node readline — 10K lines	9.13ms	8.95ms	+2.0%	1.10M rec/s, 324.5 MB/s
TextDecoderStream — 10K lines	5.49ms	5.13ms	+7.0%	1.82M rec/s, 539.0 MB/s
binary newline scan — 10K lines	7.80ms	6.25ms	+24.7% 🔴	1.28M rec/s, 379.8 MB/s
bun native stdin — 10K lines	41.96ms	43.34ms	-3.2%	238.30K rec/s, 70.6 MB/s
InputStream.fromFile — 100K lines	260.11ms	262.35ms	-0.9%	384.46K rec/s, 114.2 MB/s
InputStream.fromString — 100K lines	222.95ms	220.89ms	+0.9%	448.52K rec/s, 133.3 MB/s
manual buffer (isolated) — 100K lines	32.49ms	33.14ms	-2.0%	3.08M rec/s, 914.7 MB/s
bulk text + split — 100K lines	27.95ms	27.52ms	+1.6%	3.58M rec/s, 1063.0 MB/s
node readline — 100K lines	84.68ms	84.09ms	+0.7%	1.18M rec/s, 350.9 MB/s
TextDecoderStream — 100K lines	42.23ms	41.21ms	+2.5%	2.37M rec/s, 703.7 MB/s
binary newline scan — 100K lines	63.82ms	64.03ms	-0.3%	1.57M rec/s, 465.6 MB/s
bun native stdin — 100K lines	114.57ms	122.92ms	-6.8%	872.83K rec/s, 259.4 MB/s

feat: TUI pipeline builder (Phase 1 MVP) + auto-versioning

benbernard and others added 5 commits February 22, 2026 07:29

benbernard merged commit 06317bf into master Feb 23, 2026
4 checks passed

benbernard deleted the issue-cleanups branch February 23, 2026 05:40

benbernard added a commit that referenced this pull request Feb 23, 2026

Merge pull request #92 from benbernard/issue-cleanups

bdfdc71

feat: TUI pipeline builder (Phase 1 MVP) + auto-versioning

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: TUI pipeline builder (Phase 1 MVP) + auto-versioning#92

feat: TUI pipeline builder (Phase 1 MVP) + auto-versioning#92
benbernard merged 5 commits intomasterfrom
issue-cleanups

benbernard commented Feb 23, 2026

Uh oh!

github-actions bot commented Feb 23, 2026

JSON Parsing

JSON Serialization

KeySpec Access

Core Operations

Pipeline Overhead

Record Creation & Serialization

Chain vs Pipe

Line Reading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

benbernard commented Feb 23, 2026

Summary

TUI Pipeline Builder (Phase 1 MVP)

Auto-Versioning

Advisory Benchmarks

Test plan

Uh oh!

github-actions bot commented Feb 23, 2026

Performance Benchmark Results

JSON Parsing

JSON Serialization

KeySpec Access

Core Operations

Pipeline Overhead

Record Creation & Serialization

Chain vs Pipe

Line Reading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant