feat: {{}} compound assignments and polyglot support

[benbernard]

Summary

• Fix {{field}} += val compound assignments in JS (previously produced invalid code like __get(r, "field") += val)
• Add {{}} template expansion to Python and Perl snippet runners, so the same {{field}} syntax works across all three languages
• Add __get/__set helpers to Python runner.py and Perl runner.pl
• Add recordVar parameter to transformCode() to support Perl's $r variable

Test plan

• bun test tests/Executor.test.ts — 43 tests including 25 new compound/recordVar tests
• bun test tests/snippets/TemplateExpansion.test.ts — 36 cross-language tests (12 cases × 3 languages)
• bun test — full suite passes (1595 pass, 0 fail)
• Manual: echo '{"x":5}' | recs eval '{{x}} += 10' → {"x":15} (JS, Python, Perl)

🤖 Generated with Claude Code

[github-actions]

Performance Benchmark Results

⚠️ 11 regressions detected out of 103 benchmarks (threshold: 25%)

Benchmark	Median	Baseline	Delta
Compiled KeySpec.resolveValue — nested (address/zip)	310.0µs	174.2µs	+78.0% 🔴
Compiled KeySpec.resolveValue — deep (address/coords/lat)	237.7µs	122.5µs	+93.9% 🔴
Compiled KeySpec.resolveValue — array (tags/#0)	241.2µs	153.6µs	+57.1% 🔴
chain — 5 ops (grep	eval	grep	eval
implicit — 2 ops (grep	eval), 100 records	115.0µs	73.7µs
chain — 2 ops (grep	eval), 1K records	328.1µs	197.4µs
implicit — 3 ops (grep	eval	grep), 1K records	303.1µs
implicit — 5 ops (grep	eval	grep	eval
implicit — 5 ops (grep	eval	grep	eval
manual buffer (isolated) — 100 lines	264.5µs	178.3µs	+48.4% 🔴
binary newline scan — 100 lines	376.9µs	289.8µs	+30.0% 🔴

103 benchmarks: 15 faster, 16 slower, 72 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	151.6µs	153.3µs	-1.1%	659.49K rec/s
Record.fromJSON — 10K lines	14.16ms	13.12ms	+8.0%	706.12K rec/s, 209.1 MB/s
InputStream.fromString — 100 records	211.2µs	206.3µs	+2.4%	473.53K rec/s
InputStream.fromString — 10K records	18.39ms	18.53ms	-0.8%	543.75K rec/s, 161.0 MB/s
JSON.parse baseline — 10K lines (no Record)	13.42ms	12.76ms	+5.2%	745.21K rec/s, 220.7 MB/s
JSON.parse single array — 10K records	12.90ms	12.55ms	+2.8%	775.21K rec/s, 229.6 MB/s

JSON Serialization

Benchmark	Median	Baseline	Delta	Throughput
Record.toString — 100 records	83.3µs	71.3µs	+16.7% 🔴	1.20M rec/s
Record.toString — 10K records	8.52ms	8.98ms	-5.1%	1.17M rec/s, 347.6 MB/s
Record.toJSON — 10K records	294.7µs	298.7µs	-1.3%	33.93M rec/s
JSON.stringify baseline — 10K objects (no Record)	7.88ms	7.94ms	-0.8%	1.27M rec/s, 376.0 MB/s
Batch join — 10K records (map+join)	8.59ms	8.41ms	+2.1%	1.16M rec/s, 344.7 MB/s

KeySpec Access

Benchmark	Median	Baseline	Delta	Throughput
KeySpec — simple key (name)	228.7µs	214.4µs	+6.7%	43.73M rec/s
KeySpec — nested key (address/zip)	566.0µs	822.2µs	-31.2% 🟢	17.67M rec/s
KeySpec — deep nested (address/coords/lat)	1.09ms	1.09ms	-0.6%	9.20M rec/s
KeySpec — array index (tags/#0)	865.3µs	884.1µs	-2.1%	11.56M rec/s
Direct property access baseline (rec['name'])	76.0µs	88.2µs	-13.9% 🟢	131.64M rec/s
Direct nested access baseline (rec.address.coords.lat)	109.4µs	110.5µs	-1.0%	91.41M rec/s
KeySpec construction — cached (same spec 10K times)	411.6µs	481.3µs	-14.5% 🟢	24.30M rec/s
KeySpec construction — unique specs (10K different)	2.61ms	2.81ms	-7.0%	3.83M rec/s
Compiled KeySpec.resolveValue — nested (address/zip)	310.0µs	174.2µs	+78.0% 🔴	32.26M rec/s
Compiled KeySpec.resolveValue — deep (address/coords/lat)	237.7µs	122.5µs	+93.9% 🔴	42.08M rec/s
Compiled KeySpec.resolveValue — array (tags/#0)	241.2µs	153.6µs	+57.1% 🔴	41.46M rec/s
Compiled KeySpec.setValue — nested (address/zip)	146.7µs	201.3µs	-27.1% 🟢	68.14M rec/s

Core Operations

Benchmark	Median	Baseline	Delta	Throughput
grep — 10K records (r.age > 50)	491.2µs	442.0µs	+11.1% 🔴	20.36M rec/s
grep — 10K records (string match)	467.2µs	435.7µs	+7.2%	21.41M rec/s
eval — 10K records (add computed field)	2.45ms	2.46ms	-0.2%	4.08M rec/s
xform — 10K records (push each record)	2.13ms	2.15ms	-0.9%	4.70M rec/s
sort — 100 records (by score, numeric)	154.5µs	146.9µs	+5.2%	647.07K rec/s
sort — 10K records (by score, numeric)	17.85ms	19.51ms	-8.5%	560.21K rec/s
sort — 10K records (by name, lexical)	12.52ms	12.66ms	-1.2%	798.87K rec/s
collate — 100 records (count by city)	305.8µs	399.6µs	-23.5% 🟢	327.05K rec/s
collate — 10K records (count by city)	14.89ms	12.42ms	+19.9% 🔴	671.65K rec/s
fromcsv — 10K rows (parse CSV to records)	14.19ms	14.52ms	-2.3%	704.65K rec/s, 46.3 MB/s

Pipeline Overhead

Benchmark	Median	Baseline	Delta	Throughput
chain — single op (grep), 10K records	7.24ms	7.05ms	+2.6%	1.38M rec/s
chain — 3 ops (grep	eval	grep), 10K records	7.44ms	7.34ms
chain — 5 ops (grep	eval	grep	eval	grep), 10K records
passthrough baseline — 10K records (direct collector)	6.08ms	6.33ms	-3.9%	1.64M rec/s

Record Creation & Serialization

Benchmark	Median	Baseline	Delta	Throughput
new Record() — 10K objects	95.5µs	100.0µs	-4.5%	104.68M rec/s
new Record() empty — 10K	127.5µs	134.6µs	-5.3%	78.44M rec/s
Record.get — 10K records × 3 fields	93.0µs	99.4µs	-6.5%	322.55M rec/s
Record.set — 10K records × 1 field	125.5µs	124.1µs	+1.2%	79.66M rec/s
Record.toJSON — 10K records	287.2µs	298.7µs	-3.9%	34.82M rec/s
Record.toString — 10K records	6.77ms	8.98ms	-24.6% 🟢	1.48M rec/s
Record.clone — 10K records	6.17ms	6.60ms	-6.5%	1.62M rec/s
Record.fromJSON — 10K lines	13.05ms	13.12ms	-0.5%	766.12K rec/s, 226.9 MB/s
Record.dataRef — 10K records (zero-copy)	35.2µs	86.2µs	-59.2% 🟢	284.25M rec/s
Record.sort — 10K records (numeric field)	11.76ms	11.64ms	+1.0%	850.42K rec/s
Record.sort — 10K records (lexical field)	6.06ms	6.11ms	-0.9%	1.65M rec/s
Record.cmp — 1M comparisons (single field)	111.23ms	110.37ms	+0.8%	8.99M rec/s
Record.sort — 10K records (nested field numeric)	15.65ms	16.06ms	-2.6%	638.86K rec/s
Record.cmp — 1M comparisons (multi-field cached)	85.60ms	87.19ms	-1.8%	11.68M rec/s
Record.sort — 10K records (cached comparator reuse)	11.64ms	12.12ms	-3.9%	859.14K rec/s

Chain vs Pipe

Benchmark	Median	Baseline	Delta	Throughput
chain — 2 ops (grep	eval), 100 records	151.6µs	137.7µs	+10.1% 🔴
pipe — 2 ops (grep	eval), 100 records	420.89ms	388.66ms	+8.3%
implicit — 2 ops (grep	eval), 100 records	115.0µs	73.7µs	+56.2% 🔴
chain — 2 ops (grep	eval), 1K records	328.1µs	197.4µs	+66.2% 🔴
pipe — 2 ops (grep	eval), 1K records	381.75ms	380.07ms	+0.4%
implicit — 2 ops (grep	eval), 1K records	203.0µs	208.2µs	-2.5%
chain — 2 ops (grep	eval), 10K records	1.11ms	1.38ms	-19.7% 🟢
pipe — 2 ops (grep	eval), 10K records	376.25ms	375.39ms	+0.2%
implicit — 2 ops (grep	eval), 10K records	1.14ms	1.10ms	+3.5%
chain — 3 ops (grep	eval	grep), 100 records	103.7µs	171.1µs
pipe — 3 ops (grep	eval	grep), 100 records	580.40ms	572.39ms
implicit — 3 ops (grep	eval	grep), 100 records	116.3µs	112.2µs
chain — 3 ops (grep	eval	grep), 1K records	221.7µs	212.4µs
pipe — 3 ops (grep	eval	grep), 1K records	584.05ms	566.66ms
implicit — 3 ops (grep	eval	grep), 1K records	303.1µs	210.3µs
chain — 3 ops (grep	eval	grep), 10K records	1.18ms	1.08ms
pipe — 3 ops (grep	eval	grep), 10K records	577.80ms	576.72ms
implicit — 3 ops (grep	eval	grep), 10K records	1.41ms	1.37ms
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	501.4µs	713.4µs	-29.7% 🟢	199.45K rec/s, 58.9 MB/s
InputStream.fromString — 100 lines	190.1µs	335.0µs	-43.2% 🟢	525.90K rec/s, 155.3 MB/s
manual buffer (isolated) — 100 lines	264.5µs	178.3µs	+48.4% 🔴	378.14K rec/s, 111.7 MB/s
bulk text + split — 100 lines	69.3µs	70.3µs	-1.4%	1.44M rec/s, 425.9 MB/s
node readline — 100 lines	459.1µs	430.0µs	+6.8%	217.82K rec/s, 64.3 MB/s
TextDecoderStream — 100 lines	275.4µs	312.7µs	-11.9% 🟢	363.07K rec/s, 107.2 MB/s
binary newline scan — 100 lines	376.9µs	289.8µs	+30.0% 🔴	265.35K rec/s, 78.4 MB/s
bun native stdin — 100 lines	26.04ms	25.85ms	+0.7%	3.84K rec/s, 1.1 MB/s
InputStream.fromFile — 10K lines	24.45ms	24.10ms	+1.5%	408.98K rec/s, 121.1 MB/s
InputStream.fromString — 10K lines	17.90ms	17.97ms	-0.4%	558.80K rec/s, 165.5 MB/s
manual buffer (isolated) — 10K lines	6.50ms	5.85ms	+11.1% 🔴	1.54M rec/s, 455.9 MB/s
bulk text + split — 10K lines	2.40ms	2.45ms	-1.8%	4.16M rec/s, 1232.6 MB/s
node readline — 10K lines	9.39ms	9.33ms	+0.7%	1.06M rec/s, 315.3 MB/s
TextDecoderStream — 10K lines	4.67ms	4.72ms	-0.9%	2.14M rec/s, 633.7 MB/s
binary newline scan — 10K lines	7.40ms	9.90ms	-25.2% 🟢	1.35M rec/s, 400.1 MB/s
bun native stdin — 10K lines	42.76ms	42.60ms	+0.4%	233.88K rec/s, 69.3 MB/s
InputStream.fromFile — 100K lines	263.53ms	263.84ms	-0.1%	379.46K rec/s, 112.8 MB/s
InputStream.fromString — 100K lines	227.57ms	215.43ms	+5.6%	439.42K rec/s, 130.6 MB/s
manual buffer (isolated) — 100K lines	29.52ms	30.50ms	-3.2%	3.39M rec/s, 1006.6 MB/s
bulk text + split — 100K lines	32.06ms	30.82ms	+4.0%	3.12M rec/s, 926.8 MB/s
node readline — 100K lines	92.20ms	89.11ms	+3.5%	1.08M rec/s, 322.3 MB/s
TextDecoderStream — 100K lines	40.06ms	46.11ms	-13.1% 🟢	2.50M rec/s, 741.8 MB/s
binary newline scan — 100K lines	66.61ms	86.48ms	-23.0% 🟢	1.50M rec/s, 446.1 MB/s
bun native stdin — 100K lines	116.65ms	118.83ms	-1.8%	857.28K rec/s, 254.7 MB/s