CI job independence by frostney · Pull Request #59 · frostney/GocciaScript

frostney · 2026-03-10T09:14:11Z

Update CI workflows to ensure all jobs and matrix runs complete even if some fail.

Summary by CodeRabbit

Chores
- CI updated to avoid failing fast across matrix jobs so all builds/tests complete before reporting failures.
- Matrix adjusted to add Windows x86 coverage and selectively exclude Windows x64 in specific scenarios to better reflect supported variants.
- Release packaging and artifact handling refined to include Windows x86 entries and gracefully skip missing x64 artifacts.

- Add fail-fast: false to all matrix strategies so one platform failure doesn't cancel other platforms - Add if: !cancelled() to test, wasm-test, benchmark, and examples jobs so they run even if build had partial matrix failures - Update artifacts and release jobs to explicitly check all upstream results via needs.*.result == 'success' - Add if: !cancelled() to PR workflow downstream jobs (test, wasm-test, benchmark) for the same independence - benchmark-comment requires benchmark success explicitly Co-authored-by: Johannes Stein <frostney@users.noreply.github.com>

cursor · 2026-03-10T09:14:11Z

Cursor Agent can help with this pull request. Just @cursor in comments and I'll start working on changes in this branch.
_{Learn more about Cursor Agents}

coderabbitai · 2026-03-10T09:14:29Z

Warning

Rate limit exceeded

@cursor[bot] has exceeded the limit for the number of commits that can be reviewed per hour. Please wait 6 minutes and 44 seconds before requesting another review.

⌛ How to resolve this issue?

After the wait time has elapsed, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout.

Please see our FAQ for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: f87fca4d-9c1a-402d-877d-89eed2996208

📥 Commits

Reviewing files that changed from the base of the PR and between 1ee1067 and 6aac6f3.

📒 Files selected for processing (3)

.github/workflows/ci.yml
AGENTS.md
docs/build-system.md

📝 Walkthrough

Walkthrough

CI and PR GitHub Actions matrices were adjusted to disable fail-fast and reorganize Windows matrix entries from x64 as a primary OS to explicit x86 includes; packaging and artifact handling were updated accordingly to reference Windows x86 and skip missing x64 artifacts.

Changes

Cohort / File(s)	Summary
CI workflow `.github/workflows/ci.yml`	Added `fail-fast: false` to matrix strategies; reorganized platform matrices with `exclude` and `include` entries to remove Windows x64 from primary OS lists and add Windows x86 variants; updated packaging platform label from `windows-latest-x64` to `windows-latest-x86`; expanded artifact include lists and added logic to skip missing Windows x64 artifacts.
PR workflow `.github/workflows/pr.yml`	Added `fail-fast: false` to the strategy matrix for build and benchmark jobs so matrix cells do not short-circuit on failure.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Poem

🐰 I hopped through YAML, tidy and spry,

Fail-fast off — let tests all try,
Windows swapped to x86 in the pack,
Missing x64? We just skip back.
A tiny thump for CI on the fly 🥕✨

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'CI job independence' directly describes the main change: enabling CI workflows to complete all matrix runs independently by setting fail-fast to false.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch cursor/ci-job-independence-9f67

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.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

github-actions · 2026-03-10T09:18:29Z

Benchmark Results

254 benchmarks (no baseline)

arraybuffer.js — 14 benchmarks

Benchmark	Interpreted	Bytecode
create ArrayBuffer(0)	423,440	142,791
create ArrayBuffer(64)	408,921	143,658
create ArrayBuffer(1024)	323,049	126,129
create ArrayBuffer(8192)	139,287	81,912
slice full buffer (64 bytes)	507,076	378,913
slice half buffer (512 of 1024 bytes)	432,202	332,382
slice with negative indices	441,417	366,257
slice empty range	506,292	376,393
byteLength access	1,472,982	1,061,736
Symbol.toStringTag access	1,150,621	541,265
ArrayBuffer.isView	721,744	468,504
clone ArrayBuffer(64)	393,886	313,789
clone ArrayBuffer(1024)	293,083	251,499
clone ArrayBuffer inside object	263,047	156,001

arrays.js — 19 benchmarks

Benchmark	Interpreted	Bytecode
Array.from length 100	13,407	13,053
Array.from 10 elements	222,882	160,356
Array.of 10 elements	311,453	221,393
spread into new array	331,907	548,248
map over 50 elements	27,396	22,224
filter over 50 elements	22,888	21,600
reduce sum 50 elements	26,240	19,052
forEach over 50 elements	22,982	21,796
find in 50 elements	34,856	27,842
sort 20 elements	11,684	3,365
flat nested array	115,034	274,951
flatMap	70,928	182,922
map inside map (5x5)	20,711	67,595
filter inside map (5x10)	15,501	13,370
reduce inside map (5x10)	18,787	13,696
forEach inside forEach (5x10)	15,613	14,893
find inside some (10x10)	13,158	10,778
map+filter chain nested (5x20)	5,173	4,408
reduce flatten (10x5)	37,569	4,924

async-await.js — 6 benchmarks

Benchmark	Interpreted	Bytecode
single await	330,316	245,787
multiple awaits	152,367	101,209
await non-Promise value	719,912	694,407
await with try/catch	322,943	234,601
await Promise.all	47,900	40,930
nested async function call	162,959	176,703

classes.js — 31 benchmarks

Benchmark	Interpreted	Bytecode
simple class new	112,182	335,457
class with defaults	89,816	236,197
50 instances via Array.from	5,495	6,241
instance method call	56,663	155,262
static method call	88,760	352,216
single-level inheritance	44,208	141,976
two-level inheritance	38,222	122,587
private field access	56,059	162,043
private methods	60,568	208,065
getter/setter access	63,278	169,029
class decorator (identity)	76,986	52,171
class decorator (wrapping)	45,108	36,194
identity method decorator	55,220	42,686
wrapping method decorator	46,117	39,109
stacked method decorators (x3)	32,636	27,713
identity field decorator	61,040	45,433
field initializer decorator	51,343	38,757
getter decorator (identity)	53,117	41,115
setter decorator (identity)	43,694	35,440
static method decorator	57,086	63,112
static field decorator	67,069	66,095
private method decorator	45,789	37,110
private field decorator	49,933	38,916
plain auto-accessor (no decorator)	87,255	53,293
auto-accessor with decorator	49,323	37,385
decorator writing metadata	41,165	42,391
static getter read	97,505	386,978
static getter/setter pair	75,322	205,524
inherited static getter	53,913	261,586
inherited static setter	58,386	206,275
inherited static getter with this binding	48,930	152,575

closures.js — 11 benchmarks

Benchmark	Interpreted	Bytecode
closure over single variable	118,292	572,105
closure over multiple variables	113,182	381,275
nested closures	110,993	509,464
function as argument	88,571	472,621
function returning function	110,410	552,458
compose two functions	64,978	333,508
fn.call	143,713	129,924
fn.apply	107,856	85,843
fn.bind	130,711	139,909
recursive sum to 50	11,414	37,766
recursive tree traversal	18,905	57,687

collections.js — 12 benchmarks

Benchmark	Interpreted	Bytecode
add 50 elements	7,325	5,912
has lookup (50 elements)	93,569	90,330
delete elements	49,038	38,244
forEach iteration	15,575	15,455
spread to array	29,422	31,333
deduplicate array	40,618	31,988
set 50 entries	5,318	5,657
get lookup (50 entries)	93,344	387,621
has check	138,890	161,909
delete entries	49,070	36,222
forEach iteration	15,395	15,907
keys/values/entries	7,811	8,717

destructuring.js — 22 benchmarks

Benchmark	Interpreted	Bytecode
simple array destructuring	373,382	612,165
with rest element	261,551	457,018
with defaults	393,398	657,195
skip elements	393,155	661,072
nested array destructuring	161,317	349,762
swap variables	475,877	945,482
simple object destructuring	286,600	512,553
with defaults	341,090	320,940
with renaming	289,956	554,942
nested object destructuring	133,341	249,972
rest properties	170,346	232,747
object parameter	84,253	185,243
array parameter	113,621	316,314
mixed destructuring in map	33,564	34,756
forEach with array destructuring	62,869	134,371
map with array destructuring	62,465	165,805
filter with array destructuring	65,408	188,513
reduce with array destructuring	70,189	164,490
map with object destructuring	71,824	68,118
map with nested destructuring	60,425	57,002
map with rest in destructuring	35,229	42,308
map with defaults in destructuring	54,133	37,032

fibonacci.js — 8 benchmarks

Benchmark	Interpreted	Bytecode
recursive fib(15)	307	1,085
recursive fib(20)	28	99
recursive fib(15) typed	310	1,432
recursive fib(20) typed	28	129
iterative fib(20) via reduce	11,455	18,450
iterator fib(20)	8,910	10,256
iterator fib(20) via Iterator.from + take	8,490	9,582
iterator fib(20) last value via reduce	7,239	7,479

for-of.js — 7 benchmarks

Benchmark	Interpreted	Bytecode
for...of with 10-element array	44,123	—
for...of with 100-element array	5,190	—
for...of with string (10 chars)	34,805	—
for...of with Set (10 elements)	46,891	—
for...of with Map entries (10 entries)	28,863	10,759
for...of with destructuring	39,300	—
for-await-of with sync array	42,766	—

iterators.js — 20 benchmarks

Benchmark	Interpreted	Bytecode
Iterator.from({next}).toArray() — 20 elements	10,989	10,633
Iterator.from({next}).toArray() — 50 elements	4,701	—
spread pre-wrapped iterator — 20 elements	10,927	10,615
Iterator.from({next}).forEach — 50 elements	3,453	—
Iterator.from({next}).reduce — 50 elements	3,417	3,764
wrap array iterator	55,356	—
wrap plain {next()} object	7,299	7,824
map + toArray (50 elements)	3,049	3,506
filter + toArray (50 elements)	3,206	3,489
take(10) + toArray (50 element source)	15,920	16,070
drop(40) + toArray (50 element source)	4,423	5,605
chained map + filter + take (100 element source)	5,236	6,271
some + every (50 elements)	1,975	2,273
find (50 elements)	4,293	5,279
array.values().map().filter().toArray()	3,617	4,563
array.values().take(5).toArray()	59,938	60,841
array.values().drop(45).toArray()	15,736	23,461
map.entries() chained helpers	5,731	2,627
set.values() chained helpers	8,730	11,039
string iterator map + toArray	6,939	10,069

json.js — 20 benchmarks

Benchmark	Interpreted	Bytecode
parse simple object	155,928	—
parse nested object	96,345	—
parse array of objects	51,140	—
parse large flat object	47,350	—
parse mixed types	70,622	—
stringify simple object	148,619	—
stringify nested object	77,656	—
stringify array of objects	39,107	—
stringify mixed types	67,897	—
reviver doubles numbers	42,002	—
reviver filters properties	37,274	—
reviver on nested object	50,225	—
reviver on array	28,794	—
replacer function doubles numbers	45,299	—
replacer function excludes properties	57,224	—
array replacer (allowlist)	102,951	—
stringify with 2-space indent	76,201	—
stringify with tab indent	79,837	—
parse then stringify	45,081	—
stringify then parse	27,708	—

jsx.jsx — 21 benchmarks

Benchmark	Interpreted	Bytecode
simple element	190,983	—
self-closing element	192,630	—
element with string attribute	156,066	—
element with multiple attributes	134,045	—
element with expression attribute	148,763	—
text child	178,712	—
expression child	173,252	525,509
mixed text and expression	169,474	467,070
nested elements (3 levels)	67,084	198,473
sibling children	49,939	157,200
component element	122,282	379,191
component with children	80,097	245,663
dotted component	109,220	323,970
empty fragment	181,952	638,478
fragment with children	49,187	182,341
spread attributes	97,722	110,967
spread with overrides	84,483	81,079
shorthand props	135,487	407,116
nav bar structure	23,710	77,992
card component tree	27,540	85,317
10 list items via Array.from	12,535	23,277

numbers.js — 11 benchmarks

Benchmark	Interpreted	Bytecode
integer arithmetic	513,437	—
floating point arithmetic	528,537	—
number coercion	177,652	—
toFixed	97,589	—
toString	144,017	—
valueOf	223,490	—
toPrecision	133,563	—
Number.isNaN	292,234	—
Number.isFinite	273,828	—
Number.isInteger	274,241	—
Number.parseInt and parseFloat	228,292	—

objects.js — 7 benchmarks

Benchmark	Interpreted	Bytecode
create simple object	397,294	—
create nested object	199,434	—
create 50 objects via Array.from	8,015	—
property read	557,828	—
Object.keys	262,080	—
Object.entries	88,656	—
spread operator	170,194	—

promises.js — 12 benchmarks

Benchmark	Interpreted	Bytecode
Promise.resolve(value)	447,678	—
new Promise(resolve => resolve(value))	164,319	—
Promise.reject(reason)	447,577	—
resolve + then (1 handler)	139,247	—
resolve + then chain (3 deep)	55,979	—
resolve + then chain (10 deep)	18,273	—
reject + catch + then	75,651	—
resolve + finally + then	66,176	—
Promise.all (5 resolved)	26,305	22,729
Promise.race (5 resolved)	27,774	24,739
Promise.allSettled (5 mixed)	22,093	19,790
Promise.any (5 mixed)	25,188	22,385

strings.js — 11 benchmarks

Benchmark	Interpreted	Bytecode
string concatenation	354,687	—
template literal	386,755	—
string repeat	368,031	—
split and join	129,809	—
indexOf and includes	153,887	—
toUpperCase and toLowerCase	229,410	—
slice and substring	144,874	—
trim operations	166,515	—
replace and replaceAll	194,923	—
startsWith and endsWith	125,964	—
padStart and padEnd	185,732	—

typed-arrays.js — 22 benchmarks

Benchmark	Interpreted	Bytecode
new Int32Array(0)	268,047	120,426
new Int32Array(100)	243,820	116,076
new Int32Array(1000)	154,972	88,080
new Float64Array(100)	234,743	113,673
Int32Array.from([...])	162,834	61,820
Int32Array.of(1, 2, 3, 4, 5)	274,698	231,738
sequential write 100 elements	3,345	6,283
sequential read 100 elements	3,329	6,041
Float64Array write 100 elements	3,021	6,226
fill(42)	45,328	43,702
slice()	184,369	186,657
map(x => x * 2)	7,260	7,758
filter(x => x > 50)	7,479	7,992
reduce (sum)	7,238	6,928
sort()	170,890	151,818
indexOf()	414,399	360,394
reverse()	313,574	275,089
create view over existing buffer	341,444	135,000
subarray()	410,454	355,608
set() from array	537,733	256,345
for-of loop	4,469	8,751
spread into array	15,843	20,879

_{Measured on ubuntu-latest x64. Changes within ±7% are considered insignificant.}

Revert the overly-aggressive !cancelled() conditions added in the previous commit. The build job must fully succeed before any downstream job runs. The test/wasm-test/benchmark/examples jobs were already independent of each other (they only depend on build, not on each other). The fail-fast: false on matrix strategies is the key change that ensures one platform failure within a job doesn't cancel siblings. Co-authored-by: Johannes Stein <frostney@users.noreply.github.com>

coderabbitai

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In @.github/workflows/ci.yml:
- Around line 315-317: The CI configuration lists platforms in the "Package
releases" job using "windows-latest-x86" which contradicts docs and
FPC_TARGET_CPU usage expecting x64; decide whether Windows should be x86 or x64
and make the change consistently: update the platforms array in the "Package
releases" job to use "windows-latest-x64" if you want 64-bit builds (or update
docs/docs/build-system.md and references to FPC_TARGET_CPU to x86 if you intend
32-bit), and ensure the platform string and any mentions of FPC_TARGET_CPU and
the documented matrix all match the chosen architecture.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 60359896-ffc7-4be2-a528-620892eeff17

📥 Commits

Reviewing files that changed from the base of the PR and between b78b6f9 and 0352055.

📒 Files selected for processing (2)

.github/workflows/ci.yml
.github/workflows/pr.yml

coderabbitai

🧹 Nitpick comments (1)

.github/workflows/ci.yml (1)
273-286: Consider adding fail-fast: false for consistency.

The artifacts job matrix doesn't include fail-fast: false while the build, test, benchmark, and examples jobs all do. If one platform's artifact upload fails, it would stop other platforms from completing. This may be intentional if you want all-or-nothing artifact uploads, but for consistency with the PR objective, consider adding fail-fast: false here as well.
♻️ Suggested change
   artifacts:
     needs: [test, wasm-test, benchmark, examples]
     if: github.ref == 'refs/heads/main'
     strategy:
+      fail-fast: false
       matrix:
         os: [ubuntu-latest, macos-latest, windows-latest]
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.

In @.github/workflows/ci.yml around lines 273 - 286, Add fail-fast: false to the
artifacts job's matrix strategy so matrix failures for one platform don't cancel
other platform runs; locate the artifacts job's strategy.matrix block (the
matrix with os/arch/exclude/include shown) and insert fail-fast: false at the
same level as matrix (matching how other jobs set fail-fast) to ensure each
platform runs to completion even if one fails.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Nitpick comments:
In @.github/workflows/ci.yml:
- Around line 273-286: Add fail-fast: false to the artifacts job's matrix
strategy so matrix failures for one platform don't cancel other platform runs;
locate the artifacts job's strategy.matrix block (the matrix with
os/arch/exclude/include shown) and insert fail-fast: false at the same level as
matrix (matching how other jobs set fail-fast) to ensure each platform runs to
completion even if one fails.

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: bca2bde8-6205-49de-b533-26207e524ac0

📥 Commits

Reviewing files that changed from the base of the PR and between 0352055 and 1ee1067.

📒 Files selected for processing (1)

.github/workflows/ci.yml

The Windows build runs on windows-latest (x64 runner) but FPC's i386-win32 toolchain produces 32-bit binaries. Fix artifact and release package names to say x86 instead of x64 using a conditional expression. The matrix structure and runners are unchanged. Update docs/build-system.md and AGENTS.md to reflect the x86 label, document fail-fast: false, job independence, and the wasm-test job. Co-authored-by: Johannes Stein <frostney@users.noreply.github.com>

Build must succeed on all platforms before downstream jobs run. Keeping the default fail-fast: true so a build failure cancels sibling builds early rather than wasting runner minutes. Co-authored-by: Johannes Stein <frostney@users.noreply.github.com>

The wasm-test job now runs on all platforms (ubuntu x64/arm64, macOS x64/arm64, windows x86) matching the test/benchmark/examples jobs. FPC and Node.js are installed per-OS. Co-authored-by: Johannes Stein <frostney@users.noreply.github.com>

frostney marked this pull request as ready for review March 10, 2026 09:30

coderabbitai Bot reviewed Mar 10, 2026

View reviewed changes

Comment thread .github/workflows/ci.yml

cursor Bot force-pushed the cursor/ci-job-independence-9f67 branch from 0352055 to 1ee1067 Compare March 10, 2026 09:37

coderabbitai Bot reviewed Mar 10, 2026

View reviewed changes

cursor Bot force-pushed the cursor/ci-job-independence-9f67 branch from 1ee1067 to 6a4be74 Compare March 10, 2026 09:54

cursoragent and others added 2 commits March 10, 2026 09:56

frostney merged commit 337fd06 into main Mar 10, 2026
7 checks passed

frostney deleted the cursor/ci-job-independence-9f67 branch March 10, 2026 10:07

coderabbitai Bot mentioned this pull request Mar 22, 2026

Add cross-compilation toolchain workflow #93

Merged

coderabbitai Bot mentioned this pull request Apr 8, 2026

Build 32-bit FFI fixture DLL on Windows CI #227

Merged

frostney added the internal Refactoring, CI, tooling, cleanup label Apr 9, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

CI job independence#59

CI job independence#59
frostney merged 5 commits into
mainfrom
cursor/ci-job-independence-9f67

frostney commented Mar 10, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

cursor Bot commented Mar 10, 2026

Uh oh!

coderabbitai Bot commented Mar 10, 2026 •

edited

Loading

Rate limit exceeded

Walkthrough

Changes

Estimated code review effort

Poem

Uh oh!

github-actions Bot commented Mar 10, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot left a comment

Uh oh!

Uh oh!

coderabbitai Bot left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

frostney commented Mar 10, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary by CodeRabbit

Uh oh!

cursor Bot commented Mar 10, 2026

Uh oh!

coderabbitai Bot commented Mar 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Rate limit exceeded

Walkthrough

Changes

Estimated code review effort

Poem

Uh oh!

github-actions Bot commented Mar 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmark Results

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

frostney commented Mar 10, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Mar 10, 2026 •

edited

Loading

github-actions Bot commented Mar 10, 2026 •

edited

Loading