philihp / pcg
100%
main: 100%

LAST BUILD BRANCH: renovate/typescript-eslint-monorepo
DEFAULT BRANCH: main
Repo Added
Token zWE5kuofcwF2HQYqPzoejR5HBnSxlGlMH regen
Build 1323 Last
Files 8
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

LAST BUILD ON BRANCH claude/coverage-100
branch: claude/coverage-100
CHANGE BRANCH
x
Sync Branches

coverage: 100.0% (+4.0%) from 96.018%
26259141578

Pull #347

github

claude 
test: reach 100% line/branch/function coverage

Coverage was sitting at 98.30 line / 94.71 branch under the V8
--experimental-test-coverage reporter. The gaps were a mix of real
ones (a defensive dead branch in resolveStreamScheme, an
RXS_M_XS carry path no canonical seed hits) and tooling artifacts —
esbuild's CJS interop helpers were getting source-mapped onto random
lines of every file, making lines and branches that were always
executed report as 0-hit.

Changes:

- Switch the package to "type": "module". This is what eliminates the
  CJS interop noise: ESM-mode tsx no longer wraps modules in
  __export/__copyProps helpers, so V8 only sees our own code. dist
  filenames flip (.js is now the ESM entry, .cjs the CJS one) but the
  exports map handles both consumer modes transparently. Smoke-tested
  with both require() and dynamic import() against the built dist.
- Move the two runtime enums (OutputFnType, StreamScheme) out of
  types.ts into enums.ts so types.ts can be marked pure-types and
  excluded from coverage. Also move the type-only RandomListFn /
  RandomIntPartial1 / RandomListPartial1 interface declarations there.
- Simplify resolveStreamScheme by dropping the structurally unreachable
  INCREMENTERS[resolved] === undefined check; the throw is still
  exercised by the unknown-name test.
- Inline the seeded PCGState object literal in createPcg32 and the
  mul64 helper temporaries, working around per-property phantom-line
  reporting from V8.
- Add outputFns.coverage.test.ts with four targeted cases that exercise
  the rotate-by-zero edge of XSH_RR, the minimum-shift edge of XSH_RS,
  the xor-cancel case of XSL_RR, and the sumLo > 0xffffffff carry path
  of RXS_M_XS — none of which the seed=42/stream=54 canonical seed
  reaches naturally.
- tsconfig: set module=preserve + moduleResolution=bundler so DTS
  emission doesn't require explicit .js extensions on every relative
  import. tsup handles the actual resolution.
- Rename eslint.config.js... (continued)
Pull Request #347: test: reach 100% line/branch/function coverage

100 of 100 branches covered (100.0%)

Branch coverage included in aggregate %.

39 of 39 new or added lines in 7 files covered. (100.0%)

400 of 400 relevant lines covered (100.0%)

9010.32 hits per line

Relevant lines Covered
400 RELEVANT LINES 400 COVERED LINES
9010.32 HITS PER LINE
Source Files on claude/coverage-100

Recent builds

Builds Branch Commit Type Ran Committer Via Coverage
26259141578 claude/coverage-100 test: reach 100% line/branch/function coverage Coverage was sitting at 98.30 line / 94.71 branch under the V8 --experimental-test-coverage reporter. The gaps were a mix of real ones (a defensive dead branch in resolveStreamScheme, an RXS_M_XS car... Pull #347 claude github
100.0
See All Builds (1257)

Badge your Repo: pcg

We detected this repo isn’t badged! Grab the embed code to the right, add it to your repo to show off your code coverage, and when the badge is live hit the refresh button to remove this message.
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

Refresh