diff --git a/.github/workflows/deploy-pages.yml b/.github/workflows/deploy-pages.yml index b17ce26..428f414 100644 --- a/.github/workflows/deploy-pages.yml +++ b/.github/workflows/deploy-pages.yml @@ -46,6 +46,40 @@ jobs: id: pages uses: actions/configure-pages@v6 + # This page states rbgo's ruby/spec total in the present tense in four + # places, and on 2026-10-12 all of them said 22,488 while the gate's own + # file summed to 23,511 — and two of them described the gate as a + # "shrink-only floor of 22,475", a mechanism replaced by a per-file + # baseline in ruby#739. A number and a mechanism, both stale, on the first + # page anybody reads. + # + # The check sums the gate's own per-file baseline and compares it against + # every present-tense claim here. It is the same binary the docs site + # runs, consumed across repositories rather than copied: two readers of + # one file cannot drift, where two transcriptions of one number already + # had. A dated block ("measured ") is exempt, because a dated + # measurement going stale is not an error. + # + # It runs BEFORE hugo so a wrong figure fails the build instead of being + # published, and on pull requests as well as on main. An unreachable + # baseline exits 2 and fails the step: an unreadable baseline finds no + # disagreement either, and that silence would read as agreement. + # + # PINNED TO A TAG, not @main, and that is not tidiness. `@main` is not + # "the current main": it is whatever the module proxy last cached for the + # branch, and the lag is measurable -- + # + # $ go list -m github.com/go-embedded-ruby/docs@main + # …v0.0.0-20261012071659-5a27ddae9f8d # 07:16:59 + # $ gh api …/commits/main --jq .sha + # 3384011416dc… # 07:25:48 + # + # The first version of this step used @main and failed on a branch whose + # figures were right, because the tool it ran was nine minutes old and + # could not yet see this page's phrasing. A tag resolves deterministically. + - name: The stated conformance total matches rbgo's gate + run: go run github.com/go-embedded-ruby/docs/tools/conformancefigure@v0.1.0 -root layouts + - name: Build landing page run: hugo --minify --baseURL "${{ steps.pages.outputs.base_url }}/" diff --git a/layouts/_default/list.html b/layouts/_default/list.html index 1153651..91f6127 100644 --- a/layouts/_default/list.html +++ b/layouts/_default/list.html @@ -70,10 +70,10 @@

C · Application conformance & performance

Runtime conformance — ruby/spec

22,488 examples pass in ruby/spec's language/ + core/ suites, measured 2026-09-26 on main (3e8e3cc). Three independent runs — two through the ratchet at different timeouts, one through a separate sweep — reported the same total.

diff --git a/layouts/index.html b/layouts/index.html index 3268400..a3e5f30 100644 --- a/layouts/index.html +++ b/layouts/index.html @@ -24,7 +24,7 @@

A Ruby you can embed,
written in Go.{{ partial "icons/rocket.html" . }}Get started → {{ partial "icons/graph.html" . }}See the benchmarks

-
22,488 ruby/spec language/+core/ examples passing~100% of real Ruby parsed6 64-bit arches
+
23,511 ruby/spec language/+core/ examples passing~100% of real Ruby parsed6 64-bit arches
rbgo — zsh
@@ -59,7 +59,7 @@

Everything a Ruby runtime gives you — as a Go library.

🧩

Embeddable

A real object model — classes, mixins, method_missing, metaprogramming, Fiber. The Go API is deliberately small today: ruby.Run(src, out) runs a program on a fresh VM and writes its output to one writer. Calling a Ruby method from Go is not exposed yet.

📦

One static binary

CGO=0 by default. Cross-compiles to amd64, arm64, riscv64, loong64, ppc64le and s390x.

⚡

AOT for integer kernels

Ahead-of-time compilation of integer-bound methods to unboxed Go. Measured on this project’s own micro-benchmarks (darwin/arm64, whole-process wall time): 8.7× MRI+YJIT on the tight-loop case, 2.0× on recursive fib(35). Only methods the compiler can lower benefit.

-
✅

Conformance, measured

22,488 ruby/spec language/ + core/ examples pass, gated in CI by a shrink-only floor of 22,475 — the floor is what CI refuses to fall below, not the score. Only language/ and core/ are swept: library/, optional/ and security/ are not run, so this is not a percentage of Ruby. Front-end: ~100% of valid Ruby parsed across Rails + Puppet (2026-06).

+
✅

Conformance, measured

23,511 ruby/spec language/ + core/ examples pass, gated in CI file by file: no spec file may pass fewer examples than its recorded row, so a regression is named rather than absorbed in a total. Only language/ and core/ are swept: library/, optional/ and security/ are not run, so this is not a percentage of Ruby. Front-end: ~100% of valid Ruby parsed across Rails + Puppet (2026-06).

🔒

Closed-world builds

rbgo build --closed bakes the program to bytecode and drops the front-end for a smaller binary.

🌿

A real ecosystem

Built from the go-ruby-* family — each a 100%-coverage, 6-arch stdlib module, bound natively.

@@ -68,7 +68,7 @@

Everything a Ruby runtime gives you — as a Go library.

8.7×
rbgo+AOT vs MRI+YJIT, tight-loop benchmark (2.0× on fib(35))
-
22,488
ruby/spec language/+core/ examples passing · CI floor 22,475
+
23,511
ruby/spec language/+core/ examples passing · gated per file, not by a total
178
go-ruby-* modules bound into rbgo
CGO=0
across 6 arches · 3 OSes