Performance

go-ruby-xslt/xslt is the pure-Go library that rbgo binds for Ruby's Nokogiri::XSLT. This page records the methodology for a comparative benchmark of that module against the reference Ruby runtimes, part of the ecosystem-wide per-module parity suite.

What is measured

The same Ruby script — a Nokogiri::XSLT(ss).transform(doc) / .apply_to(doc) round-trip over a representative stylesheet and source document — is run under every runtime. rbgo's number reflects this pure-Go library doing the work (over go-ruby-nokogiri's DOM/XPath); every other column is that interpreter's own nokogiri / libxslt. So the comparison is the Ruby-visible operation, apples-to-apples across interpreters. The script prints a deterministic checksum and its output is checked byte-identical to the Nokogiri::XSLT (libxslt) oracle before timing.

• Method: best-of-5 wall time (best, not mean, to suppress scheduler noise); single-shot processes, no warm-up beyond the script's own loop.
• Runtimes (planned): ruby +PRISM (MRI, the oracle) and ruby --yjit; jruby (OpenJDK); truffleruby (GraalVM CE Native).
• The benchmark script and harness live in rbgo's repo under bench/modules/ (xslt.rb + run.sh). Reproduce: RBGO=./rbgo TRUFFLE=truffleruby bash bench/modules/run.sh 5.

Results

Measurement pending

Numbers for this module have not yet been measured on a controlled host. Rather than publish fabricated figures, the results table is intentionally omitted until a real best-of-5 run against MRI (and the other runtimes) is recorded here. The methodology above is fixed; only the measured numbers are outstanding.

When the run lands, every row will be a real measured number — the same single-shot ruby file.rb cost as rbgo and MRI, not steady-state JIT figures — with the byte-identical-to-oracle check enforced before timing.