From 93f7d8a816f3189e79e2bad340da77bf281ab61f Mon Sep 17 00:00:00 2001 From: Yuansheng Wang Date: Mon, 8 Jun 2026 01:49:25 +0800 Subject: [PATCH] docs: fix column-count mismatch in benchmark separator rows The x86_64 throughput table had an extra separator cell (8 columns vs 7 in the header), and the modify+encode table header alignment was inconsistent (first column missing the colon suffix). Fix both so the Markdown tables render correctly. Fixes #162 --- docs/benchmarks.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/benchmarks.md b/docs/benchmarks.md index 9081a74..9235631 100644 --- a/docs/benchmarks.md +++ b/docs/benchmarks.md @@ -108,7 +108,7 @@ Numbers below come from one such run. Each row is "parse + access request fields" on the named payload. | Scenario | Size | cjson | simdjson | `qjson.parse` | `qjson.decode + access content` | `qjson.decode + qjson.encode` | -|---|---|---:|---:|---:|---:|---:|---:| +|---|---|---:|---:|---:|---:|---:| | small | 2.1 KB | 92,716 | 102,602 | 128,005 | 125,815 | 260,322 | | medium | 60.4 KB | 9,007 | 82,699 | 116,198 | 219,491 | 141,563 | | github-100k | 100 KB | 1,834 | 1,909 | 4,591 | 5,643 | 6,207 | @@ -151,7 +151,7 @@ re-encode pipeline. Numbers below come from a 3-round per-scenario fresh-process run on x86_64 Linux (AMD EPYC Rome, Zen 2). | Scenario | modify top + encode | add field + encode | modify nested + encode | -|---|---|---:|---:|---:| +|---|---:|---:|---:| | small (2 KB) | 58,242 | 58,190 | 43,003 | | medium (60 KB) | 37,498 | 45,364 | 134,590 | | github-100k | 4,419 | 3,964 | 4,359 |