From 55ff93589f7737ec97017f69d5fe0c2bc74d6603 Mon Sep 17 00:00:00 2001
From: He-Pin <hepin1989@gmail.com>
Date: Tue, 12 May 2026 15:29:00 +0800
Subject: [PATCH] perf: render escaped byte strings in one pass

Motivation:
Long JSON strings that contain escape characters used to scan the UTF-8 byte array twice in ByteRenderer: once to find escapes and once to pre-compute the exact escaped length. The large_string_template benchmark spends visible time in this path.

Modification:
Render escaped long strings with one copy/escape pass, growing ByteBuilder incrementally and refreshing its backing array after capacity checks. Add a regression test that compares ByteRenderer output with the char Renderer for long escaped strings including two-byte escapes, six-byte control escapes, and a trailing plain tail.

Result:
The large_string_template JMH target improves from roughly 0.70-0.72 ms/op to roughly 0.58-0.65 ms/op in local runs, while full tests and formatting checks remain green.

References:
bench/resources/cpp_suite/large_string_template.jsonnet
---
 sjsonnet/src/sjsonnet/BaseByteRenderer.scala  | 39 +++++++++----------
 .../test/src/sjsonnet/RendererTests.scala     |  8 ++++
 2 files changed, 26 insertions(+), 21 deletions(-)

diff --git a/sjsonnet/src/sjsonnet/BaseByteRenderer.scala b/sjsonnet/src/sjsonnet/BaseByteRenderer.scala
index ce5f4907..2dbc6428 100644
--- a/sjsonnet/src/sjsonnet/BaseByteRenderer.scala
+++ b/sjsonnet/src/sjsonnet/BaseByteRenderer.scala
@@ -319,9 +319,8 @@ class BaseByteRenderer[T <: java.io.OutputStream](
       arr(pos + 1 + bLen) = '"'.toByte
       elemBuilder.length = pos + bLen + 2
     } else {
-      val escapedLen = escapedStringLength(bytes, bLen, firstEscape)
-      elemBuilder.ensureLength(escapedLen)
-      val arr = elemBuilder.arr
+      elemBuilder.ensureLength(bLen + 2 + (bLen >>> 5))
+      var arr = elemBuilder.arr
       var outPos = elemBuilder.length
       arr(outPos) = '"'.toByte
       outPos += 1
@@ -330,41 +329,39 @@ class BaseByteRenderer[T <: java.io.OutputStream](
       while (escPos >= 0) {
         if (escPos > from) {
           val chunkLen = escPos - from
+          elemBuilder.length = outPos
+          elemBuilder.ensureLength(chunkLen + 6)
+          arr = elemBuilder.arr
+          outPos = elemBuilder.length
           System.arraycopy(bytes, from, arr, outPos, chunkLen)
           outPos += chunkLen
         }
+        elemBuilder.length = outPos
+        elemBuilder.ensureLength(6)
+        arr = elemBuilder.arr
+        outPos = elemBuilder.length
         outPos = escapeByteInline(bytes(escPos) & 0xff, arr, outPos)
         from = escPos + 1
         escPos = if (from < bLen) CharSWAR.findFirstEscapeChar(bytes, from, bLen) else -1
       }
       if (from < bLen) {
         val tailLen = bLen - from
+        elemBuilder.length = outPos
+        elemBuilder.ensureLength(tailLen + 1)
+        arr = elemBuilder.arr
+        outPos = elemBuilder.length
         System.arraycopy(bytes, from, arr, outPos, tailLen)
         outPos += tailLen
       }
+      elemBuilder.length = outPos
+      elemBuilder.ensureLength(1)
+      arr = elemBuilder.arr
+      outPos = elemBuilder.length
       arr(outPos) = '"'.toByte
       elemBuilder.length = outPos + 1
     }
   }
 
-  private def escapedStringLength(bytes: Array[Byte], bLen: Int, firstEscape: Int): Int = {
-    var len = bLen + 2
-    var from = firstEscape
-    var escPos = firstEscape
-    while (escPos >= 0) {
-      len += escapeExtraLength(bytes(escPos) & 0xff)
-      from = escPos + 1
-      escPos = if (from < bLen) CharSWAR.findFirstEscapeChar(bytes, from, bLen) else -1
-    }
-    len
-  }
-
-  @inline private def escapeExtraLength(b: Int): Int =
-    (b: @scala.annotation.switch) match {
-      case '"' | '\\' | '\b' | '\f' | '\n' | '\r' | '\t' => 1
-      case _                                             => 5
-    }
-
   /** Inline JSON escape for one byte that is known to require escaping. */
   @inline private def escapeByteInline(b: Int, arr: Array[Byte], outPos0: Int): Int = {
     val outPos = outPos0
diff --git a/sjsonnet/test/src/sjsonnet/RendererTests.scala b/sjsonnet/test/src/sjsonnet/RendererTests.scala
index 65577f25..f21372b7 100644
--- a/sjsonnet/test/src/sjsonnet/RendererTests.scala
+++ b/sjsonnet/test/src/sjsonnet/RendererTests.scala
@@ -1,5 +1,6 @@
 package sjsonnet
 
+import java.io.ByteArrayOutputStream
 import utest._
 
 object RendererTests extends TestSuite {
@@ -65,6 +66,13 @@ object RendererTests extends TestSuite {
       ujson.transform(ujson.Num(1e15), new Renderer()).toString ==> "1000000000000000"
     }
 
+    test("byteRendererLongEscapedString") {
+      val s = ("abc\n\"\\\t\u0001" * 40) + "tail"
+      val out = new ByteArrayOutputStream
+      ujson.transform(ujson.Str(s), new ByteRenderer(out))
+      out.toString("UTF-8") ==> ujson.transform(ujson.Str(s), new Renderer()).toString
+    }
+
     test("indentZero") {
       // indent=0 should produce newlines but no spaces
       ujson.transform(ujson.Arr(1, 2), new Renderer(indent = 0)).toString ==>