perf: skip ordinary strings in JSON structure checks (#20232)

Co-authored-by: Orca Worker <orca-worker@localhost>

src/shared/json-text-structure-limit.test.ts

@@ -37,4 +37,28 @@ describe('JSON text structure admission', () => {
       })
     ).not.toThrow()
   })
+
+  it.each([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])('handles a quote preceded by %i backslashes', (count) => {
+    const content = `"${'\\'.repeat(count)}"[[]]`
+    const check = () =>
+      assertJsonTextStructureWithinLimits(content, {
+        structuralTokens: 3,
+        nestingDepth: 2
+      })
+    if (count % 2 === 0) {
+      expect(check).toThrowError(new JsonTextStructureCapacityError('structuralTokens', 3))
+    } else {
+      expect(check).not.toThrow()
+    }
+  })
+
+  it('resumes counting after escaped quotes and long string values', () => {
+    const content = JSON.stringify({ value: 'ordinary text [{,}] \\" '.repeat(10_000), next: [] })
+    expect(() =>
+      assertJsonTextStructureWithinLimits(content, { structuralTokens: 7, nestingDepth: 2 })
+    ).not.toThrow()
+    expect(() =>
+      assertJsonTextStructureWithinLimits(content, { structuralTokens: 6, nestingDepth: 2 })
+    ).toThrowError(new JsonTextStructureCapacityError('structuralTokens', 6))
+  })
 })

src/shared/json-text-structure-limit.ts

@@ -25,23 +25,37 @@ export function assertJsonTextStructureWithinLimits(
   assertLimit(limits.nestingDepth)
   let structuralTokens = 0
   let depth = 0
-  let inString = false
-  let escaped = false
-
   for (let index = 0; index < content.length; index += 1) {
     const character = content[index]
-    if (inString) {
-      if (escaped) {
-        escaped = false
-      } else if (character === '\\') {
-        escaped = true
-      } else if (character === '"') {
-        inString = false
-      }
-      continue
-    }
     if (character === '"') {
-      inString = true
+      let quote = content.indexOf('"', index + 1)
+      if (quote !== -1) {
+        // Only an odd backslash run escapes the quote.
+        let backslashes = 0
+        for (let at = quote - 1; at > index && content[at] === '\\'; at -= 1) {
+          backslashes += 1
+        }
+        if (backslashes % 2 !== 0) {
+          // Escape-heavy strings use the linear scan to avoid repeated native searches.
+          let escaped = false
+          for (quote += 1; quote < content.length; quote += 1) {
+            if (escaped) {
+              escaped = false
+            } else if (content[quote] === '\\') {
+              escaped = true
+            } else if (content[quote] === '"') {
+              break
+            }
+          }
+          if (quote === content.length) {
+            quote = -1
+          }
+        }
+      }
+      if (quote === -1) {
+        return
+      }
+      index = quote
       continue
     }
     if (!isStructuralToken(character)) {