4.1: an iterative canonical serializer

src/adf/editor-normal.test.ts

@@ -2,6 +2,7 @@ import assert from 'node:assert/strict'
 import test from 'node:test'
 
 import type { AdfNode } from './document.ts'
+import type { JsonValue } from '../json-value.ts'
 import { toEditorNormal } from './editor-normal.ts'
 
 test('merges adjacent text nodes carrying identical marks, at every level', () => {
@@ -59,7 +60,7 @@ test('reads an empty attrs object, marks array or content array as the absent ke
   assert.deepEqual(toEditorNormal({ content: [], type: 'doc', version: 1 }), { type: 'doc', version: 1 })
 })
 
-test('normalizes a document nesting far past the levels a recursive walk survives', () => {
+test('normalizes blocks and mark attributes nesting far past the levels a recursive walk survives', () => {
   const levels = 100000
   let node: AdfNode = { content: [], type: 'paragraph' }
   for (let level = 0; level < levels; level += 1) node = { content: [node], type: 'blockquote' }
@@ -68,4 +69,9 @@ test('normalizes a document nesting far past the levels a recursive walk survive
   for (; normal?.content !== undefined; depth += 1) normal = normal.content[0]
   assert.equal(depth, levels)
   assert.deepEqual(normal, { type: 'paragraph' })
+  let deep: JsonValue = 1
+  for (let level = 0; level < 2 * levels; level += 1) deep = [deep]
+  const marks = [{ attrs: { deep }, type: 'textColor' }]
+  const merged = toEditorNormal({ content: [{ content: [{ marks, text: 'a', type: 'text' }, { marks, text: 'b', type: 'text' }], type: 'paragraph' }], type: 'doc', version: 1 })
+  assert.deepEqual(merged.content?.[0]?.content?.map((text) => text.text), ['ab'])
 })

src/canonical-json.test.ts

@@ -1,6 +1,7 @@
 import assert from 'node:assert/strict'
 import test from 'node:test'
 
+import type { JsonValue } from './json-value.ts'
 import { serializeCanonicalJson } from './canonical-json.ts'
 
 test('sorts object keys recursively', () => {
@@ -39,3 +40,15 @@ test('leaves non-ASCII raw', () => {
 test('spells scalars in canonical JSON', () => {
   assert.equal(serializeCanonicalJson([null, true, false, 0, -1.5, 'a"b'], 'compact'), '[null,true,false,0,-1.5,"a\\"b"]')
 })
+
+test('spells a value nesting far past the levels a recursive walk survives', () => {
+  const levels = 200000
+  let array: JsonValue = 1
+  let object: JsonValue = 1
+  for (let level = 0; level < levels; level += 1) {
+    array = [array]
+    object = { a: object }
+  }
+  assert.equal(serializeCanonicalJson(array, 'compact'), `${'['.repeat(levels)}1${']'.repeat(levels)}`)
+  assert.equal(serializeCanonicalJson(object, 'compact'), `${'{"a":'.repeat(levels)}1${'}'.repeat(levels)}`)
+})

src/canonical-json.ts

@@ -2,28 +2,42 @@ import type { JsonValue } from './json-value.ts'
 
 export type JsonSpelling = 'compact' | 'two-space'
 
+type Member = { label: string; value: JsonValue }
+
+type Pending = string | { depth: number; value: JsonValue }
+
 export function serializeCanonicalJson(value: JsonValue, spelling: JsonSpelling): string {
-  return serialize(value, spelling === 'compact' ? '' : '  ', 0)
+  const indent = spelling === 'compact' ? '' : '  '
+  const text: string[] = []
+  const pending: Pending[] = [{ depth: 0, value }]
+  for (let next = pending.pop(); next !== undefined; next = pending.pop()) {
+    if (typeof next === 'string') {
+      text.push(next)
+      continue
+    }
+    const { depth, value: held } = next
+    if (Array.isArray(held)) schedule(pending, '[', held.map((item) => ({ label: '', value: item })), ']', indent, depth)
+    else if (held !== null && typeof held === 'object') schedule(pending, '{', objectMembers(held, indent), '}', indent, depth)
+    else text.push(JSON.stringify(held))
+  }
+  return text.join('')
 }
 
-function serialize(value: JsonValue, indent: string, depth: number): string {
-  if (Array.isArray(value)) {
-    if (value.length === 0) return '[]'
-    const items = value.map((item) => serialize(item, indent, depth + 1))
-    return `[${join(items, indent, depth)}]`
-  }
-  if (value !== null && typeof value === 'object') {
-    const keys = Object.keys(value).sort()
-    if (keys.length === 0) return '{}'
+function objectMembers(value: { [key: string]: JsonValue }, indent: string): Member[] {
   const separator = indent === '' ? ':' : ': '
-    const entries = keys.map((key) => `${JSON.stringify(key)}${separator}${serialize(value[key] ?? null, indent, depth + 1)}`)
-    return `{${join(entries, indent, depth)}}`
-  }
-  return JSON.stringify(value)
+  return Object.keys(value)
+    .sort()
+    .map((key) => ({ label: `${JSON.stringify(key)}${separator}`, value: value[key] ?? null }))
 }
 
-function join(parts: readonly string[], indent: string, depth: number): string {
-  if (indent === '') return parts.join(',')
-  const inner = `\n${indent.repeat(depth + 1)}`
-  return `${inner}${parts.join(`,${inner}`)}\n${indent.repeat(depth)}`
+function schedule(pending: Pending[], open: string, members: readonly Member[], close: string, indent: string, depth: number): void {
+  if (members.length === 0) {
+    pending.push(`${open}${close}`)
+    return
+  }
+  const inner = indent === '' ? '' : `\n${indent.repeat(depth + 1)}`
+  const scheduled: Pending[] = []
+  for (const [index, member] of members.entries()) scheduled.push(`${index === 0 ? open : ','}${inner}${member.label}`, { depth: depth + 1, value: member.value })
+  scheduled.push(indent === '' ? close : `\n${indent.repeat(depth)}${close}`)
+  for (const item of scheduled.reverse()) pending.push(item)
 }