Read the container blocks, and part a nested list the tight spelling would swallow

src/markdown/emit/adf-to-markdown.test.ts

@@ -159,8 +159,8 @@ test('breaks a mark run at the node it carries', () => {
 test('refuses a carried node nested deeper than the emitter carries', () => {
   let node: AdfNode = { type: 'blockCard' }
   for (let depth = 0; depth < 600; depth += 1) node = { content: [node], type: 'blockCard' }
-  assert.equal(code(adfToMarkdown(document(node))), 'unsupported-node-shape')
-  assert.equal(code(adfToMarkdown(document(paragraph(node)))), 'unsupported-node-shape')
+  assert.equal(code(adfToMarkdown(document(node))), 'unsupported-nesting-depth')
+  assert.equal(code(adfToMarkdown(document(paragraph(node)))), 'unsupported-nesting-depth')
 })
 
 test('refuses a node whose content model the canonical form cannot emit', () => {
@@ -261,19 +261,19 @@ test('refuses a node carrying one mark type twice', () => {
   assert.equal(code(adfToMarkdown(document(paragraph({ marks: [em, em], text: 'x', type: 'text' })))), 'unsupported-node-shape')
 })
 
-test('refuses a nested list the tight spelling would swallow', () => {
+test('parts a nested list the tight spelling would swallow from the block above it', () => {
   const item = (...content: AdfNode[]): AdfNode => ({ content, type: 'listItem' })
   const text = (value: string): AdfNode => ({ content: [{ text: value, type: 'text' }], type: 'paragraph' })
   const outer = (...content: AdfNode[]): AdfDocument => document({ content: [item(...content)], type: 'bulletList' })
   const ordered: AdfNode = { attrs: { order: 2 }, content: [item(text('b'))], type: 'orderedList' }
-  assert.equal(code(adfToMarkdown(outer(text('a'), ordered))), 'unspellable-line-start')
-  assert.equal(code(adfToMarkdown(outer(text('a'), { content: [item()], type: 'bulletList' }))), 'unspellable-line-start')
+  assert.equal(markdown(adfToMarkdown(outer(text('a'), ordered))), '- a\n\n  2. b\n')
+  assert.equal(markdown(adfToMarkdown(outer(text('a'), { content: [item()], type: 'bulletList' }))), '- a\n\n  -\n')
   assert.equal(markdown(adfToMarkdown(outer(text('a'), { content: [item(text('b'))], type: 'bulletList' }))), '- a\n  - b\n')
 })
 
 test('refuses marks and attributes nested deeper than the emitter carries', () => {
   const marks: AdfMark[] = Array.from({ length: 600 }, (_, index) => ({ type: index % 2 === 0 ? 'em' : 'strong' }))
-  assert.equal(code(adfToMarkdown(document(paragraph({ marks, text: 'x', type: 'text' })))), 'unsupported-node-shape')
+  assert.equal(code(adfToMarkdown(document(paragraph({ marks, text: 'x', type: 'text' })))), 'unsupported-nesting-depth')
   let attrs: AdfMark['attrs'] = { depth: 'x' }
   for (let depth = 0; depth < 600; depth += 1) attrs = { depth: attrs }
   assert.equal(code(adfToMarkdown(document(paragraph({ marks: [{ attrs, type: 'em' }], text: 'x', type: 'text' })))), 'not-an-adf-document')
@@ -380,7 +380,7 @@ test('spells one code span over a run of code-marked nodes', () => {
 test('refuses a document nested deeper than the emitter carries', () => {
   let node: AdfNode = paragraph({ text: 'x', type: 'text' })
   for (let depth = 0; depth < 600; depth += 1) node = { content: [node], type: 'blockquote' }
-  assert.equal(code(adfToMarkdown(document(node))), 'unsupported-node-shape')
+  assert.equal(code(adfToMarkdown(document(node))), 'unsupported-nesting-depth')
 })
 
 test('emits an empty list item without trailing whitespace', () => {

src/markdown/emit/adf-to-markdown.ts

@@ -31,7 +31,7 @@ export function adfToMarkdown(document: AdfDocument): Result<string> {
 }
 
 function emitBlocks(nodes: readonly AdfNode[], container: BlockContainer, path: ConvertErrorPath, depth: number): Result<EmittedBody> {
-  if (depth > largestNesting) return failure('unsupported-node-shape', `the document nests deeper than the ${largestNesting} levels the emitter carries`, path)
+  if (depth > largestNesting) return failure('unsupported-nesting-depth', `the document nests deeper than the ${largestNesting} levels the emitter carries`, path)
   const blocks: PlacedBlock[] = []
   for (const [index, node] of nodes.entries()) {
     const nodePath = [...path, 'content', index]
@@ -60,12 +60,7 @@ function separationBetween(previous: PlacedBlock, next: PlacedBlock, container:
     if (previous.node.type === next.node.type) {
       return failure('unspellable-adjacent-lists', `two adjacent ${next.node.type} nodes read back as one list`, next.path)
     }
-    if (container === 'list-item') {
-      if (!interruptsParagraph(next.node)) {
-        return failure('unspellable-line-start', `a ${next.node.type} that cannot interrupt the block above it has no tight spelling`, next.path)
-      }
-      return success('\n')
-    }
+    if (container === 'list-item') return success(interruptsParagraph(next.node) ? '\n' : '\n\n')
   }
   if (container !== 'directive' || plainPair) return success('\n\n')
   if (previous.spelling === 'directive' && next.spelling === 'directive') return success('\n')

src/markdown/emit/inline-line.ts

@@ -140,7 +140,7 @@ function refuseContentAndText(node: AdfNode, path: ConvertErrorPath): Result<nul
 
 function emitRun(nodes: readonly AdfNode[], depth: number, firstIndex: number, context: InlineContext): Result<Emission> {
   if (depth > largestNesting) {
-    return failure('unsupported-node-shape', `the marks nest deeper than the ${largestNesting} levels the emitter carries`, context.path)
+    return failure('unsupported-nesting-depth', `the marks nest deeper than the ${largestNesting} levels the emitter carries`, context.path)
   }
   const runs = inlineRuns(nodes, depth, firstIndex, context.carried)
   const segments: InlineSegment[] = []

src/markdown/opaque-carry.ts

@@ -23,7 +23,7 @@ export function carriedInline(node: AdfNode, path: ConvertErrorPath): Result<str
 
 function carriedJson(node: AdfNode, spelling: JsonSpelling, path: ConvertErrorPath): Result<string> {
   if (!isJsonValue(node)) {
-    return failure('unsupported-node-shape', `a carried node's JSON nests deeper than the ${largestNesting} levels the emitter carries`, path)
+    return failure('unsupported-nesting-depth', `a carried node's JSON nests deeper than the ${largestNesting} levels the emitter carries`, path)
   }
   return success(serializeCanonicalJson(node, spelling))
 }

src/markdown/parse/blocks.test.ts

@@ -28,6 +28,10 @@ test('keeps the link reference definitions a paragraph gives up, the first of a
     ['a', { destination: '/url\\' }],
     ['b', { destination: '/b' }],
   ])
+  assert.deepEqual(definitions('> [a]: /url\n\n- [b]: /other\n'), [
+    ['a', { destination: '/url' }],
+    ['b', { destination: '/other' }],
+  ])
   assert.deepEqual(definitions('[\u00a0a]: /one\n[a]: /two\n'), [
     ['\u00a0a', { destination: '/one' }],
     ['a', { destination: '/two' }],

src/markdown/parse/blocks.ts

@@ -1,80 +1,277 @@
 import type { LinkDefinition } from './link-reference-definitions.ts'
-import type { OpenHtmlBlock } from './html-blocks.ts'
 import { atxHeading, claimsDirectiveLine, claimsPipeLine, closingCodeFence, isThematicBreak, openingCodeFence, setextHeadingLevel } from '../commonmark-grammar.ts'
 import { openingHtmlBlock } from './html-blocks.ts'
 import { readLinkDefinitions } from './link-reference-definitions.ts'
 
 export type ClaimedConstruct = 'directive' | 'pipe-table'
 
-export type LeafBlock =
+export type Block =
+  | { blocks: Block[]; kind: 'blockquote' }
   | { construct: ClaimedConstruct; kind: 'claim' }
   | { construct: string; kind: 'html' }
+  | { items: Block[][]; kind: 'bulletList' }
+  | { items: Block[][]; kind: 'orderedList'; start: number }
   | { kind: 'code'; language: string; text: string }
   | { kind: 'heading'; level: number; text: string }
   | { kind: 'paragraph'; text: string }
   | { kind: 'rule' }
 
-export type ParsedBlocks = { blocks: LeafBlock[]; definitions: Map<string, LinkDefinition> }
+export type ParsedBlocks = { blocks: Block[]; definitions: Map<string, LinkDefinition> }
 
-type Walk = ParsedBlocks & { paragraph: string[] }
+type ListBlock = Extract<Block, { items: Block[][] }>
+
+type OpenContainer = Extract<Block, { kind: 'blockquote' }> | { blocks: Block[]; indentation: number; kind: 'item'; list: ListBlock; marker: string }
+
+type OpenLeaf =
+  | { closer: RegExp | undefined; construct: string; kind: 'html' }
+  | { held: string[]; kind: 'indented-code'; lines: string[] }
+  | { indentation: number; info: string; kind: 'fenced-code'; lines: string[]; marker: string }
+  | { kind: 'paragraph'; lines: string[] }
+
+type ContainerStart = { kind: 'blockquote'; rest: string } | { fresh: boolean; indentation: number; kind: 'item'; list: ListBlock; marker: string; rest: string }
+
+type ItemMarker = { list: ListBlock; marker: string; width: number }
+
+type Walk = ParsedBlocks & { leaf: OpenLeaf | undefined; stack: OpenContainer[] }
 
 const blankLine = /^[ \t]*$/
+const bulletMarker = /^[-*+](?=[ \t]|$)/
+const orderedMarker = /^(\d{1,9})([.)])(?=[ \t]|$)/
 const indentedCodeColumns = 4
 const largestOpenerIndentation = 3
 const tabStop = 4
 
 export function parseBlocks(markdown: string): ParsedBlocks {
-  const lines = normalizeInput(markdown).split('\n')
-  const walk: Walk = { blocks: [], definitions: new Map(), paragraph: [] }
-  let index = 0
-  while (index < lines.length) {
-    const line = lines[index] ?? ''
-    if (blankLine.test(line)) {
-      closeParagraph(walk)
-      index += 1
-      continue
-    }
-    if (leadingColumns(line) >= indentedCodeColumns && walk.paragraph.length === 0) {
-      index = readIndentedCode(walk, lines, index)
-      continue
-    }
-    const opened = openBlock(walk, lines, index, line)
-    if (opened !== undefined) {
-      index = opened
-      continue
-    }
-    walk.paragraph.push(line.replace(/^[ \t]+/, ''))
-    index += 1
-  }
-  closeParagraph(walk)
+  const walk: Walk = { blocks: [], definitions: new Map(), leaf: undefined, stack: [] }
+  for (const line of normalizeInput(markdown).split('\n')) readLine(walk, line)
+  closeLeaf(walk)
   return { blocks: walk.blocks, definitions: walk.definitions }
 }
 
-function openBlock(walk: Walk, lines: readonly string[], index: number, line: string): number | undefined {
+function readLine(walk: Walk, line: string): void {
+  const matched = matchContainers(walk, line)
+  const paragraphOpen = matched.depth === walk.stack.length && walk.leaf?.kind === 'paragraph'
+  const opened = openContainers(walk, matched.rest, paragraphOpen, matched.depth)
+  if (!opened.opened && matched.depth < walk.stack.length) {
+    if (continuesLazily(walk, opened.rest)) {
+      appendParagraph(walk, opened.rest)
+      return
+    }
+    closeContainers(walk, matched.depth)
+  }
+  readBlockLine(walk, opened.rest)
+}
+
+function matchContainers(walk: Walk, line: string): { depth: number; rest: string } {
+  let depth = 0
+  let rest = line
+  for (const container of walk.stack) {
+    const next = continuesContainer(walk, container, rest)
+    if (next === undefined) break
+    depth += 1
+    rest = next
+  }
+  return { depth, rest }
+}
+
+function continuesContainer(walk: Walk, container: OpenContainer, line: string): string | undefined {
+  if (container.kind === 'blockquote') {
+    const opener = removeColumns(line, largestOpenerIndentation)
+    return opener.startsWith('>') ? removeColumns(opener.slice(1), 1) : undefined
+  }
+  // A list item begins with at most one blank line: an empty one gives the second up.
+  if (blankLine.test(line)) return container.blocks.length === 0 && walk.leaf === undefined ? undefined : ''
+  return leadingColumns(line) < container.indentation ? undefined : removeColumns(line, container.indentation)
+}
+
+function openContainers(walk: Walk, line: string, paragraphOpen: boolean, depth: number): { opened: boolean; rest: string } {
+  let opened = false
+  let rest = line
+  while (leadingColumns(rest) < indentedCodeColumns) {
+    const start = containerStart(rest, opened ? false : paragraphOpen, opened ? undefined : walk.stack[depth])
+    if (start === undefined) break
+    if (!opened) closeContainers(walk, depth)
+    opened = true
+    openContainer(walk, start)
+    rest = start.rest
+  }
+  return { opened, rest }
+}
+
+function containerStart(line: string, paragraphOpen: boolean, enclosing: OpenContainer | undefined): ContainerStart | undefined {
+  const opener = removeColumns(line, largestOpenerIndentation)
+  if (opener.startsWith('>')) return { kind: 'blockquote', rest: removeColumns(opener.slice(1), 1) }
+  if (isThematicBreak(opener) || (paragraphOpen && setextHeadingLevel(opener) !== undefined)) return undefined
+  const marker = itemMarker(opener)
+  if (marker === undefined) return undefined
+  const after = opener.slice(marker.width)
+  const blank = blankLine.test(after)
+  if (paragraphOpen && (blank || (marker.list.kind === 'orderedList' && marker.list.start !== 1))) return undefined
+  const spaces = leadingColumns(after)
+  const padding = blank || spaces > indentedCodeColumns ? 1 : spaces
+  const continued = enclosing?.kind === 'item' && enclosing.list.kind === marker.list.kind && enclosing.marker === marker.marker
+  return {
+    fresh: !continued,
+    indentation: leadingColumns(line) + marker.width + padding,
+    kind: 'item',
+    list: continued ? enclosing.list : marker.list,
+    marker: marker.marker,
+    rest: blank ? '' : removeColumns(after, padding),
+  }
+}
+
+function itemMarker(opener: string): ItemMarker | undefined {
+  const ordered = orderedMarker.exec(opener)
+  if (ordered !== null) {
+    const digits = ordered[1] ?? ''
+    const delimiter = ordered[2] ?? ''
+    return { list: { items: [], kind: 'orderedList', start: Number(digits) }, marker: delimiter, width: digits.length + 1 }
+  }
+  const bullet = bulletMarker.exec(opener)?.[0]
+  return bullet === undefined ? undefined : { list: { items: [], kind: 'bulletList' }, marker: bullet, width: 1 }
+}
+
+function openContainer(walk: Walk, start: ContainerStart): void {
+  const blocks: Block[] = []
+  if (start.kind === 'blockquote') {
+    const blockquote: OpenContainer = { blocks, kind: 'blockquote' }
+    currentBlocks(walk).push(blockquote)
+    walk.stack.push(blockquote)
+    return
+  }
+  if (start.fresh) currentBlocks(walk).push(start.list)
+  start.list.items.push(blocks)
+  walk.stack.push({ blocks, indentation: start.indentation, kind: 'item', list: start.list, marker: start.marker })
+}
+
+function closeContainers(walk: Walk, depth: number): void {
+  closeLeaf(walk)
+  walk.stack.length = depth
+}
+
+// A claimed line ends the lazy continuation CommonMark would fold it into (spec/flavour.md).
+function continuesLazily(walk: Walk, line: string): boolean {
+  if (walk.leaf?.kind !== 'paragraph' || blankLine.test(line)) return false
+  if (leadingColumns(line) >= indentedCodeColumns) return true
+  const opener = removeColumns(line, largestOpenerIndentation)
+  if (claimedConstruct(opener) !== undefined || isThematicBreak(opener)) return false
+  return atxHeading(opener) === undefined && openingCodeFence(opener) === undefined && openingHtmlBlock(opener, false) === undefined
+}
+
+function readBlockLine(walk: Walk, line: string): void {
+  const leaf = walk.leaf
+  if (leaf?.kind === 'fenced-code') {
+    if (closingCodeFence(removeColumns(line, largestOpenerIndentation), leaf.marker)) closeLeaf(walk)
+    else leaf.lines.push(removeColumns(line, leaf.indentation))
+    return
+  }
+  if (leaf?.kind === 'html') {
+    if (leaf.closer === undefined ? blankLine.test(line) : leaf.closer.test(line)) closeLeaf(walk)
+    return
+  }
+  if (leaf?.kind === 'indented-code') {
+    if (readIndentedCodeLine(leaf, line)) return
+    closeLeaf(walk)
+  }
+  if (blankLine.test(line)) {
+    closeLeaf(walk)
+    return
+  }
+  if (walk.leaf === undefined && leadingColumns(line) >= indentedCodeColumns) {
+    walk.leaf = { held: [], kind: 'indented-code', lines: [removeColumns(line, indentedCodeColumns)] }
+    return
+  }
+  openLeaf(walk, line)
+}
+
+function readIndentedCodeLine(leaf: Extract<OpenLeaf, { kind: 'indented-code' }>, line: string): boolean {
+  if (blankLine.test(line)) {
+    leaf.held.push(removeColumns(line, indentedCodeColumns))
+    return true
+  }
+  if (leadingColumns(line) < indentedCodeColumns) return false
+  leaf.lines.push(...leaf.held, removeColumns(line, indentedCodeColumns))
+  leaf.held.length = 0
+  return true
+}
+
+function openLeaf(walk: Walk, line: string): void {
   const opener = removeColumns(line, largestOpenerIndentation)
   const claimed = claimedConstruct(opener)
   if (claimed !== undefined) {
-    closeParagraph(walk)
-    walk.blocks.push({ construct: claimed, kind: 'claim' })
-    return index + 1
+    closeLeaf(walk)
+    currentBlocks(walk).push({ construct: claimed, kind: 'claim' })
+    return
   }
-  if (readLineBlock(walk, opener)) return index + 1
+  if (readLineBlock(walk, opener)) return
   const fence = openingCodeFence(opener)
   if (fence !== undefined) {
-    closeParagraph(walk)
-    return readFencedCode(walk, lines, index + 1, fence, leadingColumns(line))
+    closeLeaf(walk)
+    walk.leaf = { indentation: leadingColumns(line), info: fence.info, kind: 'fenced-code', lines: [], marker: fence.marker }
+    return
   }
-  const html = openingHtmlBlock(opener, walk.paragraph.length > 0)
-  if (html === undefined) return undefined
-  closeParagraph(walk)
-  return readHtmlBlock(walk, lines, index, html)
+  const html = openingHtmlBlock(opener, walk.leaf?.kind === 'paragraph')
+  if (html === undefined) {
+    appendParagraph(walk, line)
+    return
+  }
+  closeLeaf(walk)
+  walk.leaf = { closer: html.closer, construct: html.construct, kind: 'html' }
+  if (html.closer?.test(line) === true) closeLeaf(walk)
 }
 
-function normalizeInput(markdown: string): string {
-  return markdown
-    .replace(/\r\n?/g, '\n')
-    .replaceAll('\u0000', '\ufffd')
-    .replace(/\n$/, '')
+// A setext underline over a paragraph the definitions emptied is no heading: it opens the next block.
+function readLineBlock(walk: Walk, opener: string): boolean {
+  const level = walk.leaf?.kind === 'paragraph' ? setextHeadingLevel(opener) : undefined
+  if (level !== undefined) {
+    const text = takeParagraph(walk)
+    if (text !== undefined) {
+      currentBlocks(walk).push({ kind: 'heading', level, text })
+      return true
+    }
+  }
+  if (isThematicBreak(opener)) {
+    closeLeaf(walk)
+    currentBlocks(walk).push({ kind: 'rule' })
+    return true
+  }
+  const heading = atxHeading(opener)
+  if (heading === undefined) return false
+  closeLeaf(walk)
+  currentBlocks(walk).push({ kind: 'heading', level: heading.level, text: heading.text })
+  return true
+}
+
+function appendParagraph(walk: Walk, line: string): void {
+  const leaf = walk.leaf
+  const text = line.replace(/^[ \t]+/, '')
+  if (leaf?.kind === 'paragraph') leaf.lines.push(text)
+  else walk.leaf = { kind: 'paragraph', lines: [text] }
+}
+
+function closeLeaf(walk: Walk): void {
+  const leaf = walk.leaf
+  if (leaf === undefined) return
+  if (leaf.kind === 'paragraph') {
+    const text = takeParagraph(walk)
+    if (text !== undefined) currentBlocks(walk).push({ kind: 'paragraph', text })
+    return
+  }
+  walk.leaf = undefined
+  if (leaf.kind === 'html') currentBlocks(walk).push({ construct: leaf.construct, kind: 'html' })
+  else currentBlocks(walk).push({ kind: 'code', language: leaf.kind === 'fenced-code' ? leaf.info : '', text: leaf.lines.join('\n') })
+}
+
+function takeParagraph(walk: Walk): string | undefined {
+  const leaf = walk.leaf
+  if (leaf?.kind !== 'paragraph') return undefined
+  walk.leaf = undefined
+  const text = readLinkDefinitions(walk.definitions, leaf.lines.join('\n'))
+  return text === '' ? undefined : text
+}
+
+function currentBlocks(walk: Walk): Block[] {
+  return walk.stack.at(-1)?.blocks ?? walk.blocks
 }
 
 function claimedConstruct(opener: string): ClaimedConstruct | undefined {
@@ -82,83 +279,11 @@ function claimedConstruct(opener: string): ClaimedConstruct | undefined {
   return claimsPipeLine(opener) ? 'pipe-table' : undefined
 }
 
-// A setext underline over a paragraph the definitions emptied is no heading: it opens the next block.
-function readLineBlock(walk: Walk, opener: string): boolean {
-  const level = walk.paragraph.length === 0 ? undefined : setextHeadingLevel(opener)
-  if (level !== undefined) {
-    const text = takeParagraph(walk)
-    if (text !== undefined) {
-      walk.blocks.push({ kind: 'heading', level, text })
-      return true
-    }
-  }
-  if (isThematicBreak(opener)) {
-    closeParagraph(walk)
-    walk.blocks.push({ kind: 'rule' })
-    return true
-  }
-  const heading = atxHeading(opener)
-  if (heading === undefined) return false
-  closeParagraph(walk)
-  walk.blocks.push({ kind: 'heading', level: heading.level, text: heading.text })
-  return true
-}
-
-function readFencedCode(walk: Walk, lines: readonly string[], start: number, fence: { info: string; marker: string }, indentation: number): number {
-  const collected: string[] = []
-  let index = start
-  while (index < lines.length) {
-    const line = lines[index] ?? ''
-    index += 1
-    if (closingCodeFence(removeColumns(line, largestOpenerIndentation), fence.marker)) break
-    collected.push(removeColumns(line, indentation))
-  }
-  walk.blocks.push({ kind: 'code', language: fence.info, text: collected.join('\n') })
-  return index
-}
-
-function readIndentedCode(walk: Walk, lines: readonly string[], start: number): number {
-  const collected: string[] = []
-  const held: string[] = []
-  let index = start
-  let end = start
-  while (index < lines.length) {
-    const line = lines[index] ?? ''
-    index += 1
-    if (blankLine.test(line)) {
-      held.push(removeColumns(line, indentedCodeColumns))
-      continue
-    }
-    if (leadingColumns(line) < indentedCodeColumns) break
-    collected.push(...held, removeColumns(line, indentedCodeColumns))
-    held.length = 0
-    end = index
-  }
-  walk.blocks.push({ kind: 'code', language: '', text: collected.join('\n') })
-  return end
-}
-
-function readHtmlBlock(walk: Walk, lines: readonly string[], start: number, html: OpenHtmlBlock): number {
-  let index = start
-  while (index < lines.length) {
-    const line = lines[index] ?? ''
-    if (html.closer === undefined && blankLine.test(line)) break
-    index += 1
-    if (html.closer !== undefined && html.closer.test(line)) break
-  }
-  walk.blocks.push({ construct: html.construct, kind: 'html' })
-  return index
-}
-
-function closeParagraph(walk: Walk): void {
-  const text = takeParagraph(walk)
-  if (text !== undefined) walk.blocks.push({ kind: 'paragraph', text })
-}
-
-function takeParagraph(walk: Walk): string | undefined {
-  const text = readLinkDefinitions(walk.definitions, walk.paragraph.join('\n'))
-  walk.paragraph = []
-  return text === '' ? undefined : text
+function normalizeInput(markdown: string): string {
+  return markdown
+    .replace(/\r\n?/g, '\n')
+    .replaceAll('\u0000', '\ufffd')
+    .replace(/\n$/, '')
 }
 
 function leadingColumns(line: string): number {

src/markdown/parse/markdown-to-adf.test.ts

@@ -25,6 +25,22 @@ function paragraph(value: string): AdfNode {
   return { content: [text(value)], type: 'paragraph' }
 }
 
+function item(...content: AdfNode[]): AdfNode {
+  return content.length === 0 ? { type: 'listItem' } : { content, type: 'listItem' }
+}
+
+function bulletList(...content: AdfNode[]): AdfNode {
+  return { content, type: 'bulletList' }
+}
+
+function orderedList(order: number, ...content: AdfNode[]): AdfNode {
+  return { attrs: { order }, content, type: 'orderedList' }
+}
+
+function quote(...content: AdfNode[]): AdfNode {
+  return content.length === 0 ? { type: 'blockquote' } : { content, type: 'blockquote' }
+}
+
 test('builds an empty document from input holding no block', () => {
   assert.deepEqual(markdownToAdf(''), { ok: true, value: { type: 'doc', version: 1 } })
   assert.deepEqual(content(markdownToAdf('\n \n\t\n')), [])
@@ -140,3 +156,72 @@ test('normalizes the line endings and the null character CommonMark replaces', (
   assert.deepEqual(content(markdownToAdf('```\r\nx\r\n```\r\n')), [{ content: [text('x')], type: 'codeBlock' }])
   assert.deepEqual(content(markdownToAdf('a\u0000b\n')), [paragraph('a\ufffdb')])
 })
+
+test('reads a blockquote and the blocks its prefix carries', () => {
+  assert.deepEqual(content(markdownToAdf('> Ship it.\n>\n> Then tell them.\n')), [quote(paragraph('Ship it.'), paragraph('Then tell them.'))])
+  assert.deepEqual(content(markdownToAdf('>Ship it.\n')), [quote(paragraph('Ship it.'))])
+  assert.deepEqual(content(markdownToAdf('   > > Deep.\n')), [quote(quote(paragraph('Deep.')))])
+  assert.deepEqual(content(markdownToAdf('>\n')), [quote()])
+  assert.deepEqual(content(markdownToAdf('> One.\n\n> Two.\n')), [quote(paragraph('One.')), quote(paragraph('Two.'))])
+  assert.deepEqual(content(markdownToAdf('Part.\n> Ship it.\n')), [paragraph('Part.'), quote(paragraph('Ship it.'))])
+  assert.deepEqual(content(markdownToAdf('    > Code.\n')), [{ content: [text('> Code.')], type: 'codeBlock' }])
+})
+
+test('reads a bullet list, the marker width setting the continuation', () => {
+  assert.deepEqual(content(markdownToAdf('- Bolt M8\n- Nut M8\n')), [bulletList(item(paragraph('Bolt M8')), item(paragraph('Nut M8')))])
+  assert.deepEqual(content(markdownToAdf('- Washer M8\n  - Fibre\n')), [bulletList(item(paragraph('Washer M8'), bulletList(item(paragraph('Fibre')))))])
+  assert.deepEqual(content(markdownToAdf('-\n')), [bulletList(item())])
+  assert.deepEqual(content(markdownToAdf('- One\n\n  Two.\n')), [bulletList(item(paragraph('One'), paragraph('Two.')))])
+  assert.deepEqual(content(markdownToAdf('-     Code.\n')), [bulletList(item({ content: [text('Code.')], type: 'codeBlock' }))])
+  assert.deepEqual(content(markdownToAdf('- a\n* b\n')), [bulletList(item(paragraph('a'))), bulletList(item(paragraph('b')))])
+  assert.deepEqual(content(markdownToAdf('-\n\n  Part.\n')), [bulletList(item()), paragraph('Part.')])
+})
+
+test('reads an ordered list, its first marker the order attribute', () => {
+  assert.deepEqual(content(markdownToAdf('9. Bolt M8\n10. Nut M8\n')), [orderedList(9, item(paragraph('Bolt M8')), item(paragraph('Nut M8')))])
+  assert.deepEqual(content(markdownToAdf('1) Loosen the clamp\n')), [orderedList(1, item(paragraph('Loosen the clamp')))])
+  assert.deepEqual(content(markdownToAdf('1. a\n1) b\n')), [orderedList(1, item(paragraph('a'))), orderedList(1, item(paragraph('b')))])
+  assert.deepEqual(content(markdownToAdf('0. Zero\n')), [orderedList(0, item(paragraph('Zero')))])
+})
+
+test('drops the tightness ADF does not record', () => {
+  assert.deepEqual(content(markdownToAdf('- a\n\n- b\n')), [bulletList(item(paragraph('a')), item(paragraph('b')))])
+  assert.deepEqual(content(markdownToAdf('- a\n\n  2. b\n')), [bulletList(item(paragraph('a'), orderedList(2, item(paragraph('b')))))])
+})
+
+test('opens a list beside a paragraph only where the marker interrupts it', () => {
+  assert.deepEqual(content(markdownToAdf('Part.\n- a\n')), [paragraph('Part.'), bulletList(item(paragraph('a')))])
+  assert.deepEqual(content(markdownToAdf('Part.\n1. a\n')), [paragraph('Part.'), orderedList(1, item(paragraph('a')))])
+  assert.deepEqual(content(markdownToAdf('Part.\n2. a\n')), [paragraph('Part. 2. a')])
+  assert.deepEqual(content(markdownToAdf('Part.\n*\n')), [paragraph('Part. *')])
+  assert.deepEqual(content(markdownToAdf('Part.\n- - -\n')), [paragraph('Part.'), { type: 'rule' }])
+  assert.deepEqual(content(markdownToAdf('Part.\n-\n')), [{ attrs: { level: 2 }, content: [text('Part.')], type: 'heading' }])
+  assert.deepEqual(content(markdownToAdf('- a\n  2. b\n')), [bulletList(item(paragraph('a 2. b')))])
+})
+
+test('folds a lazy continuation into the paragraph the container holds', () => {
+  assert.deepEqual(content(markdownToAdf('> One\ntwo.\n')), [quote(paragraph('One two.'))])
+  assert.deepEqual(content(markdownToAdf('- One\ntwo.\n')), [bulletList(item(paragraph('One two.')))])
+  assert.deepEqual(content(markdownToAdf('> One\n    two.\n')), [quote(paragraph('One two.'))])
+  assert.deepEqual(content(markdownToAdf('> One\n\ntwo.\n')), [quote(paragraph('One')), paragraph('two.')])
+  assert.deepEqual(content(markdownToAdf('> One\n# Two\n')), [quote(paragraph('One')), { attrs: { level: 1 }, content: [text('Two')], type: 'heading' }])
+  assert.deepEqual(content(markdownToAdf('> One\n---\n')), [quote(paragraph('One')), { type: 'rule' }])
+  assert.deepEqual(content(markdownToAdf('> One\n```\n')), [quote(paragraph('One')), { type: 'codeBlock' }])
+  assert.equal(code(markdownToAdf('> One\n<div>\n')), 'unmappable-html')
+})
+
+test('ends a lazy continuation at a claimed line', () => {
+  assert.equal(code(markdownToAdf('> Part.\n:::\n')), 'malformed-directive')
+  assert.deepEqual(path(markdownToAdf('> Part.\n:::\n')), ['content', 1])
+  assert.equal(code(markdownToAdf('- Part.\n| x |\n')), 'malformed-pipe-table')
+})
+
+test('names the block the claim inside a container opens', () => {
+  assert.deepEqual(path(markdownToAdf('> Part.\n>\n> :::x\n')), ['content', 0, 'content', 1])
+  assert.deepEqual(path(markdownToAdf('- Part.\n- | x |\n')), ['content', 0, 'content', 1, 'content', 0])
+})
+
+test('refuses input nested deeper than the parser carries', () => {
+  assert.equal(code(markdownToAdf('> '.repeat(501))), 'unsupported-nesting-depth')
+  assert.ok(markdownToAdf('> '.repeat(500)).ok)
+})

src/markdown/parse/markdown-to-adf.ts

@@ -1,28 +1,55 @@
 import type { AdfDocument, AdfNode } from '../../adf/document.ts'
-import type { ClaimedConstruct, LeafBlock } from './blocks.ts'
+import type { Block, ClaimedConstruct } from './blocks.ts'
 import { failure, success, type ConvertErrorPath, type Result } from '../../result.ts'
+import { largestNesting } from '../../nesting.ts'
 import { parseBlocks } from './blocks.ts'
 import { trimSpace } from '../commonmark-grammar.ts'
 
 export function markdownToAdf(markdown: string): Result<AdfDocument> {
+  const content = blockNodes(parseBlocks(markdown).blocks, [], 0)
+  if (!content.ok) return content
+  return success(content.value.length === 0 ? { type: 'doc', version: 1 } : { content: content.value, type: 'doc', version: 1 })
+}
+
+function blockNodes(blocks: readonly Block[], path: ConvertErrorPath, depth: number): Result<AdfNode[]> {
+  if (depth > largestNesting) return failure('unsupported-nesting-depth', `the input nests deeper than the ${largestNesting} levels the parser carries`, path)
   const content: AdfNode[] = []
-  for (const [index, block] of parseBlocks(markdown).blocks.entries()) {
-    const node = blockNode(block, ['content', index])
+  for (const [index, block] of blocks.entries()) {
+    const node = blockNode(block, [...path, 'content', index], depth)
     if (!node.ok) return node
     content.push(node.value)
   }
-  return success(content.length === 0 ? { type: 'doc', version: 1 } : { content, type: 'doc', version: 1 })
+  return success(content)
 }
 
-function blockNode(block: LeafBlock, path: ConvertErrorPath): Result<AdfNode> {
+function blockNode(block: Block, path: ConvertErrorPath, depth: number): Result<AdfNode> {
+  if (block.kind === 'blockquote') return containerNode({ type: 'blockquote' }, block.blocks, path, depth)
+  if (block.kind === 'bulletList') return listNode({ type: 'bulletList' }, block.items, path, depth)
   if (block.kind === 'claim') return claimFailure(block.construct, path)
   if (block.kind === 'code') return success(codeBlockNode(block.language, block.text))
   if (block.kind === 'heading') return success(withContent({ attrs: { level: block.level }, type: 'heading' }, block.text))
   if (block.kind === 'html') return failure('unmappable-html', `no ADF node carries ${block.construct}`, path)
+  if (block.kind === 'orderedList') return listNode({ attrs: { order: block.start }, type: 'orderedList' }, block.items, path, depth)
   if (block.kind === 'paragraph') return success(withContent({ type: 'paragraph' }, block.text))
   return success({ type: 'rule' })
 }
 
+function containerNode(node: AdfNode, blocks: readonly Block[], path: ConvertErrorPath, depth: number): Result<AdfNode> {
+  const content = blockNodes(blocks, path, depth + 1)
+  if (!content.ok) return content
+  return success(content.value.length === 0 ? node : { ...node, content: content.value })
+}
+
+function listNode(node: AdfNode, items: readonly Block[][], path: ConvertErrorPath, depth: number): Result<AdfNode> {
+  const content: AdfNode[] = []
+  for (const [index, blocks] of items.entries()) {
+    const item = containerNode({ type: 'listItem' }, blocks, [...path, 'content', index], depth)
+    if (!item.ok) return item
+    content.push(item.value)
+  }
+  return success({ ...node, content })
+}
+
 function claimFailure(construct: ClaimedConstruct, path: ConvertErrorPath): Result<AdfNode> {
   if (construct === 'directive') return failure('malformed-directive', 'the line claims a directive and parses as none', path)
   return failure('malformed-pipe-table', 'the line claims a pipe table and parses as none', path)

src/result.ts

@@ -11,6 +11,7 @@ export type ConvertErrorCode =
   | 'unspellable-whitespace'
   | 'unspelled-block-separation'
   | 'unsupported-document-version'
+  | 'unsupported-nesting-depth'
   | 'unsupported-node-shape'
 
 export type ConvertErrorPath = readonly (number | string)[]