Convert Vanilla Forums Rich2 format to HTML

	<?php

	// For converting Vanilla Forums Rich2 formatted posts to HTML
	// Intended for use when migrating away from Vanilla to something like Flarum

	// Doesn't capture everything. Isn't particularly elegant.
	// It evolved from something else, and I have made minimal effort to clean it up.

	// Comes with no express or implied warranty of any kind.
	// Always make backups of your data and verify this has done what you expect.

	// This is my own work to reverse engineer Rich2
	// into somewhat acceptable HTML while migrating an
	// online community off their software.

	// Sample usage:

	/*
	require __DIR__ . '/Rich2Formatter.php';
	$samplePost = file_get_contents(__DIR__ . '/sample_post.txt');
	$formatter = new Rich2Formatter();
	$output = $formatter->toHtml($samplePost);
	file_put_contents(__DIR__ . '/output.html', $output);
	*/

	class Rich2Formatter
	{
	private readonly array $supportedTypes;
	private readonly array $supportedEmbedDataSourceTypes;

	public function __construct()
	{
	$this->supportedTypes = [
	'@',
	'a',
	'code_block',
	'h2',
	'h3',
	'h4',
	'p',
	'rich_embed_card',
	'table',
	'tbody',
	'thead',
	'td',
	'th',
	'tr',
	'ul',
	'ol',
	];
	$this->supportedEmbedDataSourceTypes = [
	'image',
	'url',
	];
	}

	public function toHtml(string $content): string
	{
	$output = '';
	$parts = json_decode($content, associative: true);

	foreach ($parts ?? [] as $part)
	{
	if (! in_array($part['type'], $this->supportedTypes))
	{
	throw new \Exception("Unable to render section for " . $part['type']);
	}
	$output .= $this->handlePart($part);
	}
	return $output;
	}

	private function handlePart(array $part): string
	{
	$output = '';

	if (isset($part['text']))
	{
	$output .= $this->handleTextNode($part);
	}
	elseif (isset($part['type']))
	{
	if ($part['type'] === 'p')
	{
	$output .= $this->handleParagraph($part);
	}
	elseif ($part['type'] === 'rich_embed_card')
	{
	$output .= $this->handleRichEmbed($part, inline: false);
	}
	elseif ($part['type'] === 'rich_embed_inline')
	{
	$output .= $this->handleRichEmbed($part, inline: true);
	}
	elseif ($part['type'] === 'h2')
	{
	$output .= $this->handleHeading($part, 2);
	}
	elseif ($part['type'] === 'h3')
	{
	$output .= $this->handleHeading($part, 3);
	}
	elseif ($part['type'] === 'h4')
	{
	$output .= $this->handleHeading($part, 4);
	}
	elseif ($part['type'] === 'ol')
	{
	$output .= $this->handleOrderedList($part);
	}
	elseif ($part['type'] === 'ul')
	{
	$output .= $this->handleUnorderedList($part);
	}
	elseif ($part['type'] === 'a')
	{
	$output .= $this->handleLink($part);
	}
	elseif ($part['type'] === 'code_block')
	{
	$output .= $this->handleCodeBlock($part);
	}
	elseif ($part['type'] === 'code_line')
	{
	$output .= $this->handleCodeLine($part);
	}
	elseif ($part['type'] === '@')
	{
	$output .= $this->handleUserTag($part);
	}
	elseif ($part['type'] === 'table')
	{
	$output .= $this->handleTable($part);
	}
	elseif ($part['type'] === 'thead')
	{
	$output .= $this->handleThead($part);
	}
	elseif ($part['type'] === 'tbody')
	{
	$output .= $this->handleTbody($part);
	}
	elseif ($part['type'] === 'td')
	{
	$output .= $this->handleTd($part);
	}
	elseif ($part['type'] === 'th')
	{
	$output .= $this->handleTh($part);
	}
	elseif ($part['type'] === 'tr')
	{
	$output .= $this->handleTr($part);
	}
	else
	{
	throw new \Exception("Unsupported part " . $part['type']);
	}
	}
	return $output;
	}

	private function handleTable(array $part): string
	{
	$output = '';
	foreach($part['children'] as $child) {
	$output .= $this->handlePart($child);
	}
	$output .= '</table>';
	return $output;
	}

	private function handleTbody(array $part): string
	{
	$output = '';
	foreach($part['children'] as $child) {
	$output .= $this->handlePart($child);
	}
	$output .= '</tbody>';
	return $output;
	}

	private function handleThead(array $part): string
	{
	$output = '';
	foreach($part['children'] as $child) {
	$output .= $this->handlePart($child);
	}
	$output .= '</thead>';
	return $output;
	}

	private function handleTd(array $part): string
	{
	$output = '';
	foreach($part['children'] as $child) {
	$output .= $this->handlePart($child);
	}
	$output .= '</td>';
	return $output;
	}

	private function handleTr(array $part): string
	{
	$output = '';
	foreach($part['children'] as $child) {
	$output .= $this->handlePart($child);
	}
	$output .= '</tr>';
	return $output;
	}

	private function handleTh(array $part): string
	{
	$output = '';
	foreach($part['children'] as $child) {
	$output .= $this->handlePart($child);
	}
	$output .= '</th>';
	return $output;
	}

	private function handleUserTag(array $part): string
	{
	return $part['name'];
	}

	private function handleCodeLine(array $part): string
	{
	$output = '';
	foreach ($part['children'] as $child) {
	$output .= $this->handlePart($child) . PHP_EOL;
	}
	return $output;
	}

	private function handleCodeBlock(array $part): string
	{
	$output = ' ';
	foreach ($part['children'] as $child) {
	$output .= $this->handlePart($child) . PHP_EOL;
	}
	$output .= '</pre>';
	return $output;
	}

	private function handleLink(array $part): string
	{
	$output = '<a ';
	if (isset($part['url']))
	{
	$output .= 'href="' . $part['url'] . '" ';
	}
	if (isset($part['target']))
	{
	$output .= 'target="' . $part['target'] . '" ';
	}
	$output = trim($output);
	$output .= '>';

	$output .= $this->handleTextNode($part['children'][0]);

	$output .= '</a>';

	return $output;
	}

	private function handleUnorderedList(array $part): string
	{
	$output = '
	foreach ($part['children'] as $listItem)
	{
	$output .= ' ';
	foreach ($listItem['children'][0]['children'] as $textNode)
	{
	$output .= $this->handleTextNode($textNode);
	}
	$output .= '</li>';
	}
	$output .= '</ul>';
	return $output;
	}

	private function handleOrderedList(array $part): string
	{
	$output = '
	foreach ($part['children'] as $listItem)
	{
	$output .= ' ';
	foreach ($listItem['children'][0]['children'] as $textNode)
	{
	$output .= $this->handleTextNode($textNode);
	}
	$output .= '</li>';
	}
	$output .= '</ol>';
	return $output;
	}

	private function handleTextNode(array $textNode): string
	{
	if (! isset($textNode['text']))
	{
	return $this->handlePart($textNode);
	}

	$text = '';
	if (isset($textNode['bold']) && $textNode['bold'] === true)
	{
	$text .= '';
	}
	if (isset($textNode['italic']) && $textNode['italic'] === true)
	{
	$text .= '';
	}

	$text .= $textNode['text'];

	if (isset($textNode['italic']) && $textNode['italic'] === true)
	{
	$text .= '</em>';
	}
	if (isset($textNode['bold']) && $textNode['bold'] === true)
	{
	$text .= '</strong>';
	}
	return $text;
	}

	private function handleHeading(array $part, int $level): string
	{
	$text = '';
	foreach ($part['children'] as $child)
	{
	$text .= $this->handleTextNode($child);
	}
	return "{$text}";
	}

	private function handleParagraph(array $part): string
	{
	$output = ' ';
	foreach ($part['children'] as $child)
	{
	$output .= $this->handlePart($child);
	}
	$output .= '</p>';
	return $output;
	}

	private function handleRichEmbed(array $part, bool $inline = false): string
	{
	$output = '';

	if (array_key_exists('error', $part))
	{
	return $output;
	}

	if (! $inline)
	{
	$output .= '';
	}

	if (in_array($part['dataSourceType'], $this->supportedEmbedDataSourceTypes))
	{
	if ($part['dataSourceType'] === 'image')
	{
	$output .= <<<HTML
	<img
	src="{$part['embedData']['url']}"
	data-src="{$part['embedData']['url']}"
	alt="{$part['children'][0]['text']}"
	width="{$part['embedData']['width']}"
	height="{$part['embedData']['height']}"
	/>
	HTML;
	}
	elseif ($part['dataSourceType'] === 'url')
	{
	$output .= "";
	foreach ($part['children'] as $child) {
	$output .= $this->handlePart($child);
	}
	$output .= "</a>";
	}
	}
	else
	{
	throw new Exception('Unsupported embed type ' . $part['dataSourceType']);
	}

	if (! $inline)
	{
	$output .= '</div>';
	}

	return $output;
	}
	}

参与讨论