resolved conflicts

This commit is contained in:
2025-03-31 03:46:05 +03:00
6454 changed files with 1520539 additions and 9 deletions
@@ -0,0 +1,142 @@
<?php
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
* (c) 2015 Martin Hasoň <martin.hason@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
declare(strict_types=1);
namespace League\CommonMark\Extension\Attributes\Util;
use League\CommonMark\Node\Node;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Util\RegexHelper;
/**
* @internal
*/
final class AttributesHelper
{
private const SINGLE_ATTRIBUTE = '\s*([.]-?[_a-z][^\s}]*|[#][^\s}]+|' . RegexHelper::PARTIAL_ATTRIBUTENAME . RegexHelper::PARTIAL_ATTRIBUTEVALUESPEC . ')\s*';
private const ATTRIBUTE_LIST = '/^{:?(' . self::SINGLE_ATTRIBUTE . ')+}/i';
/**
* @return array<string, mixed>
*/
public static function parseAttributes(Cursor $cursor): array
{
$state = $cursor->saveState();
$cursor->advanceToNextNonSpaceOrNewline();
// Quick check to see if we might have attributes
if ($cursor->getCharacter() !== '{') {
$cursor->restoreState($state);
return [];
}
// Attempt to match the entire attribute list expression
// While this is less performant than checking for '{' now and '}' later, it simplifies
// matching individual attributes since they won't need to look ahead for the closing '}'
// while dealing with the fact that attributes can technically contain curly braces.
// So we'll just match the start and end braces up front.
$attributeExpression = $cursor->match(self::ATTRIBUTE_LIST);
if ($attributeExpression === null) {
$cursor->restoreState($state);
return [];
}
// Trim the leading '{' or '{:' and the trailing '}'
$attributeExpression = \ltrim(\substr($attributeExpression, 1, -1), ':');
$attributeCursor = new Cursor($attributeExpression);
/** @var array<string, mixed> $attributes */
$attributes = [];
while ($attribute = \trim((string) $attributeCursor->match('/^' . self::SINGLE_ATTRIBUTE . '/i'))) {
if ($attribute[0] === '#') {
$attributes['id'] = \substr($attribute, 1);
continue;
}
if ($attribute[0] === '.') {
$attributes['class'][] = \substr($attribute, 1);
continue;
}
/** @psalm-suppress PossiblyUndefinedArrayOffset */
[$name, $value] = \explode('=', $attribute, 2);
if ($value === 'true') {
$attributes[$name] = true;
continue;
}
$first = $value[0];
$last = \substr($value, -1);
if (($first === '"' && $last === '"') || ($first === "'" && $last === "'") && \strlen($value) > 1) {
$value = \substr($value, 1, -1);
}
if (\strtolower(\trim($name)) === 'class') {
foreach (\array_filter(\explode(' ', \trim($value))) as $class) {
$attributes['class'][] = $class;
}
} else {
$attributes[\trim($name)] = \trim($value);
}
}
if (isset($attributes['class'])) {
$attributes['class'] = \implode(' ', (array) $attributes['class']);
}
return $attributes;
}
/**
* @param Node|array<string, mixed> $attributes1
* @param Node|array<string, mixed> $attributes2
*
* @return array<string, mixed>
*/
public static function mergeAttributes($attributes1, $attributes2): array
{
$attributes = [];
foreach ([$attributes1, $attributes2] as $arg) {
if ($arg instanceof Node) {
$arg = $arg->data->get('attributes');
}
/** @var array<string, mixed> $arg */
$arg = (array) $arg;
if (isset($arg['class'])) {
if (\is_string($arg['class'])) {
$arg['class'] = \array_filter(\explode(' ', \trim($arg['class'])));
}
foreach ($arg['class'] as $class) {
$attributes['class'][] = $class;
}
unset($arg['class']);
}
$attributes = \array_merge($attributes, $arg);
}
if (isset($attributes['class'])) {
$attributes['class'] = \implode(' ', $attributes['class']);
}
return $attributes;
}
}
@@ -0,0 +1,39 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\Autolink;
use League\CommonMark\Environment\EnvironmentBuilderInterface;
use League\CommonMark\Extension\ConfigurableExtensionInterface;
use League\Config\ConfigurationBuilderInterface;
use Nette\Schema\Expect;
final class AutolinkExtension implements ConfigurableExtensionInterface
{
public function configureSchema(ConfigurationBuilderInterface $builder): void
{
$builder->addSchema('autolink', Expect::structure([
'allowed_protocols' => Expect::listOf('string')->default(['http', 'https', 'ftp'])->mergeDefaults(false),
'default_protocol' => Expect::string()->default('http'),
]));
}
public function register(EnvironmentBuilderInterface $environment): void
{
$environment->addInlineParser(new EmailAutolinkParser());
$environment->addInlineParser(new UrlAutolinkParser(
$environment->getConfiguration()->get('autolink.allowed_protocols'),
$environment->getConfiguration()->get('autolink.default_protocol'),
));
}
}
@@ -0,0 +1,157 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\Autolink;
use League\CommonMark\Extension\CommonMark\Node\Inline\Link;
use League\CommonMark\Parser\Inline\InlineParserInterface;
use League\CommonMark\Parser\Inline\InlineParserMatch;
use League\CommonMark\Parser\InlineParserContext;
final class UrlAutolinkParser implements InlineParserInterface
{
private const ALLOWED_AFTER = [null, ' ', "\t", "\n", "\x0b", "\x0c", "\x0d", '*', '_', '~', '('];
// RegEx adapted from https://github.com/symfony/symfony/blob/6.3/src/Symfony/Component/Validator/Constraints/UrlValidator.php
private const REGEX = '~
(
# Must start with a supported scheme + auth, or "www"
(?:
(?:%s):// # protocol
(?:(?:(?:[\_\.\pL\pN-]|%%[0-9A-Fa-f]{2})+:)?((?:[\_\.\pL\pN-]|%%[0-9A-Fa-f]{2})+)@)? # basic auth
|www\.)
(?:
(?:
(?:xn--[a-z0-9-]++\.)*+xn--[a-z0-9-]++ # a domain name using punycode
|
(?:[\pL\pN\pS\pM\-\_]++\.){1,127}[\pL\pN\pM]++ # a multi-level domain name; total length must be 253 bytes or less
|
[a-z0-9\-\_]++ # a single-level domain name
)\.?
| # or
\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3} # an IP address
| # or
\[
(?:(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){6})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:::(?:(?:(?:[0-9a-f]{1,4})):){5})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){4})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,1}(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){3})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,2}(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){2})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,3}(?:(?:[0-9a-f]{1,4})))?::(?:(?:[0-9a-f]{1,4})):)(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,4}(?:(?:[0-9a-f]{1,4})))?::)(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,5}(?:(?:[0-9a-f]{1,4})))?::)(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,6}(?:(?:[0-9a-f]{1,4})))?::))))
\] # an IPv6 address
)
(?::[0-9]+)? # a port (optional)
(?:/ (?:[\pL\pN\-._\~!$&\'()*+,;=:@]|%%[0-9A-Fa-f]{2})* )* # a path
(?:\? (?:[\pL\pN\-._\~!$&\'\[\]()*+,;=:@/?]|%%[0-9A-Fa-f]{2})* )? # a query (optional)
(?:\# (?:[\pL\pN\-._\~!$&\'()*+,;=:@/?]|%%[0-9A-Fa-f]{2})* )? # a fragment (optional)
)~ixu';
/**
* @var string[]
*
* @psalm-readonly
*/
private array $prefixes = ['www.'];
/**
* @psalm-var non-empty-string
*
* @psalm-readonly
*/
private string $finalRegex;
private string $defaultProtocol;
/**
* @param array<int, string> $allowedProtocols
*/
public function __construct(array $allowedProtocols = ['http', 'https', 'ftp'], string $defaultProtocol = 'http')
{
/**
* @psalm-suppress PropertyTypeCoercion
*/
$this->finalRegex = \sprintf(self::REGEX, \implode('|', $allowedProtocols));
foreach ($allowedProtocols as $protocol) {
$this->prefixes[] = $protocol . '://';
}
$this->defaultProtocol = $defaultProtocol;
}
public function getMatchDefinition(): InlineParserMatch
{
return InlineParserMatch::oneOf(...$this->prefixes);
}
public function parse(InlineParserContext $inlineContext): bool
{
$cursor = $inlineContext->getCursor();
// Autolinks can only come at the beginning of a line, after whitespace, or certain delimiting characters
$previousChar = $cursor->peek(-1);
if (! \in_array($previousChar, self::ALLOWED_AFTER, true)) {
return false;
}
// Check if we have a valid URL
if (! \preg_match($this->finalRegex, $cursor->getRemainder(), $matches)) {
return false;
}
$url = $matches[0];
// Does the URL end with punctuation that should be stripped?
if (\preg_match('/(.+?)([?!.,:*_~]+)$/', $url, $matches)) {
// Add the punctuation later
$url = $matches[1];
}
// Does the URL end with something that looks like an entity reference?
if (\preg_match('/(.+)(&[A-Za-z0-9]+;)$/', $url, $matches)) {
$url = $matches[1];
}
// Does the URL need unmatched parens chopped off?
if (\substr($url, -1) === ')' && ($diff = self::diffParens($url)) > 0) {
$url = \substr($url, 0, -$diff);
}
$cursor->advanceBy(\mb_strlen($url, 'UTF-8'));
// Auto-prefix 'http(s)://' onto 'www' URLs
if (\substr($url, 0, 4) === 'www.') {
$inlineContext->getContainer()->appendChild(new Link($this->defaultProtocol . '://' . $url, $url));
return true;
}
$inlineContext->getContainer()->appendChild(new Link($url, $url));
return true;
}
/**
* @psalm-pure
*/
private static function diffParens(string $content): int
{
// Scan the entire autolink for the total number of parentheses.
// If there is a greater number of closing parentheses than opening ones,
// we dont consider ANY of the last characters as part of the autolink,
// in order to facilitate including an autolink inside a parenthesis.
\preg_match_all('/[()]/', $content, $matches);
$charCount = ['(' => 0, ')' => 0];
foreach ($matches[0] as $char) {
$charCount[$char]++;
}
return $charCount[')'] - $charCount['('];
}
}
@@ -0,0 +1,119 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* Additional emphasis processing code based on commonmark-java (https://github.com/atlassian/commonmark-java)
* - (c) Atlassian Pty Ltd
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Delimiter\Processor;
use League\CommonMark\Delimiter\DelimiterInterface;
use League\CommonMark\Delimiter\Processor\CacheableDelimiterProcessorInterface;
use League\CommonMark\Extension\CommonMark\Node\Inline\Emphasis;
use League\CommonMark\Extension\CommonMark\Node\Inline\Strong;
use League\CommonMark\Node\Inline\AbstractStringContainer;
use League\Config\ConfigurationAwareInterface;
use League\Config\ConfigurationInterface;
final class EmphasisDelimiterProcessor implements CacheableDelimiterProcessorInterface, ConfigurationAwareInterface
{
/** @psalm-readonly */
private string $char;
/** @psalm-readonly-allow-private-mutation */
private ConfigurationInterface $config;
/**
* @param string $char The emphasis character to use (typically '*' or '_')
*/
public function __construct(string $char)
{
$this->char = $char;
}
public function getOpeningCharacter(): string
{
return $this->char;
}
public function getClosingCharacter(): string
{
return $this->char;
}
public function getMinLength(): int
{
return 1;
}
public function getDelimiterUse(DelimiterInterface $opener, DelimiterInterface $closer): int
{
// "Multiple of 3" rule for internal delimiter runs
if (($opener->canClose() || $closer->canOpen()) && $closer->getOriginalLength() % 3 !== 0 && ($opener->getOriginalLength() + $closer->getOriginalLength()) % 3 === 0) {
return 0;
}
// Calculate actual number of delimiters used from this closer
if ($opener->getLength() >= 2 && $closer->getLength() >= 2) {
if ($this->config->get('commonmark/enable_strong')) {
return 2;
}
return 0;
}
if ($this->config->get('commonmark/enable_em')) {
return 1;
}
return 0;
}
public function process(AbstractStringContainer $opener, AbstractStringContainer $closer, int $delimiterUse): void
{
if ($delimiterUse === 1) {
$emphasis = new Emphasis($this->char);
} elseif ($delimiterUse === 2) {
$emphasis = new Strong($this->char . $this->char);
} else {
return;
}
$next = $opener->next();
while ($next !== null && $next !== $closer) {
$tmp = $next->next();
$emphasis->appendChild($next);
$next = $tmp;
}
$opener->insertAfter($emphasis);
}
public function setConfiguration(ConfigurationInterface $configuration): void
{
$this->config = $configuration;
}
public function getCacheKey(DelimiterInterface $closer): string
{
return \sprintf(
'%s-%s-%d-%d',
$this->char,
$closer->canOpen() ? 'canOpen' : 'cannotOpen',
$closer->getOriginalLength() % 3,
$closer->getLength(),
);
}
}
@@ -0,0 +1,56 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Node\Block;
use League\CommonMark\Node\Block\AbstractBlock;
use League\CommonMark\Node\Block\TightBlockInterface;
class ListBlock extends AbstractBlock implements TightBlockInterface
{
public const TYPE_BULLET = 'bullet';
public const TYPE_ORDERED = 'ordered';
public const DELIM_PERIOD = 'period';
public const DELIM_PAREN = 'paren';
protected bool $tight = false; // TODO Make lists tight by default in v3
/** @psalm-readonly */
protected ListData $listData;
public function __construct(ListData $listData)
{
parent::__construct();
$this->listData = $listData;
}
public function getListData(): ListData
{
return $this->listData;
}
public function isTight(): bool
{
return $this->tight;
}
public function setTight(bool $tight): void
{
$this->tight = $tight;
}
}
@@ -0,0 +1,84 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Block;
use League\CommonMark\Extension\CommonMark\Node\Block\FencedCode;
use League\CommonMark\Parser\Block\AbstractBlockContinueParser;
use League\CommonMark\Parser\Block\BlockContinue;
use League\CommonMark\Parser\Block\BlockContinueParserInterface;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Util\ArrayCollection;
use League\CommonMark\Util\RegexHelper;
final class FencedCodeParser extends AbstractBlockContinueParser
{
/** @psalm-readonly */
private FencedCode $block;
/** @var ArrayCollection<string> */
private ArrayCollection $strings;
public function __construct(int $fenceLength, string $fenceChar, int $fenceOffset)
{
$this->block = new FencedCode($fenceLength, $fenceChar, $fenceOffset);
$this->strings = new ArrayCollection();
}
public function getBlock(): FencedCode
{
return $this->block;
}
public function tryContinue(Cursor $cursor, BlockContinueParserInterface $activeBlockParser): ?BlockContinue
{
// Check for closing code fence
if (! $cursor->isIndented() && $cursor->getNextNonSpaceCharacter() === $this->block->getChar()) {
$match = RegexHelper::matchFirst('/^(?:`{3,}|~{3,})(?=[ \t]*$)/', $cursor->getLine(), $cursor->getNextNonSpacePosition());
if ($match !== null && \strlen($match[0]) >= $this->block->getLength()) {
// closing fence - we're at end of line, so we can finalize now
return BlockContinue::finished();
}
}
// Skip optional spaces of fence offset
// Optimization: don't attempt to match if we're at a non-space position
if ($cursor->getNextNonSpacePosition() > $cursor->getPosition()) {
$cursor->match('/^ {0,' . $this->block->getOffset() . '}/');
}
return BlockContinue::at($cursor);
}
public function addLine(string $line): void
{
$this->strings[] = $line;
}
public function closeBlock(): void
{
// first line becomes info string
$firstLine = $this->strings->first();
if ($firstLine === false) {
$firstLine = '';
}
$this->block->setInfo(RegexHelper::unescape(\trim($firstLine)));
if ($this->strings->count() === 1) {
$this->block->setLiteral('');
} else {
$this->block->setLiteral(\implode("\n", $this->strings->slice(1)) . "\n");
}
}
}
@@ -0,0 +1,76 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Block;
use League\CommonMark\Extension\CommonMark\Node\Block\IndentedCode;
use League\CommonMark\Parser\Block\AbstractBlockContinueParser;
use League\CommonMark\Parser\Block\BlockContinue;
use League\CommonMark\Parser\Block\BlockContinueParserInterface;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Util\ArrayCollection;
final class IndentedCodeParser extends AbstractBlockContinueParser
{
/** @psalm-readonly */
private IndentedCode $block;
/** @var ArrayCollection<string> */
private ArrayCollection $strings;
public function __construct()
{
$this->block = new IndentedCode();
$this->strings = new ArrayCollection();
}
public function getBlock(): IndentedCode
{
return $this->block;
}
public function tryContinue(Cursor $cursor, BlockContinueParserInterface $activeBlockParser): ?BlockContinue
{
if ($cursor->isIndented()) {
$cursor->advanceBy(Cursor::INDENT_LEVEL, true);
return BlockContinue::at($cursor);
}
if ($cursor->isBlank()) {
$cursor->advanceToNextNonSpaceOrTab();
return BlockContinue::at($cursor);
}
return BlockContinue::none();
}
public function addLine(string $line): void
{
$this->strings[] = $line;
}
public function closeBlock(): void
{
$lines = $this->strings->toArray();
// Note that indented code block cannot be empty, so $lines will always have at least one non-empty element
while (\preg_match('/^[ \t]*$/', \end($lines))) { // @phpstan-ignore-line
\array_pop($lines);
}
$this->block->setLiteral(\implode("\n", $lines) . "\n");
$this->block->setEndLine($this->block->getStartLine() + \count($lines) - 1);
}
}
@@ -0,0 +1,93 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Block;
use League\CommonMark\Extension\CommonMark\Node\Block\ListBlock;
use League\CommonMark\Extension\CommonMark\Node\Block\ListData;
use League\CommonMark\Extension\CommonMark\Node\Block\ListItem;
use League\CommonMark\Node\Block\AbstractBlock;
use League\CommonMark\Parser\Block\AbstractBlockContinueParser;
use League\CommonMark\Parser\Block\BlockContinue;
use League\CommonMark\Parser\Block\BlockContinueParserInterface;
use League\CommonMark\Parser\Cursor;
final class ListBlockParser extends AbstractBlockContinueParser
{
/** @psalm-readonly */
private ListBlock $block;
public function __construct(ListData $listData)
{
$this->block = new ListBlock($listData);
}
public function getBlock(): ListBlock
{
return $this->block;
}
public function isContainer(): bool
{
return true;
}
public function canContain(AbstractBlock $childBlock): bool
{
return $childBlock instanceof ListItem;
}
public function tryContinue(Cursor $cursor, BlockContinueParserInterface $activeBlockParser): ?BlockContinue
{
// List blocks themselves don't have any markers, only list items. So try to stay in the list.
// If there is a block start other than list item, canContain makes sure that this list is closed.
return BlockContinue::at($cursor);
}
public function closeBlock(): void
{
$item = $this->block->firstChild();
while ($item instanceof AbstractBlock) {
// check for non-final list item ending with blank line:
if ($item->next() !== null && self::endsWithBlankLine($item)) {
$this->block->setTight(false);
break;
}
// recurse into children of list item, to see if there are spaces between any of them
$subitem = $item->firstChild();
while ($subitem instanceof AbstractBlock) {
if ($subitem->next() && self::endsWithBlankLine($subitem)) {
$this->block->setTight(false);
break 2;
}
$subitem = $subitem->next();
}
$item = $item->next();
}
$lastChild = $this->block->lastChild();
if ($lastChild instanceof AbstractBlock) {
$this->block->setEndLine($lastChild->getEndLine());
}
}
private static function endsWithBlankLine(AbstractBlock $block): bool
{
$next = $block->next();
return $next instanceof AbstractBlock && $block->getEndLine() !== $next->getStartLine() - 1;
}
}
@@ -0,0 +1,154 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Block;
use League\CommonMark\Extension\CommonMark\Node\Block\ListBlock;
use League\CommonMark\Extension\CommonMark\Node\Block\ListData;
use League\CommonMark\Parser\Block\BlockStart;
use League\CommonMark\Parser\Block\BlockStartParserInterface;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Parser\MarkdownParserStateInterface;
use League\CommonMark\Util\RegexHelper;
use League\Config\ConfigurationAwareInterface;
use League\Config\ConfigurationInterface;
final class ListBlockStartParser implements BlockStartParserInterface, ConfigurationAwareInterface
{
/** @psalm-readonly-allow-private-mutation */
private ?ConfigurationInterface $config = null;
/**
* @psalm-var non-empty-string|null
*
* @psalm-readonly-allow-private-mutation
*/
private ?string $listMarkerRegex = null;
public function setConfiguration(ConfigurationInterface $configuration): void
{
$this->config = $configuration;
}
public function tryStart(Cursor $cursor, MarkdownParserStateInterface $parserState): ?BlockStart
{
if ($cursor->isIndented()) {
return BlockStart::none();
}
$listData = $this->parseList($cursor, $parserState->getParagraphContent() !== null);
if ($listData === null) {
return BlockStart::none();
}
$listItemParser = new ListItemParser($listData);
// prepend the list block if needed
$matched = $parserState->getLastMatchedBlockParser();
if (! ($matched instanceof ListBlockParser) || ! $listData->equals($matched->getBlock()->getListData())) {
$listBlockParser = new ListBlockParser($listData);
// We start out with assuming a list is tight. If we find a blank line, we set it to loose later.
// TODO for 3.0: Just make them tight by default in the block so we can remove this call
$listBlockParser->getBlock()->setTight(true);
return BlockStart::of($listBlockParser, $listItemParser)->at($cursor);
}
return BlockStart::of($listItemParser)->at($cursor);
}
private function parseList(Cursor $cursor, bool $inParagraph): ?ListData
{
$indent = $cursor->getIndent();
$tmpCursor = clone $cursor;
$tmpCursor->advanceToNextNonSpaceOrTab();
$rest = $tmpCursor->getRemainder();
if (\preg_match($this->listMarkerRegex ?? $this->generateListMarkerRegex(), $rest) === 1) {
$data = new ListData();
$data->markerOffset = $indent;
$data->type = ListBlock::TYPE_BULLET;
$data->delimiter = null;
$data->bulletChar = $rest[0];
$markerLength = 1;
} elseif (($matches = RegexHelper::matchFirst('/^(\d{1,9})([.)])/', $rest)) && (! $inParagraph || $matches[1] === '1')) {
$data = new ListData();
$data->markerOffset = $indent;
$data->type = ListBlock::TYPE_ORDERED;
$data->start = (int) $matches[1];
$data->delimiter = $matches[2] === '.' ? ListBlock::DELIM_PERIOD : ListBlock::DELIM_PAREN;
$data->bulletChar = null;
$markerLength = \strlen($matches[0]);
} else {
return null;
}
// Make sure we have spaces after
$nextChar = $tmpCursor->peek($markerLength);
if (! ($nextChar === null || $nextChar === "\t" || $nextChar === ' ')) {
return null;
}
// If it interrupts paragraph, make sure first line isn't blank
if ($inParagraph && ! RegexHelper::matchAt(RegexHelper::REGEX_NON_SPACE, $rest, $markerLength)) {
return null;
}
$cursor->advanceToNextNonSpaceOrTab(); // to start of marker
$cursor->advanceBy($markerLength, true); // to end of marker
$data->padding = self::calculateListMarkerPadding($cursor, $markerLength);
return $data;
}
private static function calculateListMarkerPadding(Cursor $cursor, int $markerLength): int
{
$start = $cursor->saveState();
$spacesStartCol = $cursor->getColumn();
while ($cursor->getColumn() - $spacesStartCol < 5) {
if (! $cursor->advanceBySpaceOrTab()) {
break;
}
}
$blankItem = $cursor->peek() === null;
$spacesAfterMarker = $cursor->getColumn() - $spacesStartCol;
if ($spacesAfterMarker >= 5 || $spacesAfterMarker < 1 || $blankItem) {
$cursor->restoreState($start);
$cursor->advanceBySpaceOrTab();
return $markerLength + 1;
}
return $markerLength + $spacesAfterMarker;
}
/**
* @psalm-return non-empty-string
*/
private function generateListMarkerRegex(): string
{
// No configuration given - use the defaults
if ($this->config === null) {
return $this->listMarkerRegex = '/^[*+-]/';
}
$markers = $this->config->get('commonmark/unordered_list_markers');
\assert(\is_array($markers));
return $this->listMarkerRegex = '/^[' . \preg_quote(\implode('', $markers), '/') . ']/';
}
}
@@ -0,0 +1,82 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Block;
use League\CommonMark\Extension\CommonMark\Node\Block\ListData;
use League\CommonMark\Extension\CommonMark\Node\Block\ListItem;
use League\CommonMark\Node\Block\AbstractBlock;
use League\CommonMark\Parser\Block\AbstractBlockContinueParser;
use League\CommonMark\Parser\Block\BlockContinue;
use League\CommonMark\Parser\Block\BlockContinueParserInterface;
use League\CommonMark\Parser\Cursor;
final class ListItemParser extends AbstractBlockContinueParser
{
/** @psalm-readonly */
private ListItem $block;
public function __construct(ListData $listData)
{
$this->block = new ListItem($listData);
}
public function getBlock(): ListItem
{
return $this->block;
}
public function isContainer(): bool
{
return true;
}
public function canContain(AbstractBlock $childBlock): bool
{
return ! $childBlock instanceof ListItem;
}
public function tryContinue(Cursor $cursor, BlockContinueParserInterface $activeBlockParser): ?BlockContinue
{
if ($cursor->isBlank()) {
if ($this->block->firstChild() === null) {
// Blank line after empty list item
return BlockContinue::none();
}
$cursor->advanceToNextNonSpaceOrTab();
return BlockContinue::at($cursor);
}
$contentIndent = $this->block->getListData()->markerOffset + $this->getBlock()->getListData()->padding;
if ($cursor->getIndent() >= $contentIndent) {
$cursor->advanceBy($contentIndent, true);
return BlockContinue::at($cursor);
}
// Note: We'll hit this case for lazy continuation lines, they will get added later.
return BlockContinue::none();
}
public function closeBlock(): void
{
if (($lastChild = $this->block->lastChild()) instanceof AbstractBlock) {
$this->block->setEndLine($lastChild->getEndLine());
} else {
// Empty list item
$this->block->setEndLine($this->block->getStartLine());
}
}
}
@@ -0,0 +1,132 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Inline;
use League\CommonMark\Extension\CommonMark\Node\Inline\Code;
use League\CommonMark\Node\Inline\Text;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Parser\Inline\InlineParserInterface;
use League\CommonMark\Parser\Inline\InlineParserMatch;
use League\CommonMark\Parser\InlineParserContext;
final class BacktickParser implements InlineParserInterface
{
/**
* Max bound for backtick code span delimiters.
*
* @see https://github.com/commonmark/cmark/commit/8ed5c9d
*/
private const MAX_BACKTICKS = 1000;
/** @var \WeakReference<Cursor>|null */
private ?\WeakReference $lastCursor = null;
private bool $lastCursorScanned = false;
/** @var array<int, int> backtick count => position of known ender */
private array $seenBackticks = [];
public function getMatchDefinition(): InlineParserMatch
{
return InlineParserMatch::regex('`+');
}
public function parse(InlineParserContext $inlineContext): bool
{
$ticks = $inlineContext->getFullMatch();
$cursor = $inlineContext->getCursor();
$cursor->advanceBy($inlineContext->getFullMatchLength());
$currentPosition = $cursor->getPosition();
$previousState = $cursor->saveState();
if ($this->findMatchingTicks(\strlen($ticks), $cursor)) {
$code = $cursor->getSubstring($currentPosition, $cursor->getPosition() - $currentPosition - \strlen($ticks));
$c = \preg_replace('/\n/m', ' ', $code) ?? '';
if (
$c !== '' &&
$c[0] === ' ' &&
\substr($c, -1, 1) === ' ' &&
\preg_match('/[^ ]/', $c)
) {
$c = \substr($c, 1, -1);
}
$inlineContext->getContainer()->appendChild(new Code($c));
return true;
}
// If we got here, we didn't match a closing backtick sequence
$cursor->restoreState($previousState);
$inlineContext->getContainer()->appendChild(new Text($ticks));
return true;
}
/**
* Locates the matching closer for a backtick code span.
*
* Leverages some caching to avoid traversing the same cursor multiple times when
* we've already seen all the potential backtick closers.
*
* @see https://github.com/commonmark/cmark/commit/8ed5c9d
*
* @param int $openTickLength Number of backticks in the opening sequence
* @param Cursor $cursor Cursor to scan
*
* @return bool True if a matching closer was found, false otherwise
*/
private function findMatchingTicks(int $openTickLength, Cursor $cursor): bool
{
// Reset the seenBackticks cache if this is a new cursor
if ($this->lastCursor === null || $this->lastCursor->get() !== $cursor) {
$this->seenBackticks = [];
$this->lastCursor = \WeakReference::create($cursor);
$this->lastCursorScanned = false;
}
if ($openTickLength > self::MAX_BACKTICKS) {
return false;
}
// Return if we already know there's no closer
if ($this->lastCursorScanned && isset($this->seenBackticks[$openTickLength]) && $this->seenBackticks[$openTickLength] <= $cursor->getPosition()) {
return false;
}
while ($ticks = $cursor->match('/`{1,' . self::MAX_BACKTICKS . '}/m')) {
$numTicks = \strlen($ticks);
// Did we find the closer?
if ($numTicks === $openTickLength) {
return true;
}
// Store position of closer
if ($numTicks <= self::MAX_BACKTICKS) {
$this->seenBackticks[$numTicks] = $cursor->getPosition() - $numTicks;
}
}
// Got through whole input without finding closer
$this->lastCursorScanned = true;
return false;
}
}
@@ -0,0 +1,44 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Inline;
use League\CommonMark\Node\Inline\Text;
use League\CommonMark\Parser\Inline\InlineParserInterface;
use League\CommonMark\Parser\Inline\InlineParserMatch;
use League\CommonMark\Parser\InlineParserContext;
final class BangParser implements InlineParserInterface
{
public function getMatchDefinition(): InlineParserMatch
{
return InlineParserMatch::string('![');
}
public function parse(InlineParserContext $inlineContext): bool
{
$cursor = $inlineContext->getCursor();
$cursor->advanceBy(2);
$node = new Text('![', ['delim' => true]);
$inlineContext->getContainer()->appendChild($node);
// Add entry to stack for this opener
$inlineContext->getDelimiterStack()->addBracket($node, $cursor->getPosition(), true);
return true;
}
}
@@ -0,0 +1,214 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Inline;
use League\CommonMark\Delimiter\Bracket;
use League\CommonMark\Environment\EnvironmentAwareInterface;
use League\CommonMark\Environment\EnvironmentInterface;
use League\CommonMark\Extension\CommonMark\Node\Inline\AbstractWebResource;
use League\CommonMark\Extension\CommonMark\Node\Inline\Image;
use League\CommonMark\Extension\CommonMark\Node\Inline\Link;
use League\CommonMark\Extension\Mention\Mention;
use League\CommonMark\Node\Inline\AdjacentTextMerger;
use League\CommonMark\Node\Inline\Text;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Parser\Inline\InlineParserInterface;
use League\CommonMark\Parser\Inline\InlineParserMatch;
use League\CommonMark\Parser\InlineParserContext;
use League\CommonMark\Reference\ReferenceInterface;
use League\CommonMark\Reference\ReferenceMapInterface;
use League\CommonMark\Util\LinkParserHelper;
use League\CommonMark\Util\RegexHelper;
final class CloseBracketParser implements InlineParserInterface, EnvironmentAwareInterface
{
/** @psalm-readonly-allow-private-mutation */
private EnvironmentInterface $environment;
public function getMatchDefinition(): InlineParserMatch
{
return InlineParserMatch::string(']');
}
public function parse(InlineParserContext $inlineContext): bool
{
// Look through stack of delimiters for a [ or !
$opener = $inlineContext->getDelimiterStack()->getLastBracket();
if ($opener === null) {
return false;
}
if (! $opener->isImage() && ! $opener->isActive()) {
// no matched opener; remove from stack
$inlineContext->getDelimiterStack()->removeBracket();
return false;
}
$cursor = $inlineContext->getCursor();
$startPos = $cursor->getPosition();
$previousState = $cursor->saveState();
$cursor->advanceBy(1);
// Check to see if we have a link/image
// Inline link?
if ($result = $this->tryParseInlineLinkAndTitle($cursor)) {
$link = $result;
} elseif ($link = $this->tryParseReference($cursor, $inlineContext->getReferenceMap(), $opener, $startPos)) {
$reference = $link;
$link = ['url' => $link->getDestination(), 'title' => $link->getTitle()];
} else {
// No match; remove this opener from stack
$inlineContext->getDelimiterStack()->removeBracket();
$cursor->restoreState($previousState);
return false;
}
$inline = $this->createInline($link['url'], $link['title'], $opener->isImage(), $reference ?? null);
$opener->getNode()->replaceWith($inline);
while (($label = $inline->next()) !== null) {
// Is there a Mention or Link contained within this link?
// CommonMark does not allow nested links, so we'll restore the original text.
if ($label instanceof Mention) {
$label->replaceWith($replacement = new Text($label->getPrefix() . $label->getIdentifier()));
$inline->appendChild($replacement);
} elseif ($label instanceof Link) {
foreach ($label->children() as $child) {
$label->insertBefore($child);
}
$label->detach();
} else {
$inline->appendChild($label);
}
}
// Process delimiters such as emphasis inside link/image
$delimiterStack = $inlineContext->getDelimiterStack();
$stackBottom = $opener->getPosition();
$delimiterStack->processDelimiters($stackBottom, $this->environment->getDelimiterProcessors());
$delimiterStack->removeBracket();
$delimiterStack->removeAll($stackBottom);
// Merge any adjacent Text nodes together
AdjacentTextMerger::mergeChildNodes($inline);
// processEmphasis will remove this and later delimiters.
// Now, for a link, we also remove earlier link openers (no links in links)
if (! $opener->isImage()) {
$inlineContext->getDelimiterStack()->deactivateLinkOpeners();
}
return true;
}
public function setEnvironment(EnvironmentInterface $environment): void
{
$this->environment = $environment;
}
/**
* @return array<string, string>|null
*/
private function tryParseInlineLinkAndTitle(Cursor $cursor): ?array
{
if ($cursor->getCurrentCharacter() !== '(') {
return null;
}
$previousState = $cursor->saveState();
$cursor->advanceBy(1);
$cursor->advanceToNextNonSpaceOrNewline();
if (($dest = LinkParserHelper::parseLinkDestination($cursor)) === null) {
$cursor->restoreState($previousState);
return null;
}
$cursor->advanceToNextNonSpaceOrNewline();
$previousCharacter = $cursor->peek(-1);
// We know from previous lines that we've advanced at least one space so far, so this next call should never be null
\assert(\is_string($previousCharacter));
$title = '';
// make sure there's a space before the title:
if (\preg_match(RegexHelper::REGEX_WHITESPACE_CHAR, $previousCharacter)) {
$title = LinkParserHelper::parseLinkTitle($cursor) ?? '';
}
$cursor->advanceToNextNonSpaceOrNewline();
if ($cursor->getCurrentCharacter() !== ')') {
$cursor->restoreState($previousState);
return null;
}
$cursor->advanceBy(1);
return ['url' => $dest, 'title' => $title];
}
private function tryParseReference(Cursor $cursor, ReferenceMapInterface $referenceMap, Bracket $opener, int $startPos): ?ReferenceInterface
{
$savePos = $cursor->saveState();
$beforeLabel = $cursor->getPosition();
$n = LinkParserHelper::parseLinkLabel($cursor);
if ($n > 2) {
$start = $beforeLabel + 1;
$length = $n - 2;
} elseif (! $opener->hasNext()) {
// Empty or missing second label means to use the first label as the reference.
// The reference must not contain a bracket. If we know there's a bracket, we don't even bother checking it.
$start = $opener->getPosition();
$length = $startPos - $start;
} else {
$cursor->restoreState($savePos);
return null;
}
$referenceLabel = $cursor->getSubstring($start, $length);
if ($n === 0) {
// If shortcut reference link, rewind before spaces we skipped
$cursor->restoreState($savePos);
}
return $referenceMap->get($referenceLabel);
}
private function createInline(string $url, string $title, bool $isImage, ?ReferenceInterface $reference = null): AbstractWebResource
{
if ($isImage) {
$inline = new Image($url, null, $title);
} else {
$inline = new Link($url, null, $title);
}
if ($reference) {
$inline->data->set('reference', $reference);
}
return $inline;
}
}
@@ -0,0 +1,42 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Parser\Inline;
use League\CommonMark\Node\Inline\Text;
use League\CommonMark\Parser\Inline\InlineParserInterface;
use League\CommonMark\Parser\Inline\InlineParserMatch;
use League\CommonMark\Parser\InlineParserContext;
final class OpenBracketParser implements InlineParserInterface
{
public function getMatchDefinition(): InlineParserMatch
{
return InlineParserMatch::string('[');
}
public function parse(InlineParserContext $inlineContext): bool
{
$inlineContext->getCursor()->advanceBy(1);
$node = new Text('[', ['delim' => true]);
$inlineContext->getContainer()->appendChild($node);
// Add entry to stack for this opener
$inlineContext->getDelimiterStack()->addBracket($node, $inlineContext->getCursor()->getPosition(), false);
return true;
}
}
@@ -0,0 +1,80 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\CommonMark\Renderer\Block;
use League\CommonMark\Extension\CommonMark\Node\Block\ListItem;
use League\CommonMark\Node\Block\AbstractBlock;
use League\CommonMark\Node\Block\Paragraph;
use League\CommonMark\Node\Block\TightBlockInterface;
use League\CommonMark\Node\Node;
use League\CommonMark\Renderer\ChildNodeRendererInterface;
use League\CommonMark\Renderer\NodeRendererInterface;
use League\CommonMark\Util\HtmlElement;
use League\CommonMark\Xml\XmlNodeRendererInterface;
final class ListItemRenderer implements NodeRendererInterface, XmlNodeRendererInterface
{
/**
* @param ListItem $node
*
* {@inheritDoc}
*
* @psalm-suppress MoreSpecificImplementedParamType
*/
public function render(Node $node, ChildNodeRendererInterface $childRenderer): \Stringable
{
ListItem::assertInstanceOf($node);
$contents = $childRenderer->renderNodes($node->children());
$inTightList = ($parent = $node->parent()) && $parent instanceof TightBlockInterface && $parent->isTight();
if ($this->needsBlockSeparator($node->firstChild(), $inTightList)) {
$contents = "\n" . $contents;
}
if ($this->needsBlockSeparator($node->lastChild(), $inTightList)) {
$contents .= "\n";
}
$attrs = $node->data->get('attributes');
return new HtmlElement('li', $attrs, $contents);
}
public function getXmlTagName(Node $node): string
{
return 'item';
}
/**
* {@inheritDoc}
*/
public function getXmlAttributes(Node $node): array
{
return [];
}
private function needsBlockSeparator(?Node $child, bool $inTightList): bool
{
if ($child instanceof Paragraph && $inTightList) {
return false;
}
return $child instanceof AbstractBlock;
}
}
@@ -0,0 +1,98 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com>
*
* Original code based on the CommonMark JS reference parser (http://bitly.com/commonmark-js)
* - (c) John MacFarlane
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\SmartPunct;
use League\CommonMark\Delimiter\Delimiter;
use League\CommonMark\Parser\Inline\InlineParserInterface;
use League\CommonMark\Parser\Inline\InlineParserMatch;
use League\CommonMark\Parser\InlineParserContext;
use League\CommonMark\Util\RegexHelper;
final class QuoteParser implements InlineParserInterface
{
/**
* @deprecated This constant is no longer used and will be removed in a future major release
*/
public const DOUBLE_QUOTES = [Quote::DOUBLE_QUOTE, Quote::DOUBLE_QUOTE_OPENER, Quote::DOUBLE_QUOTE_CLOSER];
/**
* @deprecated This constant is no longer used and will be removed in a future major release
*/
public const SINGLE_QUOTES = [Quote::SINGLE_QUOTE, Quote::SINGLE_QUOTE_OPENER, Quote::SINGLE_QUOTE_CLOSER];
public function getMatchDefinition(): InlineParserMatch
{
return InlineParserMatch::oneOf(Quote::SINGLE_QUOTE, Quote::DOUBLE_QUOTE);
}
/**
* Normalizes any quote characters found and manually adds them to the delimiter stack
*/
public function parse(InlineParserContext $inlineContext): bool
{
$char = $inlineContext->getFullMatch();
$cursor = $inlineContext->getCursor();
$index = $cursor->getPosition();
$charBefore = $cursor->peek(-1);
if ($charBefore === null) {
$charBefore = "\n";
}
$cursor->advance();
$charAfter = $cursor->getCurrentCharacter();
if ($charAfter === null) {
$charAfter = "\n";
}
[$leftFlanking, $rightFlanking] = $this->determineFlanking($charBefore, $charAfter);
$canOpen = $leftFlanking && ! $rightFlanking;
$canClose = $rightFlanking;
$node = new Quote($char, ['delim' => true]);
$inlineContext->getContainer()->appendChild($node);
// Add entry to stack to this opener
$inlineContext->getDelimiterStack()->push(new Delimiter($char, 1, $node, $canOpen, $canClose, $index));
return true;
}
/**
* @return bool[]
*/
private function determineFlanking(string $charBefore, string $charAfter): array
{
$afterIsWhitespace = \preg_match('/\pZ|\s/u', $charAfter);
$afterIsPunctuation = \preg_match(RegexHelper::REGEX_PUNCTUATION, $charAfter);
$beforeIsWhitespace = \preg_match('/\pZ|\s/u', $charBefore);
$beforeIsPunctuation = \preg_match(RegexHelper::REGEX_PUNCTUATION, $charBefore);
$leftFlanking = ! $afterIsWhitespace &&
! ($afterIsPunctuation &&
! $beforeIsWhitespace &&
! $beforeIsPunctuation);
$rightFlanking = ! $beforeIsWhitespace &&
! ($beforeIsPunctuation &&
! $afterIsWhitespace &&
! $afterIsPunctuation);
return [$leftFlanking, $rightFlanking];
}
}
@@ -0,0 +1,69 @@
<?php
declare(strict_types=1);
/*
* This file is part of the league/commonmark package.
*
* (c) Colin O'Dell <colinodell@gmail.com> and uAfrica.com (http://uafrica.com)
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\Strikethrough;
use League\CommonMark\Delimiter\DelimiterInterface;
use League\CommonMark\Delimiter\Processor\CacheableDelimiterProcessorInterface;
use League\CommonMark\Node\Inline\AbstractStringContainer;
final class StrikethroughDelimiterProcessor implements CacheableDelimiterProcessorInterface
{
public function getOpeningCharacter(): string
{
return '~';
}
public function getClosingCharacter(): string
{
return '~';
}
public function getMinLength(): int
{
return 1;
}
public function getDelimiterUse(DelimiterInterface $opener, DelimiterInterface $closer): int
{
if ($opener->getLength() > 2 && $closer->getLength() > 2) {
return 0;
}
if ($opener->getLength() !== $closer->getLength()) {
return 0;
}
// $opener and $closer are the same length so we just return one of them
return $opener->getLength();
}
public function process(AbstractStringContainer $opener, AbstractStringContainer $closer, int $delimiterUse): void
{
$strikethrough = new Strikethrough(\str_repeat('~', $delimiterUse));
$tmp = $opener->next();
while ($tmp !== null && $tmp !== $closer) {
$next = $tmp->next();
$strikethrough->appendChild($tmp);
$tmp = $next;
}
$opener->insertAfter($strikethrough);
}
public function getCacheKey(DelimiterInterface $closer): string
{
return '~' . $closer->getLength();
}
}
@@ -0,0 +1,63 @@
<?php
declare(strict_types=1);
/*
* This is part of the league/commonmark package.
*
* (c) Martin Hasoň <martin.hason@gmail.com>
* (c) Webuni s.r.o. <info@webuni.cz>
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\Table;
use League\CommonMark\Environment\EnvironmentBuilderInterface;
use League\CommonMark\Extension\ConfigurableExtensionInterface;
use League\CommonMark\Renderer\HtmlDecorator;
use League\Config\ConfigurationBuilderInterface;
use Nette\Schema\Expect;
final class TableExtension implements ConfigurableExtensionInterface
{
public function configureSchema(ConfigurationBuilderInterface $builder): void
{
$attributeArraySchema = Expect::arrayOf(
Expect::type('string|string[]|bool'), // attribute value(s)
'string' // attribute name
)->mergeDefaults(false);
$builder->addSchema('table', Expect::structure([
'wrap' => Expect::structure([
'enabled' => Expect::bool()->default(false),
'tag' => Expect::string()->default('div'),
'attributes' => Expect::arrayOf(Expect::string()),
]),
'alignment_attributes' => Expect::structure([
'left' => (clone $attributeArraySchema)->default(['align' => 'left']),
'center' => (clone $attributeArraySchema)->default(['align' => 'center']),
'right' => (clone $attributeArraySchema)->default(['align' => 'right']),
]),
'max_autocompleted_cells' => Expect::int()->min(0)->default(TableParser::DEFAULT_MAX_AUTOCOMPLETED_CELLS),
]));
}
public function register(EnvironmentBuilderInterface $environment): void
{
$tableRenderer = new TableRenderer();
if ($environment->getConfiguration()->get('table/wrap/enabled')) {
$tableRenderer = new HtmlDecorator($tableRenderer, $environment->getConfiguration()->get('table/wrap/tag'), $environment->getConfiguration()->get('table/wrap/attributes'));
}
$environment
->addBlockStartParser(new TableStartParser($environment->getConfiguration()->get('table/max_autocompleted_cells')))
->addRenderer(Table::class, $tableRenderer)
->addRenderer(TableSection::class, new TableSectionRenderer())
->addRenderer(TableRow::class, new TableRowRenderer())
->addRenderer(TableCell::class, new TableCellRenderer($environment->getConfiguration()->get('table/alignment_attributes')));
}
}
@@ -0,0 +1,212 @@
<?php
declare(strict_types=1);
/*
* This is part of the league/commonmark package.
*
* (c) Martin Hasoň <martin.hason@gmail.com>
* (c) Webuni s.r.o. <info@webuni.cz>
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\Table;
use League\CommonMark\Parser\Block\AbstractBlockContinueParser;
use League\CommonMark\Parser\Block\BlockContinue;
use League\CommonMark\Parser\Block\BlockContinueParserInterface;
use League\CommonMark\Parser\Block\BlockContinueParserWithInlinesInterface;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Parser\InlineParserEngineInterface;
use League\CommonMark\Util\ArrayCollection;
final class TableParser extends AbstractBlockContinueParser implements BlockContinueParserWithInlinesInterface
{
/**
* @internal
*/
public const DEFAULT_MAX_AUTOCOMPLETED_CELLS = 10_000;
/** @psalm-readonly */
private Table $block;
/**
* @var ArrayCollection<string>
*
* @psalm-readonly-allow-private-mutation
*/
private ArrayCollection $bodyLines;
/**
* @var array<int, string|null>
* @psalm-var array<int, TableCell::ALIGN_*|null>
* @phpstan-var array<int, TableCell::ALIGN_*|null>
*
* @psalm-readonly
*/
private array $columns;
/**
* @var array<int, string>
*
* @psalm-readonly-allow-private-mutation
*/
private array $headerCells;
/** @psalm-readonly-allow-private-mutation */
private bool $nextIsSeparatorLine = true;
private int $remainingAutocompletedCells;
/**
* @param array<int, string|null> $columns
* @param array<int, string> $headerCells
*
* @psalm-param array<int, TableCell::ALIGN_*|null> $columns
*
* @phpstan-param array<int, TableCell::ALIGN_*|null> $columns
*/
public function __construct(array $columns, array $headerCells, int $remainingAutocompletedCells = self::DEFAULT_MAX_AUTOCOMPLETED_CELLS)
{
$this->block = new Table();
$this->bodyLines = new ArrayCollection();
$this->columns = $columns;
$this->headerCells = $headerCells;
$this->remainingAutocompletedCells = $remainingAutocompletedCells;
}
public function canHaveLazyContinuationLines(): bool
{
return true;
}
public function getBlock(): Table
{
return $this->block;
}
public function tryContinue(Cursor $cursor, BlockContinueParserInterface $activeBlockParser): ?BlockContinue
{
if (\strpos($cursor->getLine(), '|') === false) {
return BlockContinue::none();
}
return BlockContinue::at($cursor);
}
public function addLine(string $line): void
{
if ($this->nextIsSeparatorLine) {
$this->nextIsSeparatorLine = false;
} else {
$this->bodyLines[] = $line;
}
}
public function parseInlines(InlineParserEngineInterface $inlineParser): void
{
$headerColumns = \count($this->headerCells);
$head = new TableSection(TableSection::TYPE_HEAD);
$this->block->appendChild($head);
$headerRow = new TableRow();
$head->appendChild($headerRow);
for ($i = 0; $i < $headerColumns; $i++) {
$cell = $this->headerCells[$i];
$tableCell = $this->parseCell($cell, $i, $inlineParser);
$tableCell->setType(TableCell::TYPE_HEADER);
$headerRow->appendChild($tableCell);
}
$body = null;
foreach ($this->bodyLines as $rowLine) {
$cells = self::split($rowLine);
$row = new TableRow();
// Body can not have more columns than head
for ($i = 0; $i < $headerColumns; $i++) {
// It can have less columns though, in which case we'll autocomplete the empty ones (up to some limit)
if (! isset($cells[$i]) && $this->remainingAutocompletedCells-- <= 0) {
// Too many cells were auto-completed, so we'll just stop here
return;
}
$cell = $cells[$i] ?? '';
$tableCell = $this->parseCell($cell, $i, $inlineParser);
$row->appendChild($tableCell);
}
if ($body === null) {
// It's valid to have a table without body. In that case, don't add an empty TableBody node.
$body = new TableSection();
$this->block->appendChild($body);
}
$body->appendChild($row);
}
}
private function parseCell(string $cell, int $column, InlineParserEngineInterface $inlineParser): TableCell
{
$tableCell = new TableCell(TableCell::TYPE_DATA, $this->columns[$column] ?? null);
if ($cell !== '') {
$inlineParser->parse(\trim($cell), $tableCell);
}
return $tableCell;
}
/**
* @internal
*
* @return array<int, string>
*/
public static function split(string $line): array
{
$cursor = new Cursor(\trim($line));
if ($cursor->getCurrentCharacter() === '|') {
$cursor->advanceBy(1);
}
$cells = [];
$sb = '';
while (! $cursor->isAtEnd()) {
switch ($c = $cursor->getCurrentCharacter()) {
case '\\':
if ($cursor->peek() === '|') {
// Pipe is special for table parsing. An escaped pipe doesn't result in a new cell, but is
// passed down to inline parsing as an unescaped pipe. Note that that applies even for the `\|`
// in an input like `\\|` - in other words, table parsing doesn't support escaping backslashes.
$sb .= '|';
$cursor->advanceBy(1);
} else {
// Preserve backslash before other characters or at end of line.
$sb .= '\\';
}
break;
case '|':
$cells[] = $sb;
$sb = '';
break;
default:
$sb .= $c;
}
$cursor->advanceBy(1);
}
if ($sb !== '') {
$cells[] = $sb;
}
return $cells;
}
}
@@ -0,0 +1,165 @@
<?php
declare(strict_types=1);
/*
* This is part of the league/commonmark package.
*
* (c) Martin Hasoň <martin.hason@gmail.com>
* (c) Webuni s.r.o. <info@webuni.cz>
* (c) Colin O'Dell <colinodell@gmail.com>
*
* For the full copyright and license information, please view the LICENSE
* file that was distributed with this source code.
*/
namespace League\CommonMark\Extension\Table;
use League\CommonMark\Parser\Block\BlockStart;
use League\CommonMark\Parser\Block\BlockStartParserInterface;
use League\CommonMark\Parser\Block\ParagraphParser;
use League\CommonMark\Parser\Cursor;
use League\CommonMark\Parser\MarkdownParserStateInterface;
final class TableStartParser implements BlockStartParserInterface
{
private int $maxAutocompletedCells;
public function __construct(int $maxAutocompletedCells = TableParser::DEFAULT_MAX_AUTOCOMPLETED_CELLS)
{
$this->maxAutocompletedCells = $maxAutocompletedCells;
}
public function tryStart(Cursor $cursor, MarkdownParserStateInterface $parserState): ?BlockStart
{
$paragraph = $parserState->getParagraphContent();
if ($paragraph === null || \strpos($paragraph, '|') === false) {
return BlockStart::none();
}
$columns = self::parseSeparator($cursor);
if (\count($columns) === 0) {
return BlockStart::none();
}
$lastLineBreak = \strrpos($paragraph, "\n");
$lastLine = $lastLineBreak === false ? $paragraph : \substr($paragraph, $lastLineBreak + 1);
$headerCells = TableParser::split($lastLine);
if (\count($headerCells) > \count($columns)) {
return BlockStart::none();
}
$cursor->advanceToEnd();
$parsers = [];
if ($lastLineBreak !== false) {
$p = new ParagraphParser();
$p->addLine(\substr($paragraph, 0, $lastLineBreak));
$parsers[] = $p;
}
$parsers[] = new TableParser($columns, $headerCells, $this->maxAutocompletedCells);
return BlockStart::of(...$parsers)
->at($cursor)
->replaceActiveBlockParser();
}
/**
* @return array<int, string|null>
*
* @psalm-return array<int, TableCell::ALIGN_*|null>
*
* @phpstan-return array<int, TableCell::ALIGN_*|null>
*/
private static function parseSeparator(Cursor $cursor): array
{
$columns = [];
$pipes = 0;
$valid = false;
while (! $cursor->isAtEnd()) {
switch ($c = $cursor->getCurrentCharacter()) {
case '|':
$cursor->advanceBy(1);
$pipes++;
if ($pipes > 1) {
// More than one adjacent pipe not allowed
return [];
}
// Need at least one pipe, even for a one-column table
$valid = true;
break;
case '-':
case ':':
if ($pipes === 0 && \count($columns) > 0) {
// Need a pipe after the first column (first column doesn't need to start with one)
return [];
}
$left = false;
$right = false;
if ($c === ':') {
$left = true;
$cursor->advanceBy(1);
}
if ($cursor->match('/^-+/') === null) {
// Need at least one dash
return [];
}
if ($cursor->getCurrentCharacter() === ':') {
$right = true;
$cursor->advanceBy(1);
}
$columns[] = self::getAlignment($left, $right);
// Next, need another pipe
$pipes = 0;
break;
case ' ':
case "\t":
// White space is allowed between pipes and columns
$cursor->advanceToNextNonSpaceOrTab();
break;
default:
// Any other character is invalid
return [];
}
}
if (! $valid) {
return [];
}
return $columns;
}
/**
* @psalm-return TableCell::ALIGN_*|null
*
* @phpstan-return TableCell::ALIGN_*|null
*
* @psalm-pure
*/
private static function getAlignment(bool $left, bool $right): ?string
{
if ($left && $right) {
return TableCell::ALIGN_CENTER;
}
if ($left) {
return TableCell::ALIGN_LEFT;
}
if ($right) {
return TableCell::ALIGN_RIGHT;
}
return null;
}
}