Select Git revision
ModuleHandler.php
Code owners
Assign users and groups as approvers for specific file changes. Learn more.
DocCommentSniff.php 21.68 KiB
<?php
/**
* Ensures doc blocks follow basic formatting.
*
* @category PHP
* @package PHP_CodeSniffer
* @link http://pear.php.net/package/PHP_CodeSniffer
*/
namespace Drupal\Sniffs\Commenting;
use PHP_CodeSniffer\Files\File;
use PHP_CodeSniffer\Sniffs\Sniff;
/**
* Ensures doc blocks follow basic formatting.
*
* Largely copied from
* \PHP_CodeSniffer\Standards\Generic\Sniffs\Commenting\DocCommentSniff,
* but Drupal @file comments are different.
*
* @category PHP
* @package PHP_CodeSniffer
* @link http://pear.php.net/package/PHP_CodeSniffer
*/
class DocCommentSniff implements Sniff
{
/**
* A list of tokenizers this sniff supports.
*
* @var array
*/
public $supportedTokenizers = [
'PHP',
'JS',
];
/**
* Returns an array of tokens this test wants to listen for.
*
* @return array
*/
public function register()
{
return [T_DOC_COMMENT_OPEN_TAG];
}//end register()
/**
* Processes this test, when one of its tokens is encountered.
*
* @param \PHP_CodeSniffer\Files\File $phpcsFile The file being scanned.
* @param int $stackPtr The position of the current token
* in the stack passed in $tokens.
*
* @return void
*/
public function process(File $phpcsFile, $stackPtr)
{
$tokens = $phpcsFile->getTokens();
$commentEnd = $phpcsFile->findNext(T_DOC_COMMENT_CLOSE_TAG, ($stackPtr + 1));
$commentStart = $tokens[$commentEnd]['comment_opener'];
$empty = [
T_DOC_COMMENT_WHITESPACE,
T_DOC_COMMENT_STAR,
];
$short = $phpcsFile->findNext($empty, ($stackPtr + 1), $commentEnd, true);
if ($short === false) {
// No content at all.
$error = 'Doc comment is empty';
$phpcsFile->addError($error, $stackPtr, 'Empty');
return;
}
// Ignore doc blocks in functions, this is handled by InlineCommentSniff.
if (empty($tokens[$stackPtr]['conditions']) === false && in_array(T_FUNCTION, $tokens[$stackPtr]['conditions']) === true) {
return;
}
// The first line of the comment should just be the /** code.
// In JSDoc there are cases with @lends that are on the same line as code.
if ($tokens[$short]['line'] === $tokens[$stackPtr]['line'] && $phpcsFile->tokenizerType !== 'JS') {
$error = 'The open comment tag must be the only content on the line';
$fix = $phpcsFile->addFixableError($error, $stackPtr, 'ContentAfterOpen');
if ($fix === true) {
$phpcsFile->fixer->beginChangeset();
$phpcsFile->fixer->addNewline($stackPtr);
$phpcsFile->fixer->addContentBefore($short, '* ');
$phpcsFile->fixer->endChangeset();
}
}
// The last line of the comment should just be the */ code.
$prev = $phpcsFile->findPrevious($empty, ($commentEnd - 1), $stackPtr, true);
if ($tokens[$commentEnd]['content'] !== '*/') {
$error = 'Wrong function doc comment end; expected "*/", found "%s"';
$phpcsFile->addError($error, $commentEnd, 'WrongEnd', [$tokens[$commentEnd]['content']]);
}
// Check for additional blank lines at the end of the comment.
if ($tokens[$prev]['line'] < ($tokens[$commentEnd]['line'] - 1)) {
$error = 'Additional blank lines found at end of doc comment';
$fix = $phpcsFile->addFixableError($error, $commentEnd, 'SpacingAfter');
if ($fix === true) {
$phpcsFile->fixer->beginChangeset();
for ($i = ($prev + 1); $i < $commentEnd; $i++) {
if ($tokens[($i + 1)]['line'] === $tokens[$commentEnd]['line']) {
break;
}
$phpcsFile->fixer->replaceToken($i, '');
}
$phpcsFile->fixer->endChangeset();
}
}
// The short description of @file comments is one line below.
if ($tokens[$short]['code'] === T_DOC_COMMENT_TAG && $tokens[$short]['content'] === '@file') {
$next = $phpcsFile->findNext($empty, ($short + 1), $commentEnd, true);
if ($next !== false) {
$fileShort = $short;
$short = $next;
}
}
// Do not check defgroup sections, they have no short description. Also don't
// check PHPUnit tests doc blocks because they might not have a description.
if (in_array($tokens[$short]['content'], ['@defgroup', '@addtogroup', '@}', '@coversDefaultClass']) === true) {
return;
}
// Check for a comment description.
if ($tokens[$short]['code'] !== T_DOC_COMMENT_STRING) {
// JSDoc has many cases of @type declaration that don't have a // description.
if ($phpcsFile->tokenizerType === 'JS') {
return;
}
// PHPUnit test methods are allowed to skip the short description and
// only provide an @covers annotation.
if ($tokens[$short]['content'] === '@covers') {
return;
}
$error = 'Missing short description in doc comment';
$phpcsFile->addError($error, $stackPtr, 'MissingShort');
return;
}
if (isset($fileShort) === true) {
$start = $fileShort;
} else {
$start = $stackPtr;
}
// No extra newline before short description.
if ($tokens[$short]['line'] !== ($tokens[$start]['line'] + 1)) {
$error = 'Doc comment short description must be on the first line';
$fix = $phpcsFile->addFixableError($error, $short, 'SpacingBeforeShort');
if ($fix === true) {
// Move file comment short description to the next line.
if (isset($fileShort) === true && $tokens[$short]['line'] === $tokens[$start]['line']) {
$phpcsFile->fixer->addContentBefore($fileShort, "\n *");
} else {
$phpcsFile->fixer->beginChangeset();
for ($i = $start; $i < $short; $i++) {
if ($tokens[$i]['line'] === $tokens[$start]['line']) {
continue;
} else if ($tokens[$i]['line'] === $tokens[$short]['line']) {
break;
}
$phpcsFile->fixer->replaceToken($i, '');
}
$phpcsFile->fixer->endChangeset();
}
}
}//end if
if ($tokens[($short - 1)]['content'] !== ' '
&& strpos($tokens[($short - 1)]['content'], $phpcsFile->eolChar) === false
) {
$error = 'Function comment short description must start with exactly one space';
$fix = $phpcsFile->addFixableError($error, $short, 'ShortStartSpace');
if ($fix === true) {
if ($tokens[($short - 1)]['code'] === T_DOC_COMMENT_WHITESPACE) {
$phpcsFile->fixer->replaceToken(($short - 1), ' ');
} else {
$phpcsFile->fixer->addContent(($short - 1), ' ');
}
}
}
// Account for the fact that a short description might cover
// multiple lines.
$shortContent = $tokens[$short]['content'];
$shortEnd = $short;
for ($i = ($short + 1); $i < $commentEnd; $i++) {
if ($tokens[$i]['code'] === T_DOC_COMMENT_STRING) {
if ($tokens[$i]['line'] === ($tokens[$shortEnd]['line'] + 1)) {
$shortContent .= $tokens[$i]['content'];
$shortEnd = $i; } else {
break;
}
}
if ($tokens[$i]['code'] === T_DOC_COMMENT_TAG) {
break;
}
}
// Remove any trailing white spaces which are detected by other sniffs.
$shortContent = trim($shortContent);
if (preg_match('|\p{Lu}|u', $shortContent[0]) === 0
// Allow both variants of inheritdoc comments.
&& $shortContent !== '{@inheritdoc}'
&& $shortContent !== '{@inheritDoc}'
// Ignore Features module export files that just use the file name as
// comment.
&& $shortContent !== basename($phpcsFile->getFilename())
) {
$error = 'Doc comment short description must start with a capital letter';
// If we cannot capitalize the first character then we don't have a
// fixable error.
if ($tokens[$short]['content'] === ucfirst($tokens[$short]['content'])) {
$phpcsFile->addError($error, $short, 'ShortNotCapital');
} else {
$fix = $phpcsFile->addFixableError($error, $short, 'ShortNotCapital');
if ($fix === true) {
$phpcsFile->fixer->replaceToken($short, ucfirst($tokens[$short]['content']));
}
}
}
$lastChar = substr($shortContent, -1);
if (in_array($lastChar, ['.', '!', '?', ')']) === false
// Allow both variants of inheritdoc comments.
&& $shortContent !== '{@inheritdoc}'
&& $shortContent !== '{@inheritDoc}'
// Ignore Features module export files that just use the file name as
// comment.
&& $shortContent !== basename($phpcsFile->getFilename())
) {
$error = 'Doc comment short description must end with a full stop';
$fix = $phpcsFile->addFixableError($error, $shortEnd, 'ShortFullStop');
if ($fix === true) {
$phpcsFile->fixer->addContent($shortEnd, '.');
}
}
if ($tokens[$short]['line'] !== $tokens[$shortEnd]['line']) {
$error = 'Doc comment short description must be on a single line, further text should be a separate paragraph';
$phpcsFile->addError($error, $shortEnd, 'ShortSingleLine');
}
$long = $phpcsFile->findNext($empty, ($shortEnd + 1), ($commentEnd - 1), true);
if ($long === false) {
return;
}
if ($tokens[$long]['code'] === T_DOC_COMMENT_STRING) {
if ($tokens[$long]['line'] !== ($tokens[$shortEnd]['line'] + 2)) {
$error = 'There must be exactly one blank line between descriptions in a doc comment';
$fix = $phpcsFile->addFixableError($error, $long, 'SpacingBetween');
if ($fix === true) {
$phpcsFile->fixer->beginChangeset();
for ($i = ($shortEnd + 1); $i < $long; $i++) {
if ($tokens[$i]['line'] === $tokens[$shortEnd]['line']) {
continue;
} else if ($tokens[$i]['line'] === ($tokens[$long]['line'] - 1)) { break;
}
$phpcsFile->fixer->replaceToken($i, '');
}
$phpcsFile->fixer->endChangeset();
}
}
if (preg_match('|\p{Lu}|u', $tokens[$long]['content'][0]) === 0
&& $tokens[$long]['content'] !== ucfirst($tokens[$long]['content'])
) {
$error = 'Doc comment long description must start with a capital letter';
$fix = $phpcsFile->addFixableError($error, $long, 'LongNotCapital');
if ($fix === true) {
$phpcsFile->fixer->replaceToken($long, ucfirst($tokens[$long]['content']));
}
}
// Account for the fact that a description might cover multiple lines.
$longContent = $tokens[$long]['content'];
$longEnd = $long;
for ($i = ($long + 1); $i < $commentEnd; $i++) {
if ($tokens[$i]['code'] === T_DOC_COMMENT_STRING) {
if ($tokens[$i]['line'] <= ($tokens[$longEnd]['line'] + 1)) {
$longContent .= $tokens[$i]['content'];
$longEnd = $i;
} else {
break;
}
}
if ($tokens[$i]['code'] === T_DOC_COMMENT_TAG) {
if ($tokens[$i]['line'] <= ($tokens[$longEnd]['line'] + 1)
// Allow link tags within the long comment itself.
&& ($tokens[$i]['content'] === '@link' || $tokens[$i]['content'] === '@endlink')
) {
$longContent .= $tokens[$i]['content'];
$longEnd = $i;
} else {
break;
}
}
}//end for
// Remove any trailing white spaces which are detected by other sniffs.
$longContent = trim($longContent);
if (preg_match('/[a-zA-Z]$/', $longContent) === 1) {
$error = 'Doc comment long description must end with a full stop';
$fix = $phpcsFile->addFixableError($error, $longEnd, 'LongFullStop');
if ($fix === true) {
$phpcsFile->fixer->addContent($longEnd, '.');
}
}
}//end if
if (empty($tokens[$commentStart]['comment_tags']) === true) {
// No tags in the comment.
return;
}
$firstTag = $tokens[$commentStart]['comment_tags'][0];
$prev = $phpcsFile->findPrevious($empty, ($firstTag - 1), $stackPtr, true);
// This does not apply to @file, @code, @link and @endlink tags.
if ($tokens[$firstTag]['line'] !== ($tokens[$prev]['line'] + 2)
&& isset($fileShort) === false
&& in_array($tokens[$firstTag]['content'], ['@code', '@link', '@endlink']) === false
) { $error = 'There must be exactly one blank line before the tags in a doc comment';
$fix = $phpcsFile->addFixableError($error, $firstTag, 'SpacingBeforeTags');
if ($fix === true) {
$phpcsFile->fixer->beginChangeset();
for ($i = ($prev + 1); $i < $firstTag; $i++) {
if ($tokens[$i]['line'] === $tokens[$firstTag]['line']) {
break;
}
$phpcsFile->fixer->replaceToken($i, '');
}
$indent = str_repeat(' ', $tokens[$stackPtr]['column']);
$phpcsFile->fixer->addContent($prev, $phpcsFile->eolChar.$indent.'*'.$phpcsFile->eolChar);
$phpcsFile->fixer->endChangeset();
}
}
// Break out the tags into groups and check alignment within each.
// A tag group is one where there are no blank lines between tags.
// The param tag group is special as it requires all @param tags to be inside.
$tagGroups = [];
$groupid = 0;
$paramGroupid = null;
$currentTag = null;
$previousTag = null;
$isNewGroup = null;
$ignoreTags = [
'@code',
'@endcode',
'@see',
];
foreach ($tokens[$commentStart]['comment_tags'] as $pos => $tag) {
if ($pos > 0) {
$prev = $phpcsFile->findPrevious(
T_DOC_COMMENT_STRING,
($tag - 1),
$tokens[$commentStart]['comment_tags'][($pos - 1)]
);
if ($prev === false) {
$prev = $tokens[$commentStart]['comment_tags'][($pos - 1)];
}
$isNewGroup = $tokens[$prev]['line'] !== ($tokens[$tag]['line'] - 1);
if ($isNewGroup === true) {
$groupid++;
}
}
$currentTag = $tokens[$tag]['content'];
if ($currentTag === '@param') {
if (($paramGroupid === null
&& empty($tagGroups[$groupid]) === false)
|| ($paramGroupid !== null
&& $paramGroupid !== $groupid)
) {
$error = 'Parameter tags must be grouped together in a doc comment';
$phpcsFile->addError($error, $tag, 'ParamGroup');
}
if ($paramGroupid === null) {
$paramGroupid = $groupid;
}
// The @param, @return and @throws tag sections should be
// separated by a blank line both before and after these sections.
} else if ($isNewGroup === false
&& (in_array($currentTag, ['@param', '@return', '@throws']) === true
|| in_array($previousTag, ['@param', '@return', '@throws']) === true) && $previousTag !== $currentTag
// Ignore code blocks in comments, they can be anywhere.
&& in_array($previousTag, $ignoreTags) === false
&& in_array($currentTag, $ignoreTags) === false
) {
$error = 'Separate the %s and %s sections by a blank line.';
$fix = $phpcsFile->addFixableError($error, $tag, 'TagGroupSpacing', [$previousTag, $currentTag]);
if ($fix === true) {
$phpcsFile->fixer->replaceToken(($tag - 1), "\n".str_repeat(' ', ($tokens[$tag]['column'] - 3)).'* ');
}
}//end if
$previousTag = $currentTag;
$tagGroups[$groupid][] = $tag;
}//end foreach
foreach ($tagGroups as $group) {
$maxLength = 0;
$paddings = [];
foreach ($group as $pos => $tag) {
$tagLength = strlen($tokens[$tag]['content']);
if ($tagLength > $maxLength) {
$maxLength = $tagLength;
}
// Check for a value. No value means no padding needed.
$string = $phpcsFile->findNext(T_DOC_COMMENT_STRING, $tag, $commentEnd);
if ($string !== false && $tokens[$string]['line'] === $tokens[$tag]['line']) {
$paddings[$tag] = strlen($tokens[($tag + 1)]['content']);
}
}
// Check that there was single blank line after the tag block
// but account for a multi-line tag comments.
$lastTag = $group[$pos];
$next = $phpcsFile->findNext(T_DOC_COMMENT_TAG, ($lastTag + 3), $commentEnd);
if ($next !== false) {
$prev = $phpcsFile->findPrevious([T_DOC_COMMENT_TAG, T_DOC_COMMENT_STRING], ($next - 1), $commentStart);
if ($tokens[$next]['line'] !== ($tokens[$prev]['line'] + 2)) {
$error = 'There must be a single blank line after a tag group';
$fix = $phpcsFile->addFixableError($error, $lastTag, 'SpacingAfterTagGroup');
if ($fix === true) {
$phpcsFile->fixer->beginChangeset();
for ($i = ($prev + 1); $i < $next; $i++) {
if ($tokens[$i]['line'] === $tokens[$next]['line']) {
break;
}
$phpcsFile->fixer->replaceToken($i, '');
}
$indent = str_repeat(' ', $tokens[$stackPtr]['column']);
$phpcsFile->fixer->addContent($prev, $phpcsFile->eolChar.$indent.'*'.$phpcsFile->eolChar);
$phpcsFile->fixer->endChangeset();
}
}
}//end if
// Now check paddings.
foreach ($paddings as $tag => $padding) {
if ($padding !== 1) {
$error = 'Tag value indented incorrectly; expected 1 space but found %s';
$data = [$padding];
$fix = $phpcsFile->addFixableError($error, ($tag + 1), 'TagValueIndent', $data);
if ($fix === true) {
$phpcsFile->fixer->replaceToken(($tag + 1), ' ');
}
}
} }//end foreach
// If there is a param group, it needs to be first; with the exception of
// @code, @todo and link tags.
if ($paramGroupid !== null && $paramGroupid !== 0
&& in_array($tokens[$tokens[$commentStart]['comment_tags'][0]]['content'], ['@code', '@todo', '@link', '@endlink', '@codingStandardsIgnoreStart']) === false
// In JSDoc we can have many other valid tags like @function or
// @constructor before the param tags.
&& $phpcsFile->tokenizerType !== 'JS'
) {
$error = 'Parameter tags must be defined first in a doc comment';
$phpcsFile->addError($error, $tagGroups[$paramGroupid][0], 'ParamNotFirst');
}
$foundTags = [];
$lastPos = 0;
foreach ($tokens[$stackPtr]['comment_tags'] as $pos => $tag) {
$tagName = $tokens[$tag]['content'];
// Skip code tags, they can be anywhere.
if (in_array($tagName, $ignoreTags) === true) {
continue;
}
if (isset($foundTags[$tagName]) === true) {
$lastTag = $tokens[$stackPtr]['comment_tags'][$lastPos];
if ($tokens[$lastTag]['content'] !== $tagName) {
$error = 'Tags must be grouped together in a doc comment';
$phpcsFile->addError($error, $tag, 'TagsNotGrouped');
}
}
$foundTags[$tagName] = true;
$lastPos = $pos;
}
}//end process()
}//end class