<?php
/**
* WordPress Coding Standard.
*
* @package WPCS\WordPressCodingStandards
* @link https://github.com/WordPress/WordPress-Coding-Standards
* @license https://opensource.org/licenses/MIT MIT
*/
namespace WordPressCS\WordPress\Sniffs\WP;
use PHP_CodeSniffer\Util\Tokens;
use PHPCSUtils\BackCompat\Helper;
use PHPCSUtils\Utils\MessageHelper;
use PHPCSUtils\Utils\PassedParameters;
use PHPCSUtils\Utils\TextStrings;
use WordPressCS\WordPress\AbstractFunctionParameterSniff;
use WordPressCS\WordPress\Helpers\RulesetPropertyHelper;
use XMLReader;
/**
* Makes sure WP internationalization functions are used properly.
*
* @link https://make.wordpress.org/core/handbook/best-practices/internationalization/
* @link https://developer.wordpress.org/plugins/internationalization/
*
* @since 0.10.0
* @since 0.11.0 - Now also checks for translators comments.
* - Now has the ability to handle text domain set via the command-line
* as a comma-delimited list.
* `phpcs --runtime-set text_domain my-slug,default`
* @since 0.13.0 Class name changed: this class is now namespaced.
* @since 1.0.0 This class now extends the WordPressCS native
* `AbstractFunctionRestrictionSniff` class.
* The parent `exclude` property is, however, disabled as it
* would disable the whole sniff.
* @since 3.0.0 This class now extends the WordPressCS native
* `AbstractFunctionParameterSniff` class.
* The parent `exclude` property is still disabled.
*/
final class I18nSniff extends AbstractFunctionParameterSniff {
/**
* These Regexes were originally copied from https://www.php.net/function.sprintf#93552
* and adjusted for better precision and updated specs.
*/
const SPRINTF_PLACEHOLDER_REGEX = '/(?:
(?<!%) # Don\'t match a literal % (%%).
(
% # Start of placeholder.
(?:[0-9]+\$)? # Optional ordering of the placeholders.
[+-]? # Optional sign specifier.
(?:
(?:0|\'.)? # Optional padding specifier - excluding the space.
-? # Optional alignment specifier.
[0-9]* # Optional width specifier.
(?:\.(?:[ 0]|\'.)?[0-9]+)? # Optional precision specifier with optional padding character.
| # Only recognize the space as padding in combination with a width specifier.
(?:[ ])? # Optional space padding specifier.
-? # Optional alignment specifier.
[0-9]+ # Width specifier.
(?:\.(?:[ 0]|\'.)?[0-9]+)? # Optional precision specifier with optional padding character.
)
[bcdeEfFgGhHosuxX] # Type specifier.
)
)/x';
/**
* "Unordered" means there's no position specifier: '%s', not '%2$s'.
*/
const UNORDERED_SPRINTF_PLACEHOLDER_REGEX = '/(?:
(?<!%) # Don\'t match a literal % (%%).
% # Start of placeholder.
[+-]? # Optional sign specifier.
(?:
(?:0|\'.)? # Optional padding specifier - excluding the space.
-? # Optional alignment specifier.
[0-9]* # Optional width specifier.
(?:\.(?:[ 0]|\'.)?[0-9]+)? # Optional precision specifier with optional padding character.
| # Only recognize the space as padding in combination with a width specifier.
(?:[ ])? # Optional space padding specifier.
-? # Optional alignment specifier.
[0-9]+ # Width specifier.
(?:\.(?:[ 0]|\'.)?[0-9]+)? # Optional precision specifier with optional padding character.
)
[bcdeEfFgGhHosuxX] # Type specifier.
)/x';
/**
* Text domain.
*
* @var string[]
*/
public $text_domain;
/**
* The I18N functions in use in WP.
*
* @since 0.10.0
* @since 0.11.0 Changed visibility from public to protected.
*
* @var array<string, string> Key is function name, value is the function type.
*/
protected $i18n_functions = array(
'translate' => 'simple',
'__' => 'simple',
'esc_attr__' => 'simple',
'esc_html__' => 'simple',
'_e' => 'simple',
'esc_attr_e' => 'simple',
'esc_html_e' => 'simple',
'translate_with_gettext_context' => 'context',
'_x' => 'context',
'_ex' => 'context',
'esc_attr_x' => 'context',
'esc_html_x' => 'context',
'_n' => 'number',
'_nx' => 'number_context',
'_n_noop' => 'noopnumber',
'_nx_noop' => 'noopnumber_context',
);
/**
* Whether or not the `default` text domain is one of the allowed text domains.
*
* @since 0.14.0
*
* @var bool
*/
private $text_domain_contains_default = false;
/**
* Whether or not the `default` text domain is the only allowed text domain.
*
* @since 0.14.0
*
* @var bool
*/
private $text_domain_is_default = false;
/**
* Parameter specifications for the functions in each group.
*
* {@internal Even when not all parameters will be examined, the parameter list should still
* be complete in the below array to allow for a correct "total parameters" calculation.}
*
* @since 3.0.0
*
* @var array<string, array> Array of the parameter positions and names.
*/
private $parameter_specs = array(
'simple' => array(
1 => 'text',
2 => 'domain',
),
'context' => array(
1 => 'text',
2 => 'context',
3 => 'domain',
),
'number' => array(
1 => 'single',
2 => 'plural',
3 => 'number',
4 => 'domain',
),
'number_context' => array(
1 => 'single',
2 => 'plural',
3 => 'number',
4 => 'context',
5 => 'domain',
),
'noopnumber' => array(
1 => 'singular',
2 => 'plural',
3 => 'domain',
),
'noopnumber_context' => array(
1 => 'singular',
2 => 'plural',
3 => 'context',
4 => 'domain',
),
);
/**
* Groups of functions to restrict.
*
* Example: groups => array(
* 'lambda' => array(
* 'type' => 'error' | 'warning',
* 'message' => 'Use anonymous functions instead please!',
* 'functions' => array( 'file_get_contents', 'create_function' ),
* )
* )
*
* @return array
*/
public function getGroups() {
return array(
'i18n' => array(
'functions' => array_keys( $this->i18n_functions ),
),
'typos' => array(
'functions' => array(
'_',
),
),
);
}
/**
* Processes this test, when one of its tokens is encountered.
*
* @since 1.0.0 Defers to the abstractFunctionRestriction sniff for determining
* whether something is a function call. The logic after that has
* been split off to the `process_matched_token()` method.
*
* @param int $stackPtr The position of the current token in the stack.
*
* @return void
*/
public function process_token( $stackPtr ) {
// Reset defaults.
$this->text_domain_contains_default = false;
$this->text_domain_is_default = false;
// Allow overruling the text_domain set in a ruleset via the command line.
$cl_text_domain = Helper::getConfigData( 'text_domain' );
if ( ! empty( $cl_text_domain ) ) {
$cl_text_domain = trim( $cl_text_domain );
if ( '' !== $cl_text_domain ) {
$this->text_domain = array_filter( array_map( 'trim', explode( ',', $cl_text_domain ) ) );
}
}
$this->text_domain = RulesetPropertyHelper::merge_custom_array( $this->text_domain, array(), false );
if ( ! empty( $this->text_domain ) ) {
if ( \in_array( 'default', $this->text_domain, true ) ) {
$this->text_domain_contains_default = true;
if ( \count( $this->text_domain ) === 1 ) {
$this->text_domain_is_default = true;
}
}
}
// Prevent exclusion of the i18n group.
$this->exclude = array();
parent::process_token( $stackPtr );
}
/**
* Process a matched token.
*
* @since 1.0.0 Logic split off from the `process_token()` method.
*
* @param int $stackPtr The position of the current token in the stack.
* @param string $group_name The name of the group which was matched.
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
*
* @return void
*/
public function process_matched_token( $stackPtr, $group_name, $matched_content ) {
$func_open_paren_token = $this->phpcsFile->findNext( Tokens::$emptyTokens, ( $stackPtr + 1 ), null, true );
if ( ! isset( $this->tokens[ $func_open_paren_token ]['parenthesis_closer'] ) ) {
// Live coding, parse error or not a function call.
return;
}
if ( 'typos' === $group_name && '_' === $matched_content ) {
$this->phpcsFile->addError(
'Found single-underscore "_()" function when double-underscore expected.',
$stackPtr,
'SingleUnderscoreGetTextFunction'
);
return;
}
if ( 'translate' === $matched_content || 'translate_with_gettext_context' === $matched_content ) {
$this->phpcsFile->addWarning(
'Use of the "%s()" function is reserved for low-level API usage.',
$stackPtr,
'LowLevelTranslationFunction',
array( $matched_content )
);
}
parent::process_matched_token( $stackPtr, $group_name, $matched_content );
}
/**
* Process the function if no parameters were found.
*
* @since 3.0.0
*
* @param int $stackPtr The position of the current token in the stack.
* @param string $group_name The name of the group which was matched.
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
*
* @return void
*/
public function process_no_parameters( $stackPtr, $group_name, $matched_content ) {
$function_param_specs = $this->parameter_specs[ $this->i18n_functions[ $matched_content ] ];
foreach ( $function_param_specs as $param_name ) {
$error_code = MessageHelper::stringToErrorcode( 'MissingArg' . ucfirst( $param_name ) );
$this->phpcsFile->addError(
'Missing $%s parameter in function call to %s().',
$stackPtr,
$error_code,
array( $param_name, $matched_content )
);
}
}
/**
* Process the parameters of a matched function.
*
* @since 3.0.0
*
* @param int $stackPtr The position of the current token in the stack.
* @param string $group_name The name of the group which was matched.
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param array $parameters Array with information about the parameters.
*
* @return void
*/
public function process_parameters( $stackPtr, $group_name, $matched_content, $parameters ) {
/*
* Retrieve the individual parameters from the array in a way that we know which is which.
*/
$parameter_details = array();
$function_param_specs = $this->parameter_specs[ $this->i18n_functions[ $matched_content ] ];
$expected_args = count( $function_param_specs );
foreach ( $function_param_specs as $position => $name ) {
if ( 'number' === $name ) {
// This sniff does not examine the $number parameter.
continue;
}
$parameter_details[ $name ] = PassedParameters::getParameterFromStack( $parameters, $position, $name );
}
/*
* Examine the individual parameters.
*/
$this->check_argument_count( $stackPtr, $matched_content, $parameters, $expected_args );
foreach ( $parameter_details as $param_name => $param_info ) {
$is_string_literal = $this->check_argument_is_string_literal( $stackPtr, $matched_content, $param_name, $param_info );
/*
* If the parameter exists, remember whether the argument was a valid string literal.
* This is used in a few places to determine whether the checks which examine a text string should run.
*/
if ( false !== $param_info ) {
$parameter_details[ $param_name ]['is_string_literal'] = $is_string_literal;
}
if ( false === $is_string_literal ) {
continue;
}
if ( 'domain' === $param_name ) {
$this->check_textdomain_matches( $matched_content, $param_name, $param_info );
}
if ( \in_array( $param_name, array( 'text', 'single', 'singular', 'plural' ), true ) ) {
$this->check_placeholders_in_string( $matched_content, $param_name, $param_info );
$has_content = $this->check_string_has_translatable_content( $matched_content, $param_name, $param_info );
if ( true === $has_content ) {
$this->check_string_has_no_html_wrapper( $matched_content, $param_name, $param_info );
}
}
}
/*
* For _n*() calls, compare the singular and plural strings.
*
* If either of the arguments is missing, empty or has more than 1 token, skip out.
* An error for that will already have been reported via the `check_argument_is_string_literal()` method.
*/
$single_details = null;
if ( isset( $parameter_details['single'] ) ) {
$single_details = $parameter_details['single'];
} elseif ( isset( $parameter_details['singular'] ) ) {
$single_details = $parameter_details['singular'];
}
if ( isset( $single_details, $parameter_details['plural'] )
&& false !== $single_details
&& false !== $parameter_details['plural']
&& true === $single_details['is_string_literal']
&& true === $parameter_details['plural']['is_string_literal']
) {
$this->compare_single_and_plural_arguments( $stackPtr, $single_details, $parameter_details['plural'] );
}
/*
* Check if a translators comments is necessary and if so, if it exists.
*/
$this->check_for_translator_comment( $stackPtr, $matched_content, $parameter_details );
}
/**
* Verify that there are no superfluous function arguments.
*
* @since 3.0.0 Check moved from the `process_matched_token()` method to this method.
*
* @param int $stackPtr The position of the current token in the stack.
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param array $parameters The parameters array.
* @param int $expected_count The expected number of passed arguments.
*
* @return void
*/
private function check_argument_count( $stackPtr, $matched_content, $parameters, $expected_count ) {
$actual_count = count( $parameters );
if ( $actual_count > $expected_count ) {
$this->phpcsFile->addError(
'Too many parameters passed to function "%s()". Expected: %s parameters, received: %s',
$stackPtr,
'TooManyFunctionArgs',
array( $matched_content, $expected_count, $actual_count )
);
}
}
/**
* Check if an arbitrary function call parameter is a text string literal suitable for use in the translation functions.
*
* Will also check and warn about missing parameters.
*
* @since 3.0.0 Most of the logic in this method used to be contained in the, now removed, `check_argument_tokens()` method.
*
* @param int $stackPtr The position of the current token in the stack.
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param string $param_name The name of the parameter being examined.
* @param array|false $param_info Parameter info array for an individual parameter,
* as received from the PassedParameters class.
*
* @return bool Whether or not the argument is a string literal.
*/
private function check_argument_is_string_literal( $stackPtr, $matched_content, $param_name, $param_info ) {
/*
* Check if the parameter was supplied.
*/
if ( false === $param_info || '' === $param_info['clean'] ) {
$error_code = MessageHelper::stringToErrorcode( 'MissingArg' . ucfirst( $param_name ) );
/*
* Special case the text domain parameter, which is allowed to be "missing"
* when set to `default` (= WP Core translation).
*/
if ( 'domain' === $param_name ) {
if ( empty( $this->text_domain ) ) {
// If no text domain is passed, presume WP Core.
return false;
}
if ( true === $this->text_domain_is_default ) {
return false;
}
if ( true === $this->text_domain_contains_default ) {
$this->phpcsFile->addWarning(
'Missing $%s parameter in function call to %s(). If this text string is supposed to use a WP Core translation, use the "default" text domain.',
$stackPtr,
$error_code . 'Default',
array( $param_name, $matched_content )
);
return false;
}
}
$this->phpcsFile->addError(
'Missing $%s parameter in function call to %s().',
$stackPtr,
$error_code,
array( $param_name, $matched_content )
);
return false;
}
/*
* Check if the parameter consists of one singular text string literal.
* Heredoc/nowdocs not allowed. Multi-line single/double quoted strings are allowed.
*/
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
for ( $i = $first_non_empty; $i <= $param_info['end']; $i++ ) {
if ( isset( Tokens::$emptyTokens[ $this->tokens[ $i ]['code'] ] ) ) {
continue;
}
if ( isset( Tokens::$stringTokens[ $this->tokens[ $i ]['code'] ] ) === false ) {
$error_code = MessageHelper::stringToErrorcode( 'NonSingularStringLiteral' . ucfirst( $param_name ) );
$this->phpcsFile->addError(
'The $%s parameter must be a single text string literal. Found: %s',
$first_non_empty,
$error_code,
array( $param_name, $param_info['clean'] )
);
return false;
}
}
/*
* Make sure the text string does not contain any interpolated variable.
*/
if ( \T_DOUBLE_QUOTED_STRING === $this->tokens[ $first_non_empty ]['code'] ) {
$error_code = MessageHelper::stringToErrorcode( 'InterpolatedVariable' . ucfirst( $param_name ) );
$interpolated_variables = TextStrings::getEmbeds( $param_info['clean'] );
foreach ( $interpolated_variables as $interpolated_variable ) {
$this->phpcsFile->addError(
'The $%s parameter must not contain interpolated variables or expressions. Found: %s',
$first_non_empty,
$error_code,
array( $param_name, $interpolated_variable )
);
}
if ( ! empty( $interpolated_variables ) ) {
return false;
}
}
return true;
}
/**
* Check the correct text domain is being used.
*
* @since 3.0.0 The logic in this method used to be contained in the, now removed, `check_argument_tokens()` method.
*
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param string $param_name The name of the parameter being examined.
* @param array|false $param_info Parameter info array for an individual parameter,
* as received from the PassedParameters class.
*
* @return void
*/
private function check_textdomain_matches( $matched_content, $param_name, $param_info ) {
$stripped_content = TextStrings::stripQuotes( $param_info['clean'] );
if ( empty( $this->text_domain ) && '' === $stripped_content ) {
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
$this->phpcsFile->addError(
'The passed $domain should never be an empty string. Either pass a text domain or remove the parameter.',
$first_non_empty,
'EmptyTextDomain'
);
}
if ( empty( $this->text_domain ) ) {
// Nothing more to do, the other checks all depend on a text domain being known.
return;
}
if ( ! \in_array( $stripped_content, $this->text_domain, true ) ) {
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
$this->phpcsFile->addError(
'Mismatched text domain. Expected \'%s\' but got %s.',
$first_non_empty,
'TextDomainMismatch',
array( implode( "' or '", $this->text_domain ), $param_info['clean'] )
);
return;
}
if ( true === $this->text_domain_is_default && 'default' === $stripped_content ) {
$fixable = false;
$error = 'No need to supply the text domain in function call to %s() when the only accepted text domain is "default".';
$error_code = 'SuperfluousDefaultTextDomain';
$data = array( $matched_content );
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
// Prevent removing comments when auto-fixing.
$remove_from = ( $param_info['start'] - 1 );
$remove_to = $first_non_empty;
if ( isset( $param_info['name_token'] ) ) {
$remove_from = $this->phpcsFile->findPrevious( Tokens::$emptyTokens, ( $param_info['name_token'] - 1 ), null, true );
if ( \T_OPEN_PARENTHESIS === $this->tokens[ $remove_from ]['code'] ) {
++$remove_from; // Don't remove the open parenthesis.
/*
* Named param as first param in the function call, if we fix this, we need to
* remove the comma _after_ the parameter as well to prevent creating a parse error.
*/
$remove_to = $param_info['end'];
if ( \T_COMMA === $this->tokens[ ( $param_info['end'] + 1 ) ]['code'] ) {
++$remove_to; // Include the comma.
}
}
}
// Now, make sure there are no comments in the tokens we want to remove.
if ( $this->phpcsFile->findNext( Tokens::$commentTokens, $remove_from, ( $remove_to + 1 ) ) === false ) {
$fixable = true;
}
if ( false === $fixable ) {
$this->phpcsFile->addWarning( $error, $first_non_empty, $error_code, $data );
return;
}
$fix = $this->phpcsFile->addFixableWarning( $error, $first_non_empty, $error_code, $data );
if ( true === $fix ) {
$this->phpcsFile->fixer->beginChangeset();
for ( $i = $remove_from; $i <= $remove_to; $i++ ) {
$this->phpcsFile->fixer->replaceToken( $i, '' );
}
$this->phpcsFile->fixer->endChangeset();
}
}
}
/**
* Check the placeholders used in translatable text for common problems.
*
* @since 3.0.0 The logic in this method used to be contained in the, now removed, `check_text()` method.
*
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param string $param_name The name of the parameter being examined.
* @param array|false $param_info Parameter info array for an individual parameter,
* as received from the PassedParameters class.
*
* @return void
*/
private function check_placeholders_in_string( $matched_content, $param_name, $param_info ) {
$content = $param_info['clean'];
// UnorderedPlaceholders: Check for multiple unordered placeholders.
$unordered_matches_count = preg_match_all( self::UNORDERED_SPRINTF_PLACEHOLDER_REGEX, $content, $unordered_matches );
$unordered_matches = $unordered_matches[0];
$all_matches_count = preg_match_all( self::SPRINTF_PLACEHOLDER_REGEX, $content, $all_matches );
if ( $unordered_matches_count > 0
&& $unordered_matches_count !== $all_matches_count
&& $all_matches_count > 1
) {
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
$error_code = MessageHelper::stringToErrorcode( 'MixedOrderedPlaceholders' . ucfirst( $param_name ) );
$this->phpcsFile->addError(
'Multiple placeholders in translatable strings should be ordered. Mix of ordered and non-ordered placeholders found. Found: "%s" in %s.',
$first_non_empty,
$error_code,
array( implode( ', ', $all_matches[0] ), $param_info['clean'] )
);
return;
}
if ( $unordered_matches_count >= 2 ) {
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
$error_code = MessageHelper::stringToErrorcode( 'UnorderedPlaceholders' . ucfirst( $param_name ) );
$suggestions = array();
$replace_regexes = array();
$replacements = array();
for ( $i = 0; $i < $unordered_matches_count; $i++ ) {
$to_insert = ( $i + 1 );
$to_insert .= ( '"' !== $content[0] ) ? '$' : '\$';
$suggestions[ $i ] = substr_replace( $unordered_matches[ $i ], $to_insert, 1, 0 );
// Prepare the strings for use in a regex.
$replace_regexes[ $i ] = '`\Q' . $unordered_matches[ $i ] . '\E`';
// Note: the initial \\ is a literal \, the four \ in the replacement translate also to a literal \.
$replacements[ $i ] = str_replace( '\\', '\\\\', $suggestions[ $i ] );
// Note: the $ needs escaping to prevent numeric sequences after the $ being interpreted as match replacements.
$replacements[ $i ] = str_replace( '$', '\\$', $replacements[ $i ] );
}
$fix = $this->phpcsFile->addFixableError(
'Multiple placeholders in translatable strings should be ordered. Expected "%s", but got "%s" in %s.',
$first_non_empty,
$error_code,
array( implode( ', ', $suggestions ), implode( ', ', $unordered_matches ), $param_info['clean'] )
);
if ( true === $fix ) {
$this->phpcsFile->fixer->beginChangeset();
$fixed_str = preg_replace( $replace_regexes, $replacements, $content, 1 );
$this->phpcsFile->fixer->replaceToken( $first_non_empty, $fixed_str );
$i = ( $first_non_empty + 1 );
while ( $i <= $param_info['end'] && isset( Tokens::$stringTokens[ $this->tokens[ $i ]['code'] ] ) ) {
$this->phpcsFile->fixer->replaceToken( $i, '' );
++$i;
}
$this->phpcsFile->fixer->endChangeset();
}
}
}
/**
* Check if a parameter which is supposed to hold translatable text actually has translatable text.
*
* @since 3.0.0 The logic in this method used to be contained in the, now removed, `check_text()` method.
*
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param string $param_name The name of the parameter being examined.
* @param array|false $param_info Parameter info array for an individual parameter,
* as received from the PassedParameters class.
*
* @return bool Whether or not the text string has translatable content.
*/
private function check_string_has_translatable_content( $matched_content, $param_name, $param_info ) {
// Strip placeholders and surrounding quotes.
$content_without_quotes = trim( TextStrings::stripQuotes( $param_info['clean'] ) );
$non_placeholder_content = preg_replace( self::SPRINTF_PLACEHOLDER_REGEX, '', $content_without_quotes );
if ( '' === $non_placeholder_content ) {
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
$this->phpcsFile->addError(
'The $%s text string should have translatable content. Found: %s',
$first_non_empty,
'NoEmptyStrings',
array( $param_name, $param_info['clean'] )
);
return false;
}
return true;
}
/**
* Ensure that a translatable text string is not wrapped in HTML code.
*
* If the text is wrapped in HTML, the HTML should be moved out of the translatable text string.
*
* @since 3.0.0 The logic in this method used to be contained in the, now removed, `check_text()` method.
*
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param string $param_name The name of the parameter being examined.
* @param array|false $param_info Parameter info array for an individual parameter,
* as received from the PassedParameters class.
*
* @return void
*/
private function check_string_has_no_html_wrapper( $matched_content, $param_name, $param_info ) {
// Strip surrounding quotes.
$content_without_quotes = trim( TextStrings::stripQuotes( $param_info['clean'] ) );
$reader = new XMLReader();
$reader->XML( $content_without_quotes, 'UTF-8', \LIBXML_NOERROR | \LIBXML_ERR_NONE | \LIBXML_NOWARNING );
// Is the first node an HTML element?
if ( ! $reader->read() || XMLReader::ELEMENT !== $reader->nodeType ) {
return;
}
// If the opening HTML element includes placeholders in its attributes, we don't warn.
// E.g. '<option id="%1$s" value="%2$s">Translatable option name</option>'.
$i = 0;
while ( $attr = $reader->getAttributeNo( $i ) ) {
if ( preg_match( self::SPRINTF_PLACEHOLDER_REGEX, $attr ) === 1 ) {
return;
}
++$i;
}
// We don't flag strings wrapped in `<a href="...">...</a>`, as the link target might actually need localization.
if ( 'a' === $reader->name && $reader->getAttribute( 'href' ) ) {
return;
}
// Does the entire string only consist of this HTML node?
if ( $reader->readOuterXml() === $content_without_quotes ) {
$first_non_empty = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info['start'], ( $param_info['end'] + 1 ), true );
$this->phpcsFile->addWarning(
'Translatable string should not be wrapped in HTML. Found: %s',
$first_non_empty,
'NoHtmlWrappedStrings',
array( $param_info['clean'] )
);
}
}
/**
* Check for inconsistencies in the placeholders between single and plural form of the translatable text string.
*
* @since 3.0.0 - The parameter names and expected format for the $param_info_single
* and the $param_info_plural parameters has changed.
* - The method visibility has been changed from `protected` to `private`.
*
* @param int $stackPtr The position of the function call token in the stack.
* @param array $param_info_single Parameter info array for the `$single` parameter,
* as received from the PassedParameters class.
* @param array $param_info_plural Parameter info array for the `$plural` parameter,
* as received from the PassedParameters class.
*
* @return void
*/
private function compare_single_and_plural_arguments( $stackPtr, $param_info_single, $param_info_plural ) {
$single_content = $param_info_single['clean'];
$plural_content = $param_info_plural['clean'];
preg_match_all( self::SPRINTF_PLACEHOLDER_REGEX, $single_content, $single_placeholders );
$single_placeholders = $single_placeholders[0];
preg_match_all( self::SPRINTF_PLACEHOLDER_REGEX, $plural_content, $plural_placeholders );
$plural_placeholders = $plural_placeholders[0];
// English conflates "singular" with "only one", described in the codex:
// https://codex.wordpress.org/I18n_for_WordPress_Developers#Plurals .
if ( \count( $single_placeholders ) < \count( $plural_placeholders ) ) {
$error_string = 'Missing singular placeholder, needed for some languages. See https://codex.wordpress.org/I18n_for_WordPress_Developers#Plurals';
$first_non_empty_single = $this->phpcsFile->findNext( Tokens::$emptyTokens, $param_info_single['start'], ( $param_info_single['end'] + 1 ), true );
$this->phpcsFile->addError( $error_string, $first_non_empty_single, 'MissingSingularPlaceholder' );
return;
}
// Reordering is fine, but mismatched placeholders is probably wrong.
sort( $single_placeholders, \SORT_NATURAL );
sort( $plural_placeholders, \SORT_NATURAL );
if ( $single_placeholders !== $plural_placeholders ) {
$this->phpcsFile->addWarning( 'Mismatched placeholders is probably an error', $stackPtr, 'MismatchedPlaceholders' );
}
}
/**
* Check for the presence of a translators comment if one of the text strings contains a placeholder.
*
* @since 3.0.0 - The parameter names and expected format for the $parameters parameter has changed.
* - The method visibility has been changed from `protected` to `private`.
*
* @param int $stackPtr The position of the gettext call token in the stack.
* @param string $matched_content The token content (function name) which was matched
* in lowercase.
* @param array $parameters Array with information about the parameters.
*
* @return void
*/
private function check_for_translator_comment( $stackPtr, $matched_content, $parameters ) {
$needs_translators_comment = false;
foreach ( $parameters as $param_name => $param_info ) {
if ( false === \in_array( $param_name, array( 'text', 'single', 'singular', 'plural' ), true ) ) {
continue;
}
if ( false === $param_info || false === $param_info['is_string_literal'] ) {
continue;
}
if ( preg_match( self::SPRINTF_PLACEHOLDER_REGEX, $param_info['clean'] ) === 1 ) {
$needs_translators_comment = true;
break;
}
}
if ( false === $needs_translators_comment ) {
// No text string with placeholders found, no translation comment needed.
return;
}
$previous_comment = $this->phpcsFile->findPrevious( Tokens::$commentTokens, ( $stackPtr - 1 ) );
if ( false !== $previous_comment ) {
/*
* Check that the comment is either on the line before the gettext call or
* if it's not, that there is only whitespace between.
*/
$correctly_placed = false;
if ( ( $this->tokens[ $previous_comment ]['line'] + 1 ) === $this->tokens[ $stackPtr ]['line'] ) {
$correctly_placed = true;
} else {
$next_non_whitespace = $this->phpcsFile->findNext( \T_WHITESPACE, ( $previous_comment + 1 ), $stackPtr, true );
if ( false === $next_non_whitespace || $this->tokens[ $next_non_whitespace ]['line'] === $this->tokens[ $stackPtr ]['line'] ) {
// No non-whitespace found or next non-whitespace is on same line as gettext call.
$correctly_placed = true;
}
unset( $next_non_whitespace );
}
/*
* Check that the comment starts with 'translators:'.
*/
if ( true === $correctly_placed ) {
if ( \T_COMMENT === $this->tokens[ $previous_comment ]['code'] ) {
$comment_text = trim( $this->tokens[ $previous_comment ]['content'] );
// If it's multi-line /* */ comment, collect all the parts.
if ( '*/' === substr( $comment_text, -2 ) && '/*' !== substr( $comment_text, 0, 2 ) ) {
for ( $i = ( $previous_comment - 1 ); 0 <= $i; $i-- ) {
if ( \T_COMMENT !== $this->tokens[ $i ]['code'] ) {
break;
}
$comment_text = trim( $this->tokens[ $i ]['content'] ) . $comment_text;
}
}
if ( true === $this->is_translators_comment( $comment_text ) ) {
// Comment is ok.
return;
}
}
if ( \T_DOC_COMMENT_CLOSE_TAG === $this->tokens[ $previous_comment ]['code'] ) {
// If it's docblock comment (wrong style) make sure that it's a translators comment.
if ( isset( $this->tokens[ $previous_comment ]['comment_opener'] ) ) {
$db_start = $this->tokens[ $previous_comment ]['comment_opener'];
} else {
$db_start = $this->phpcsFile->findPrevious( \T_DOC_COMMENT_OPEN_TAG, ( $previous_comment - 1 ) );
}
$db_first_text = $this->phpcsFile->findNext( \T_DOC_COMMENT_STRING, ( $db_start + 1 ), $previous_comment );
if ( true === $this->is_translators_comment( $this->tokens[ $db_first_text ]['content'] ) ) {
$this->phpcsFile->addWarning(
'A "translators:" comment must be a "/* */" style comment. Docblock comments will not be picked up by the tools to generate a ".pot" file.',
$stackPtr,
'TranslatorsCommentWrongStyle'
);
return;
}
}
}
}
// Found placeholders but no translators comment.
$this->phpcsFile->addWarning(
'A function call to %s() with texts containing placeholders was found, but was not accompanied by a "translators:" comment on the line above to clarify the meaning of the placeholders.',
$stackPtr,
'MissingTranslatorsComment',
array( $matched_content )
);
}
/**
* Check if a (collated) comment string starts with 'translators:'.
*
* @since 0.11.0
*
* @param string $content Comment string content.
*
* @return bool
*/
private function is_translators_comment( $content ) {
if ( preg_match( '`^(?:(?://|/\*{1,2}) )?translators:`i', $content ) === 1 ) {
return true;
}
return false;
}
}