wp-includes/SimplePie/src/IRI.php:374Replace invalid character with percent encoding
$stringstring$extra_charsstring$iprivatebooloptionalfalsestring protected function replace_invalid_with_pct_encoding(string $string, string $extra_chars, bool $iprivate = false) { // Normalize as many pct-encoded sections as possible $string = preg_replace_callback('/(?:%[A-Fa-f0-9]{2})+/', [$this, 'remove_iunreserved_percent_encoded'], $string); \assert(\is_string($string), "For PHPStan: Should not occur, the regex is valid"); // Replace invalid percent characters $string = preg_replace('/%(?![A-Fa-f0-9]{2})/', '%25', $string); \assert(\is_string($string), "For PHPStan: Should not occur, the regex is valid"); // Add unreserved and % to $extra_chars (the latter is safe because all // pct-encoded sections are now valid). $extra_chars .= 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-._~%'; // Now replace any bytes that aren't allowed with their pct-encoded versions $position = 0; $strlen = strlen($string); while (($position += strspn($string, $extra_chars, $position)) < $strlen) { $value = ord($string[$position]); $character = 0; // Start position $start = $position; // By default we are valid $valid = true; // No one byte sequences are valid due to the while. // Two byte sequence: if (($value & 0xE0) === 0xC0) { $character = ($value & 0x1F) << 6; $length = 2; $remaining = 1; } // Three byte sequence: elseif (($value & 0xF0) === 0xE0) { $character = ($value & 0x0F) << 12; $length = 3; $remaining = 2; } // Four byte sequence: elseif (($value & 0xF8) === 0xF0) { $character = ($value & 0x07) << 18; $length = 4; $remaining = 3; } // Invalid byte: else { $valid = false; $length = 1; $remaining = 0; } if ($remaining) { if ($position + $length <= $strlen) { for ($position++; $remaining; $position++) { $value = ord($string[$position]); // Check that the byte is valid, then add it to the character: if (($value & 0xC0) === 0x80) { $character |= ($value & 0x3F) << (--$remaining * 6); } // If it is invalid, count the sequence as invalid and reprocess the current byte: else { $valid = false; $position--; break; } } } else { $position = $strlen - 1; $valid = false; } } // Percent encode anything invalid or not in ucschar if ( // Invalid sequences !$valid // Non-shortest form sequences are invalidUnchanged from 6.7.7 through 7.1.0.
Signature, return type and hooks compared across 5 parsed releases.
src/wp-includes/SimplePie/src/IRI.php, and regenerated for each WordPress release so it tracks the code rather than a snapshot of it.