ANSI sanitizing user input by removing unsupported Unicode characters for a Latin1 encoded PostgreSQL database.

test.php

<?php
function sanitizeLatin1($input) {
    $output = '';
    $unsupportedChars = [];
    
    // Loop through each character in the input string
    for ($i = 0; $i < mb_strlen($input, 'UTF-8'); $i++) {
        $char = mb_substr($input, $i, 1, 'UTF-8');
        
        // Convert the character to Latin1 encoding
        $latin1Char = @iconv('UTF-8', 'ISO-8859-1//IGNORE', $char);
        
        // If conversion is successful, append to output string
        if ($latin1Char !== false && strlen($latin1Char) > 0) {
            $output .= $latin1Char;
        } else {
            // If conversion fails, add to unsupported characters list
            $unsupportedChars[] = $char;
        }
    }
    
    return [
        'sanitized' => $output,
        'unsupported' => $unsupportedChars
    ];
}

// Example usage
$input = "This is a test string with unsupported chars: 😀, 你好, привет";
$result = sanitizeLatin1($input);

echo "Sanitized string: " . $result['sanitized'] . "\n";
echo "Unsupported characters: " . implode(', ', $result['unsupported']) . "\n";