Java正则表达式在数据清洗中的高级应用技巧

import java.util.regex.Matcher;
import java.util.regex.Pattern;

public class DataExtractor {
    public static void main(String[] args) {
        String[] orderDescriptions = {
            "订单ID: ABC12345, 金额 $123.45, 日期 2023-10-26",
            "2023/11/01 订单号 XYZ98765, 支付了 99.99元",
            "ID: PQR67890, ￥50.00 (2023.12.05)",
            "订单: DEF00000, $200.00" // 缺少日期
        };

        // 尝试匹配订单ID、金额和日期。注意日期和金额的多种可能格式
        // 我会尽量让模式更灵活，例如金额符号、日期分隔符
        String regex = "(?:订单ID|订单号|ID|订单)[:\s]*?(?<orderId>[A-Z0-9]+)" +
                       "(?:.*?(?:金额|支付了)?\s*?[¥$](?<amount>\d+\.?\d*))?" +
                       "(?:.*?(?:日期|\()\s*?(?<date>\d{4}[-/.]\d{2}[-/.]\d{2}))?";

        Pattern pattern = Pattern.compile(regex, Pattern.CASE_INSENSITIVE);

        for (String desc : orderDescriptions) {
            Matcher matcher = pattern.matcher(desc);
            if (matcher.find()) {
                String orderId = matcher.group("orderId");
                String amount = matcher.group("amount");
                String date = matcher.group("date");

                System.out.println("描述: "" + desc + """);
                System.out.println("  订单ID: " + (orderId != null ? orderId : "N/A"));
                System.out.println("  金额: " + (amount != null ? amount : "N/A"));
                System.out.println("  日期: " + (date != null ? date : "N/A"));
                System.out.println("---");
            } else {
                System.out.println("描述: "" + desc + "" - 未匹配到有效信息");
                System.out.println("---");
            }
        }
    }
}

import java.util.regex.Pattern;

public class DataStandardizer {
    public static void main(String[] args) {
        // 电话号码标准化：统一为 123-456-7890 格式
        String[] phoneNumbers = {
            "(123) 456-7890",
            "123.456.7890",
            "1234567890",
            "+1-123-456-7890",
            "电话: 123 456 7890" // 包含非数字字符
        };
        // 匹配任意非数字字符，并替换为空
        Pattern nonDigitPattern = Pattern.compile("[^\d]+");
        // 匹配10位数字，并用捕获组重新格式化
        Pattern phoneFormatPattern = Pattern.compile("(\d{3})(\d{3})(\d{4})");

        System.out.println("--- 电话号码标准化 ---");
        for (String phone : phoneNumbers) {
            String cleanedPhone = nonDigitPattern.matcher(phone).replaceAll("");
            String standardizedPhone = phoneFormatPattern.matcher(cleanedPhone).replaceAll("$1-$2-$3");
            System.out.println("原始: "" + phone + "" -> 标准化: "" + standardizedPhone + """);
        }

        // 日期格式标准化：统一为 YYYY-MM-DD
        String[] dates = {
            "2023/10/26",
            "26-10-2023",
            "2023.10.26",
            "October 26, 2023" // 这个稍微复杂，需要映射月份
        };
        // 匹配 YYYY/MM/DD, YYYY.MM.DD
        Pattern datePattern1 = Pattern.compile("(\d{4})[./](\d{2})[./](\d{2})");
        // 匹配 DD-MM-YYYY
        Pattern datePattern2 = Pattern.compile("(\d{2})-(\d{2})-(\d{4})");

        System.out.println("
--- 日期格式标准化 (部分) ---");
        for (String date : dates) {
            String result = date;
            if (datePattern1.matcher(result).matches()) {
                result = datePattern1.matcher(result).replaceAll("$1-$2-$3");
            } else if (datePattern2.matcher(result).matches()) {
                // 调换捕获组顺序
                result = datePattern2.matcher(result).replaceAll("$3-$2-$1");
            }
            // 对于 "October 26, 2023" 这种，需要更复杂的逻辑，可能要结合Map进行月份转换，单纯正则替换有局限性
            System.out.println("原始: "" + date + "" -> 标准化: "" + result + """);
        }
    }
}