如何用Java 8按表头读取CSV文件中的两列数据

问题 #

需要通过表头col5和col7读取CSV文件中的两列数据，现有代码仅能读取单列（col5），希望将这两列数据存入同一个Map中（选择Map是因为后续需要对数据进行处理）。

现有代码 #

public static List<String> getData() throws IOException {
        String titleToSearchFor = "col5";
        Path path = Paths.get("D:/blackduck_sample.txt");

        if (Files.exists(path)) {
            List<String> lines = Files.readAllLines(path);

            List<String> columns = Arrays.asList(lines.get(0).split(","));

            int titleIndex = columns.indexOf(titleToSearchFor);

            List<String> values = lines.stream().skip(1).map(line -> Arrays.asList(line.split(",")))
                    .map(list -> list.get(titleIndex)).filter(Objects::nonNull).filter(s -> s.trim().length() > 0)
                    .collect(Collectors.toList());

            return values;
        }

        return new ArrayList<>();

    }

样本CSV文件 #

col1,col2,col3,col4,col5,col6,col7,col8
CVM Finding,12345,sdc,CAP,12345:sdc-posistion-svc:SnakeYAML:2.1:3,TRUE,FALSE,OverDue
CVM Finding,12345,sdc,CAP,12345:sdc-event-generation-svc:SpringBoot:2.1.2.Release:1,TRUE,FALSE,Overdue
CVM Finding,12345,sdc,CAP,12345:sdc-event-subscription-svc:Apache Tomcat:9.0.70:1,TRUE,FALSE,Due in <= 90
CVM Finding,12345,sdc,CAP,12345:sdc-bh-registry-svc:Jettison - Json Stax implementation:1.5.3:1,TRUE,FALSE,Due in <=90
CVM Finding,12345,sdc,CAP,12345:sdc-event-generation-svc:Apache Tomcat:9.0.70:2,TRUE,FALSE,Due in >90
CVM Finding,12345,sdc,CAP,12345:sdc-posistion-svc:jackson-databind:2.14:1,TRUE,FALSE,OverDue
CVM Finding,12345,sdc,CAP,12345:sdc-event-subscription-svc:jackson-databind:2.14:1,TRUE,FALSE,Due in <= 90
CVM Finding,12345,sdc,CAP,12345:sdc-event-subscription-svc:SpringBoot:2.1.2.Release:1,TRUE,FALSE,Overdue
CVM Finding,12345,sdc,CAP,12345:sdc-posistion-svc:SpringBoot:2.1.2.Release:1,TRUE,FALSE,OverDue

之前找到的参考方法（仅按索引读取） #

String[] cols = line.split(cvsSplitBy);
System.out.println("Column 5= " + cols[5] + " , Column 7=" + cols[7]);

期望输出 #

12345:sdc-posistion-svc:SnakeYAML:2.1:3,OverDue
12345:sdc-event-generation-svc:SpringBoot:2.1.2.Release:1,Overdue
12345:sdc-event-subscription-svc:Apache Tomcat:9.0.70:1,Due in <= 90
12345:sdc-bh-registry-svc:Jettison - Json Stax implementation:1.5.3:1,Due in <=90
12345:sdc-event-generation-svc:Apache Tomcat:9.0.70:2,Due in >90
12345:sdc-posistion-svc:jackson-databind:2.14:1,OverDue
12345:sdc-event-subscription-svc:jackson-databind:2.14:1,Due in <= 90
12345:sdc-event-subscription-svc:SpringBoot:2.1.2.Release:1,Overdue
12345:sdc-posistion-svc:SpringBoot:2.1.2.Release:1,OverDue

解决方案 #

完全可以将两列数据存入同一个Map中，以下提供两种常见实现方式：

方式1：将col5作为Key，目标列作为Value存入Map #

（注：你的期望输出第二列对应样本CSV的col8，若需匹配期望输出，只需将代码中的col7Title改为col8即可）

import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Paths;
import java.util.Map;
import java.util.Objects;
import java.util.stream.Collectors;

public class CSVReader {
    public static Map<String, String> getDataAsMap() throws IOException {
        String col5Title = "col5";
        String targetColTitle = "col7"; // 若匹配期望输出，改为"col8"
        Path path = Paths.get("D:/blackduck_sample.txt");

        if (Files.exists(path)) {
            var lines = Files.readAllLines(path);
            if (lines.isEmpty()) return Map.of();

            // 获取表头索引
            var columns = java.util.Arrays.asList(lines.get(0).split(","));
            int col5Index = columns.indexOf(col5Title);
            int targetColIndex = columns.indexOf(targetColTitle);

            // 校验表头是否存在
            if (col5Index == -1 || targetColIndex == -1) {
                throw new IllegalArgumentException("指定表头不存在：" + col5Title + " 或 " + targetColTitle);
            }

            // 读取数据并封装为Map
            return lines.stream()
                    .skip(1)
                    .map(line -> line.split(","))
                    .filter(parts -> parts.length > Math.max(col5Index, targetColIndex))
                    .map(parts -> {
                        String col5Val = parts[col5Index].trim();
                        String targetVal = parts[targetColIndex].trim();
                        return Map.entry(col5Val, targetVal);
                    })
                    .filter(entry -> !entry.getKey().isEmpty() && !entry.getValue().isEmpty())
                    .collect(Collectors.toMap(
                            Map.Entry::getKey,
                            Map.Entry::getValue,
                            (existing, replacement) -> existing // 重复Key保留第一个值，可按需修改
                    ));
        }
        return Map.of();
    }

    // 生成符合期望的字符串格式输出
    public static java.util.List<String> getFormattedOutput() throws IOException {
        var dataMap = getDataAsMap();
        return dataMap.entrySet().stream()
                .map(entry -> entry.getKey() + "," + entry.getValue())
                .collect(Collectors.toList());
    }

    public static void main(String[] args) throws IOException {
        getFormattedOutput().forEach(System.out::println);
    }
}

方式2：每行数据封装为Map（键为表头名，值为对应列数据） #

适合需要保留表头与值对应关系的场景：

public static java.util.List<Map<String, String>> getRowData() throws IOException {
    String col5Title = "col5";
    String targetColTitle = "col7"; // 若匹配期望输出，改为"col8"
    Path path = Paths.get("D:/blackduck_sample.txt");

    if (Files.exists(path)) {
        var lines = Files.readAllLines(path);
        if (lines.isEmpty()) return new java.util.ArrayList<>();

        var columns = java.util.Arrays.asList(lines.get(0).split(","));
        int col5Index = columns.indexOf(col5Title);
        int targetColIndex = columns.indexOf(targetColTitle);

        if (col5Index == -1 || targetColIndex == -1) {
            throw new IllegalArgumentException("指定表头不存在：" + col5Title + " 或 " + targetColTitle);
        }

        return lines.stream()
                .skip(1)
                .map(line -> line.split(","))
                .filter(parts -> parts.length > Math.max(col5Index, targetColIndex))
                .map(parts -> {
                    String col5Val = parts[col5Index].trim();
                    String targetVal = parts[targetColIndex].trim();
                    return Map.of(col5Title, col5Val, targetColTitle, targetVal);
                })
                .filter(map -> !map.get(col5Title).isEmpty() && !map.get(targetColTitle).isEmpty())
                .collect(Collectors.toList());
    }
    return new java.util.ArrayList<>();
}

代码说明 #

• 避免硬编码索引，通过表头名动态获取列位置，适配表头变动场景
• 增加数据校验，防止数组越界和空值干扰
• 提供两种Map存储方式，可根据后续处理需求选择

内容的提问来源于stack exchange，提问作者Selenium New bie