Java导出CSV文件

---

以前导出总是用POI导出为Excel文件，后来当我了解到CSV以后，我发现速度飞快。

如果导出的数据不要求格式、样式、公式等等，建议最好导成CSV文件，因为真的很快。

虽然我们可以用Java再带的文件相关的类去操作以生成一个CSV文件，但事实上有好多第三方类库也提供了类似的功能。

这里我们使用apache提供的commons-csv组件

Commons CSV

文档在这里

http://commons.apache.org/

http://commons.apache.org/proper/commons-csv/

http://commons.apache.org/proper/commons-csv/user-guide.html

先看一下具体用法

@Test
    public void testWrite() throws Exception {
        FileOutputStream fos = new FileOutputStream("E:/cjsworkspace/cjs-excel-demo/target/abc.csv");
        OutputStreamWriter osw = new OutputStreamWriter(fos, "GBK");

        CSVFormat csvFormat = CSVFormat.DEFAULT.withHeader("姓名", "年龄", "家乡");
        CSVPrinter csvPrinter = new CSVPrinter(osw, csvFormat);

//        csvPrinter = CSVFormat.DEFAULT.withHeader("姓名", "年龄", "家乡").print(osw);

        for (int i = 0; i < 10; i++) {
            csvPrinter.printRecord("张三", 20, "湖北");
        }

        csvPrinter.flush();
        csvPrinter.close();

    }

    @Test
    public void testRead() throws IOException {
        InputStream is = new FileInputStream("E:/cjsworkspace/cjs-excel-demo/target/abc.csv");
        InputStreamReader isr = new InputStreamReader(is, "GBK");
        Reader reader = new BufferedReader(isr);

        CSVParser parser = CSVFormat.EXCEL.withHeader("name", "age", "jia").parse(reader);
//        CSVParser csvParser = CSVParser.parse(reader, CSVFormat.DEFAULT.withHeader("name", "age", "jia"));
        List<CSVRecord> list = parser.getRecords();
        for (CSVRecord record : list) {
            System.out.println(record.getRecordNumber()
                    + ":" + record.get("name")
                    + ":" + record.get("age")
                    + ":" + record.get("jia"));
        }

        parser.close();
    }

    /**
     * Parsing an Excel CSV File
     */
    @Test
    public void testParse() throws Exception {
        Reader reader = new FileReader("C:/Users/Administrator/Desktop/abc.csv");
        CSVParser parser = CSVFormat.EXCEL.parse(reader);
        for (CSVRecord record : parser.getRecords()) {
            System.out.println(record);
        }
        parser.close();
    }

    /**
     * Defining a header manually
     */
    @Test
    public void testParseWithHeader() throws Exception {
        Reader reader = new FileReader("C:/Users/Administrator/Desktop/abc.csv");
        CSVParser parser = CSVFormat.EXCEL.withHeader("id", "name", "code").parse(reader);
        for (CSVRecord record : parser.getRecords()) {
            System.out.println(record.get("id") + ","
                    + record.get("name") + ","
                    + record.get("code"));
        }
        parser.close();
    }

    /**
     * Using an enum to define a header
     */
    enum MyHeaderEnum {
        ID, NAME, CODE;
    }

    @Test
    public void testParseWithEnum() throws Exception {
        Reader reader = new FileReader("C:/Users/Administrator/Desktop/abc.csv");
        CSVParser parser = CSVFormat.EXCEL.withHeader(MyHeaderEnum.class).parse(reader);
        for (CSVRecord record : parser.getRecords()) {
            System.out.println(record.get(MyHeaderEnum.ID) + ","
                    + record.get(MyHeaderEnum.NAME) + ","
                    + record.get(MyHeaderEnum.CODE));
        }
        parser.close();
    }

    private List<Map<String, String>> recordList = new ArrayList<>();

    @Before
    public void init() {
        for (int i = 0; i < 5; i++) {
            Map<String, String> map = new HashMap<>();
            map.put("name", "zhangsan");
            map.put("code", "001");
            recordList.add(map);
        }
    }

    @Test
    public void writeMuti() throws InterruptedException {
        ExecutorService executorService = Executors.newFixedThreadPool(3);
        CountDownLatch doneSignal = new CountDownLatch(2);

        executorService.submit(new exprotThread("E:/0.csv", recordList, doneSignal));
        executorService.submit(new exprotThread("E:/1.csv", recordList, doneSignal));

        doneSignal.await();
        System.out.println("Finish!!!");
    }

    class exprotThread implements Runnable {

        private String filename;
        private List<Map<String, String>> list;
        private CountDownLatch countDownLatch;

        public exprotThread(String filename, List<Map<String, String>> list, CountDownLatch countDownLatch) {
            this.filename = filename;
            this.list = list;
            this.countDownLatch = countDownLatch;
        }

        @Override
        public void run() {
            try {
                CSVPrinter printer = new CSVPrinter(new FileWriter(filename), CSVFormat.EXCEL.withHeader("NAME", "CODE"));
                for (Map<String, String> map : list) {
                    printer.printRecord(map.values());
                }
                printer.close();
                countDownLatch.countDown();
            } catch (IOException e) {
                e.printStackTrace();
            }
        }
    }

CSV与EXCEL

　　/**
     * 测试写100万数据需要花费多长时间
     */
    @Test
    public void testMillion() throws Exception {
        int times = 10000 * 10;
        Object[] cells = {"满100减15元", "100011", 15};

        //  导出为CSV文件
        long t1 = System.currentTimeMillis();
        FileWriter writer = new FileWriter("G:/test1.csv");
        CSVPrinter printer = CSVFormat.EXCEL.print(writer);
        for (int i = 0; i < times; i++) {
            printer.printRecord(cells);
        }
        printer.flush();
        printer.close();
        long t2 = System.currentTimeMillis();
        System.out.println("CSV: " + (t2 - t1));

        //  导出为Excel文件
        long t3 = System.currentTimeMillis();
        XSSFWorkbook workbook = new XSSFWorkbook();
        XSSFSheet sheet = workbook.createSheet();
        for (int i = 0; i < times; i++) {
            XSSFRow row = sheet.createRow(i);
            for (int j = 0; j < cells.length; j++) {
                XSSFCell cell = row.createCell(j);
                cell.setCellValue(String.valueOf(cells[j]));
            }
        }
        FileOutputStream fos = new FileOutputStream("G:/test2.xlsx");
        workbook.write(fos);
        fos.flush();
        fos.close();
        long t4 = System.currentTimeMillis();
        System.out.println("Excel: " + (t4 - t3));
    }

Maven依赖

<dependencies>
    <dependency>
        <groupId>org.apache.commons</groupId>
        <artifactId>commons-csv</artifactId>
        <version>1.5</version>
    </dependency>

    <dependency>
        <groupId>org.apache.poi</groupId>
        <artifactId>poi</artifactId>
        <version>3.17</version>
    </dependency>

    <dependency>
        <groupId>org.apache.poi</groupId>
        <artifactId>poi-ooxml</artifactId>
        <version>3.17</version>
    </dependency>

    <dependency>
        <groupId>junit</groupId>
        <artifactId>junit</artifactId>
        <version>4.12</version>
        <scope>test</scope>
    </dependency>
</dependencies>

最后，刚才的例子中只写了3个字段，100万行，生成的CSV文件有十几二十兆，太多的话建议分多个文件打包下周，不然想象一个打开一个几百兆的excel都费劲。