fast-csv转换似乎正在剥离列

Question

我正在处理一个项目，该项目在底层解析csv文件并对其进行转换，以便仅使用解析文件中的某些列将某些列映射到新文件。包括输入.csv以供参考。

Co Code,Batch ID,File #,Reg Hours,O/T Hours,Reg Earnings,O/T 
Earnings,Hours 3 Code,Hours 3 Amount,Hours 3 Code,Hours 3 Amount,Hours 4 
Code,Hours 4 Amount,Hours 4 Code,Hours 4 Amount,Earnings 3 Code,Earnings 3 
Amount,Earnings 3 Code,Earnings 3 Amount,Earnings 4 Code,Earnings 4 
Amount,Earnings 4 Code,Earnings 4 Amount,Earnings 5 Code,Earnings 5 
Amount,Earnings 5 Code,Earnings 5 Amount,Temp Dept

83M,1,102,70.00,14.00,,,,,,,,,,,,,,,,,,,,,,,

83M,1,109,92.50,16.50,,,D,5.00,,,,,,,,,,,,,,,,,,,

83M,1,61,70.00,11.50,,,,,,,V,24.00,,,,,,,,,,,,,,,

83M,1,6384,69.00,15.50,,,D,9.50,,,,,,,,,,,,,,,,,,,

83M,1,6568,78.50,29.50,,,D,18.00,,,,,,,,,,,,,,,,,,,

83M,1,6583,84.50,32.50,,,D,10.00,,,,,,,,,,,,,,,,,,,

83M,1,6596,70.50,27.50,,,D,11.50,,,,,,,,,,,,,,,,,,,

83M,1,6627,95.50,23.50,,,D,12.50,,,,,,,,,,,,,,,,,,,

83M,1,6628,83.50,31.50,,,D,12.50,,,,,,,,,,,,,,,,,,,

83M,1,93,45.00,21.50,,,,,,,11,16.00,V,16.00,,,,,,,,,,,,,,,

我发现，当我使用fast-csv提供的transform函数时，有一列特别的'Hours 3 Amount‘映射到'D/T hours’，它甚至在返回转换后的对象(参见对象输出示例和csv输出示例)之前也没有返回任何值(不包括某些值)。

{ 'Co Code': '83M',
  'Batch ID': '1',
  'File #': '93',
  'Reg Hours': '36.50',
  'O/T Hours': '10.00',
  'Reg Earnings': '',
  'O/T Earnings': '',
  'Hours 3 Code': '',
  'Hours 3 Amount': '',
  'Hours 4 Code': 'H',
  'Hours 4 Amount': '8.00',
  'Earnings 3 Code': '',
  'Earnings 3 Amount': '',
  'Earnings 4 Code': '',
  'Earnings 4 Amount': '',
  'Earnings 5 Code': '',
  'Earnings 5 Amount': '',
  'Temp Dept': '' }

Co Code,Batch Id,File #,Reg Hours,O/T Hours,D/T Hours,Reg Earnings,OT 
Earnings,DT Earnigs,Pay Period Start,Pay Period End,Total
83M,1,6596,30.00,,,,,,12/30/18,1/12/19,
83M,1,6628,66.50,19.00,,,,,12/30/18,1/12/19,
83M,1,6650,20.50,18.50,,,,,12/30/18,1/12/19,

有趣的是，我可以输入一个字符串文字作为'D/T小时‘的值，它会出现在输出的.csv文件中。仔细检查后，我发现如果我注释掉转换函数，输出的csv，虽然是原始的未转换形式，但填充了’小时3金额‘的值(见下文)。我还包含了我的代码。

// Process the file by moving the cells and inserting the start and end date
const processFile = (fileName, file) => {
  csv
    .fromPath(basePath + file, {
      ignoreEmpty: false,
      headers: true,
      discardUnmappedColumns: true,
      oobjectMode: true
    })
    .transform(obj => {
      [startDate, endDate] = fileName;

      console.log(obj); // Does not include the Hours 3 Amount property

      if (obj && obj['Co Code'] != '') {
        return {
          'Co Code': obj['Co Code'],
          'Batch Id': obj['Batch ID'],
          'File #': obj['File #'],
          'Reg Hours': obj['Reg Hours'],
          'O/T Hours': obj['O/T Hours'],
          'D/T Hours': obj['Hours 3 Amount'],
          'Reg Earnings': '',
          'OT Earnings': '',
          'DT Earnigs': '',
          'Pay Period Start': startDate,
          'Pay Period End': endDate,
          Total: ''
        };
      }
    })
    .on('error', () => {
      return false;
    })
    .pipe(csv.createWriteStream())
    .pipe(
      fs.createWriteStream('./working-files/Payroll-files/output/out.csv', {
        encoding: 'utf8',
        flags: 'a',
        headers: true
      })
    );
};

有没有人知道为什么转换函数会在返回转换后的对象之前剥离列？

Answer 1

不要紧。我确定我的源.csv有一个重复的hours 3 Amount列。如果存在重复的值，transform函数显然不会返回值。