如何在匯總另一個值的同時按特定項目/列的值對基於數組或列的數據進行分組？

Question

我們正在 SAAS 產品中構建集成，該產品從表中導出數據並將其寫入 CSV 文件。 該工具允許編寫 JavaScript（我不太熟悉）代碼，該代碼可以從表中讀取數據並將其寫入 CSV 文件並將文件傳輸到 S3 存儲桶。

直到現在我才能開始工作。 示例 JavaScript -

// Manually invoke this method via 'Test connection'
function testConnection(context) {
  return true;
}
// Manually invoke this method via 'Run manually'
function exportData(context) {

  // Write the message to the log.
  ai.log.logInfo("Script Execution Starting...");
  var dataSource = context.getDataSource();

  var bucketName = dataSource.getSetting("Bucket Name").getValue();
  var awsAccessKey = dataSource.getSetting("Access Key").getValue();
  var awsSecretKey = dataSource.getSetting("Access Secret Key").getValue();
  var region = dataSource.getSetting("Region").getValue();
  var date = new Date();
  var currentDate = date.toLocaleDateString().replaceAll("/","");
  //Append unique number and current date to the file name.
  var key = dataSource.getSetting("Output File Name").getValue() + "_" + Date.now() + "_" + currentDate + ".csv";
  ai.log.logInfo("Preparing Output File : " + key);

  // Step 1: Build an array with the data that needs to be written to the csv file.
  var reader = context.createTableReader(['SqlProjectCodeOnly', 'ResCodeCategory', 'Period', 'SqlAWSAccountID', 'SqlAWSAccountName', 'SqlAWSAccountRegion', 'Value']);
  // Step 2: Write the header of the output file.
  var result = "Project|Product Name|Report Month|AWS Account ID|AWS Account Name|AWS Region|Adjusted Cost\n";
  var row = null;
  // If reader.readRow() returns null, then there are no more records.
  while ((row = reader.readRow()) !== null) {
    //Separate each field by pipe.
    result += row.join("|") + '\n';
  }
  ai.awss3.putFile(bucketName, key, result, region, awsAccessKey, awsSecretKey);
  ai.log.logInfo("Script Execution Completed.");
}

它以以下格式輸出數據 -

Project|Product Name|Report Month|AWS Account ID|AWS Account Name|AWS Region|Adjusted Cost

XXXXXXX|AWS Elastic Compute Cloud|01/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|2588.598833

XXXXXXX|AWS Elastic Compute Cloud|01/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|1553.1593004

XXXXXXX|AWS Elastic Compute Cloud|02/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|4141.7581328

XXXXXXX|AWS Elastic Compute Cloud|02/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|1553.1593004

現在它需要按前六列對數據進行分組，並對最后一列的值求和，以獲得以下格式的數據 -

Project|Product Name|Report Month|AWS Account ID|AWS Account Name|AWS Region|Adjusted Cost
XXXXXXX|AWS Elastic Compute Cloud|01/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|4,141.7581334
XXXXXXX|AWS Elastic Compute Cloud|02/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|5,694.9174338

我嘗試使用 reduce 功能，但無法使其正常工作。 當我在下面寫日志時-

ai.log.logVerbose( " row: " + row);

它寫入如下數據 -

row: XXXXXXX|AWS Elastic Compute Cloud|01/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|2588.598833

row: XXXXXXX|AWS Elastic Compute Cloud|01/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|1553.1593004

row: XXXXXXX|AWS Elastic Compute Cloud|02/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|4141.7581328

row: XXXXXXX|AWS Elastic Compute Cloud|02/2022|AWS Account ID (Uncategorized)|AWS Account Name (Uncategorized)|AWS Region (Uncategorized)|1553.1593004.

如果那里的專家可以提供幫助，我將不勝感激。

Answer 1

在讀取每一row ，必須將其push入例如rows數組中。 然后可以將此reduce歸約為一個對象，該對象包含按同一報告月份分組的聚合列組。 然后可以將此類groups對象的values map並join最終結果。

 function collectAndAggregateSameReportMonthAdjustedCost(groups, recordColumns) { const reportMonth = recordColumns[2]; const adjustedCost = parseFloat(recordColumns[6]); let groupedColumns = groups[reportMonth]; if (!groupedColumns) { groupedColumns = groups[reportMonth] = [...recordColumns]; groupedColumns[6] = adjustedCost; } else { groupedColumns[6] = groupedColumns[6] + adjustedCost; } return groups; } // const rows = []; let row; // // If reader.readRow() returns null, then there are no more records. // while ((row = reader.readRow()) !== null) { // rows.push(row); // } // `rows` after being aggregated by the above commented code // will loke like the next provided demonstrator ... const rows = [[ 'XXXXXXX', 'AWS Elastic Compute Cloud', '01/2022', 'AWS Account ID (Uncategorized)', 'AWS Account Name (Uncategorized)', 'AWS Region (Uncategorized)', '2588.598833', ], [ 'XXXXXXX', 'AWS Elastic Compute Cloud', '01/2022', 'AWS Account ID (Uncategorized)', 'AWS Account Name (Uncategorized)', 'AWS Region (Uncategorized)', '1553.1593004', ], [ 'XXXXXXX', 'AWS Elastic Compute Cloud', '02/2022', 'AWS Account ID (Uncategorized)', 'AWS Account Name (Uncategorized)', 'AWS Region (Uncategorized)', '4141.7581328', ], [ 'XXXXXXX', 'AWS Elastic Compute Cloud', '02/2022', 'AWS Account ID (Uncategorized)', 'AWS Account Name (Uncategorized)', 'AWS Region (Uncategorized)', '1553.1593004', ]]; const aggregatedRows = Object .values( rows .reduce(collectAndAggregateSameReportMonthAdjustedCost, {}) ); console.log({ aggregatedRows }); const result = [ 'Project|Product Name|Report Month|AWS Account ID|AWS Account Name|AWS Region|Adjusted Cost', ...aggregatedRows .map(columns => { columns[6] = columns[6] .toLocaleString('en', { useGrouping: 'always', maximumFractionDigits: 7, }); return columns.join('|'); }), ].join('\n'); console.log(result);

 .as-console-wrapper { min-height: 100%!important; top: 0; }