Mongodb – How to create incremental range aggregation in mongoose from a query parameter. i.e. create an incremental range from a field value

So was able to come up with a working solution using a mongo aggregate method called bucket, but one of the problem is that if you want an aggregation of like week 1 - week 20 in chunks of 5, i.e, 1-5 (1 included, 5 excluded), 5- 10,10-15 and 15-20, you will have to give it an array like; boundaries: [1,5,10,15,20] as part of the argument and from the question, i have to create a JS function to return an array of numbers between start week and end week with the range given also. Written in typescript, the return array from this question would look like : [2020-W01,2020-W05,2020-W10,2020-W15,2020-W20], Also there are certain edge cases you have to account for since all the parameters are dynamic, like if the week spans more than one year, also, the fact that mongo to the best of my knowledge don't have date format like "2020-W10" makes it a bit more complex

export function customLoop(
  startWeek: number,
  endWeek: number,
  rangeNum: number,
  year: number
): returnData {
  const boundaryRange: string[] = [];
  let skip = 0;
  for (let i = startWeek; i <= endWeek; i += rangeNum) {
    const currentNum: string = i < 10 ? `0${i}` : `${i}`;
    const result = `${year}-W${currentNum}`;
    boundaryRange.push(result);
    //if all the weeks in a year, Check where the last loop stops to determine skip
    if (endWeek === 53 && i + rangeNum > 53) {
      skip = i + rangeNum - 53 - 1;
    }
  }
  return {
    skip,
    theRange: boundaryRange,
  };
}

After this i opened my mongo compass on local to construct and chain aggregate methods and function to satisfy the task given:

  const result = await VaccinationModel.aggregate([
      {
        $match: {
          ReportingCountry: c,
        },
      },
      {
        $bucket: {
          groupBy: "$YearWeekISO",
          boundaries: [...boundaryRange],
          default: "others",
          output: {
            NumberDosesReceived: {
              $sum: "$NumberDosesReceived",
            },
          },
        },
      },
      {
        $addFields: {
          range: rangeNum,
        },
      },
      {
        $addFields: {
          weekStart: "$_id",
          weekEnd: {
            $function: {
              body: "function(id,range) {n      const arr = id.split('-')n      const year = arr[0];n      let week;n      let result=''n      if(arr[1]){n       week = arr[1].slice(1);n        result=`${year}-W${Number(week) + range}`nn              }nn               return resultn        }",
              args: ["$_id", "$range"],
              lang: "js",
            },
          },
        },
      },
      {
        $unset: ["_id", "range"],
      },
      {
        $match: {
          weekEnd: {
            $ne: "",
          },
        },
      },
      {
        $sort: {
          weekStart: 1,
        },
      },
    ]);

In that aggregation:

1. Match the country code.
2. I basically called the bucket aggregation with the array of boundaries, then summing the results of each chunk/range using its NumberDosesReceived field while naming it NumberDosesReceived.
3. since i needed extra two fields to complete the number of fields to return, namely weekStart and weekEnd that isnt in the dataset, the weekStart is the _id field from the bucket aggregation, to get the weekEnd, i added the range as a field.
4. If for instance the current mongo iteration is 2020-W5, which would be the in the _id, that means the weekend would be 5 + range = 10, so i used the mongo function method to extract that passing _id and range as argument.
5. Used the unset method to remove the _id and range field as it wouldn't be part of the return data.
6. Get this new weekEnd field excluding empty ones.
7. sort using it.

here is the link to the repo: link