Uploaded image for project: 'Camunda Optimize'
  1. Camunda Optimize
  2. OPT-2266

Extendedbounds are not set on default limiting startDate histogram filter

    Details

    • Type: Bug Report
    • Status: Done
    • Priority: L3 - Default
    • Resolution: Won't Fix
    • Affects Version/s: None
    • Fix Version/s: 2.5.0-alpha2
    • Component/s: backend
    • Labels:
      None

      Description

      given:

      • I have a group by startDate report and 1000 process instances distributed over more than 1000 days

      when:

      • I group by days

      then:

      • I get less than 1000 day buckets in the result

      expected:

      • I get exactly 1000 day buckets in the result

      Context:
      When there is no user defined startDateFilter we create a default filter windows from the latest entry - 1000 (groupByUnit). However we do not apply the extendedBounds to ensure potentially leading empty buckets are present in the result.

        Activity

        Hide
        michael.wagner Michael Wagner added a comment -

        After further discussion about the issue we came to the conclusion that the current behavior probably makes more sense than the alternative.

        In the original implementation we decided to apply the extended bounds only for filters set by the user, but not on default as this would mean a lot of empty buckets, also for small data.
        The problem with the described expected behavior is that for smaller amount of process instances this would also mean that 1000 day buckets are returned if the extended bounds are also applied for the default filter.
        e.g. when we have 2 process instances, one being inside the 1000 days, one being outside, that would mean 1 process instance result is returned, plus 999 empty buckets.
        Currently only 1 bucket with the 1 result is returned in that case.

        Show
        michael.wagner Michael Wagner added a comment - After further discussion about the issue we came to the conclusion that the current behavior probably makes more sense than the alternative. In the original implementation we decided to apply the extended bounds only for filters set by the user, but not on default as this would mean a lot of empty buckets, also for small data. The problem with the described expected behavior is that for smaller amount of process instances this would also mean that 1000 day buckets are returned if the extended bounds are also applied for the default filter. e.g. when we have 2 process instances, one being inside the 1000 days, one being outside, that would mean 1 process instance result is returned, plus 999 empty buckets. Currently only 1 bucket with the 1 result is returned in that case.

          People

          • Assignee:
            Unassigned
            Reporter:
            sebastian.bathke Sebastian Bathke
          • Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved: