Elasticsearch Dataset

Aggregation Override

CBoard provides three types of common Bucket aggregation override assist inputs. For specific configurable parameters, please refer to the official documentation: Bucket Aggregations

1. date_hist: Date histogram aggregation
2. number_range: Number range aggregation
3. number_hist: Number histogram

Syntax (multiple column aggregations can be overridden repeatedly):
{
  "columnname":{
    "<aggregation_type>" : {
      <aggregation_body>
    }
  }
}
---------------------------------
Example:
Aggregate the timestamp field (long type) in 10-minute intervals, and aggregate the numeric memory field with a custom range.
{
  "timestamp": {
    "date_histogram": {
      "field": "timestamp",
      "format": "yyyy-MM-dd HH:mm:ss",
      "interval": "10m",
      "time_zone": "+08:00"
    }
  },
  "memory": {
    "range": {
      "field": "memory",
      "ranges": [
        {
          "to": 10000
        },
        {
          "from": 10000,
          "to": 30000
        },
        {
          "from": 30000
        }
      ]
    }
  }
}

Date Field + Interval

Used for generating index date suffixes and query filtering based on the index. This feature is for maintenance and may not be very stable.

When a date field is set, the default index naming follows the pattern: indexName-yyyy.MM.dd.

• When fetching fields, a wildcard path is used: http://localhost:9200/index_name-*/_mapping
• For query searches, if the date field does not have a filter condition, the default date component is used: http://localhost:9200/index_name-2023.01.01,index_name-2023.01.02,index_name-2023.01.03/_mapping
• When the date field has a filter condition, the date suffix is automatically generated based on the filter condition.

TimeZone

Index date timezone configuration. Supports configurations like +02:00 or timezone names like Asia/Shanghai.

Global Filter Conditions

You can set one or multiple global filter conditions. Built-in date variables can be used within filter conditions to achieve dynamic global filtering.

{
  "range": {
    "time": {
      "gt": "${cdt.addMonth(-1, 'yyyy-MM-dd')}"
    }
  }
}

[
  {
    "range": {
      "time": {
        "gt": "xx"
      }
    }
  },
  {
    "term": {
      "status": "published"
    }
  }
]

Script Derived Dimension Configuration

The aggregation override settings mentioned earlier are applied at the dataset query level, equivalent to a global query column aggregation override. If you want to apply multiple bucketing strategies to a single column, you can reference the optional column multiple times in the Schema tree and then edit the custom information.

{
  "esBucket": {
    "<aggregation_type>" : {
      <aggregation_body>
    }
  }
}

Advanced Aggregation Expressions

Besides conventional aggregate statistics, ES aggregation expressions also support Filter Aggregation, used for conditional statistics or static proportions.

Syntax:
count/sum/avg/max/min("{
  'column': 'Column used for aggregation',
  'filter': <filter_body>
}")
-------------------------------
count("{
    'column': 'orderId',
    'filter': {
        'term': {
          'order.platform': 'online'
        }
    }
}")