Search

Query Scheme

Query scheme for interaction between back- and front-end

Overall speaking, a query is represented as JSON object with a mandatory op field (means operation):

{
  "op": string,
  ...
}

Two types of queries: atomic and compound.

• atomic: a simple query on one field with one condition, e.g. date > 2015, title match "agile". This kind of query is defined by such JSON object:

    {
      "op": string,  // operation of query, condition, from pre-defined list
      "field": string,
      "value": any,
      ...      // operation may require more than one values
    }
  

  pre-defined operations:

    + `$eq`: equals
    + `$neq`: non equals
    + `$gt`: great than
    + `$ge`: great than or equal to
    + `$lt`: little than
    + `$le`: little than or equal to
    + `$match`: a string field matches given substring, i.e. a field contains the given substring. This can be extended to regex matching
  

  examples:

  • date > 2015:

      {
        "op": "$gt",
        "field": "date",
        "value": 2015,
      }
    

  • title match "agile":

      {
        "op": "$match",
        "field": "title",
        "value": "agile",
      }
    

• compound:

  a compound query represents logical combination (and, or, not) of queries. field is not necessary in compound queries. The scheme is

    {
      "op": "$and",  // or `"$or"`, same format
      "queries": []  // list of sub-queries
    }
  
    {
      "op": "$not",
      "query": object  // object of a query
    }
  

• An example:

    {
        “op”: “$and”,
        “queries”: [
            {
                “op”: “$match”,
                “field”: “title”,
                “value”: “tdd”
            },
            {
                “op”: “$and”,
                “queries”: [
                    {
                        “op”: “$ge”,
                        “field”: “date”,
                        “value”: 2009
                    },
                    {
                        “op”: “$le”,
                        “field”: “date”,
                        “value”: 2019
                    }
                ]
            }
        ]
    }