Trouble Matching nested objects with multi_match Query Using Wildcards in Elasticsearch

[Morzaram (Chris King)]

Hey everyone,

I'm encountering an issue where my documents aren't being matched by a multi_match query when I include a wildcard with the search string. I'm trying to search across multiple fields, including nested names and email addresses, with a case-insensitive approach. Despite using a custom analyzer that applies the standard tokenizer and lowercase filter, my searches with the string "Nicole" (hoping to match "Nicole L" and "nicole@***.org" in my document) are not returning the expected results.

---document

{"_index"=>"appointments",
 "_id"=>"22676738",
 "_version"=>1,
 "_seq_no"=>4185,
 "_primary_term"=>1,
 "found"=>true,
 "_source"=>
  {"starts_at"=>"2024-03-18T02:00:00.000-07:00",
   "ends_at"=>"2024-03-18T04:00:00.000-07:00",
   "status"=>"Active",
   "planned_duration"=>7200,
   "users"=>[],
   "contact"=>{"id"=>222, "full_name"=>"Nicole L***", "emails"=>["nicole@***.org"], "phone_numbers"=>["+****"]},
   "address"=>{"line1"=>"***", "city"=>"**", "state"=>"**", "postal_code"=>"**"},
   "account_id"=>2}}

---query

client.explain(index: 'appointments', id: 22676738, body:  {
        query: {
      bool: {
        should: [
          {
            bool: {
              must: {
                multi_match: {
                query: "Nicole*",
                fields: ["contact.full_name^10", "contact.emails^10", "contact.phone_numbers", "users.name", "status", "address.line1", "address.line2", "address.city", "address.state", "address.postal_code"]
              }
            },
            filter: {
              term: {
                account_id: 2
              }
            }
          }
          }
        ]
      }
    }

---explaination

{"_index"=>"appointments",
 "_id"=>"22676738",
 "matched"=>false,
 "explanation"=>
  {"value"=>0.0,
   "description"=>"Failure to meet condition(s) of required/prohibited clause(s)",
   "details"=>
    [{"value"=>0.0,
      "description"=>
       "no match on required clause (((contact.full_name:nicole)^10.0 | address.line2:nicole | address.postal_code:nicole | address.line1:nicole | (contact.emails:nicole)^10.0 | users.name:nicole | address.city:nicole | address.state:nicole | contact.phone_numbers:nicole | status:nicole))",
      "details"=>[{"value"=>0.0, "description"=>"No matching clause", "details"=>[]}]},
     {"value"=>0.0,
      "description"=>"match on required clause, product of:",
      "details"=>[{"value"=>0.0, "description"=>"# clause", "details"=>[]}, {"value"=>1.0, "description"=>"account_id:[2 TO 2]", "details"=>[]}]}]}}

--- index mapping

INDEX_MAPPING = {
        settings: { 
          analysis: {
            analyzer: {
              default: {
                type: "custom",
                tokenizer: "standard",
                filter: ["lowercase"]
              }
            }
          }
        },
        mappings: {
          properties: {
            starts_at: {
              type: "date"
            },
            ends_at: {
              type: "date"
            },
            contact_name: {
              type: "text"
            },
            status: {
              type: "text"
            },
            planned_duration: {
              type: "long"
            },
            users: {
              type: "nested",
              properties: {
                id: {type: "long"},
                initials: {type: "text"},
                color: {type: "text"},
                name: {type: "text"}
              }},
            contact: {
              type: "nested",
              properties: {
                id: {type: "long"},
                full_name: {type: "text"},
                emails: {type: "text"},
                phone_numbers: {type: "text"}
            }},
            address: {
              type: "nested",
              properties: {
              line1: {type: "text"},
              line2: {type: "text"},
              city: {type: "text"},
              state: {type: "text"},
              postal_code: {type: "text"}}
            },
            account_id: {type: "long"},
          }
        }
      }

Sorry if this is blatantly obvious. I'm struggling with some of the concepts I guess that lead me to me not finding this on my own.

Thank you in advance

[Morzaram (Chris King)]

Alright this is the thing that I got to work

{
          size: 10,
          query: {
            bool: {
              should: [
                {
                  bool: {
                    must: {
                      nested: {
                        path: "contact,
                        query: {
                          multi_match: {
                            query: "Nicole*",
                            fields: ["contact.full_name^10", "contact.emails^10", "contact.phone_numbers"]
                          }
                        }
                      }
                    },
                    filter: {
                      term: {
                        account_id: 2
                      }
                    }
                  }
                }
              ]
            }
          }
        }

And I would do that for each nested object. However I'm not sure if there's a better way to do this so I can include multiple nested objects. ie contact, address