Search fields in a collection via Ref

MongoDB Atlas Atlas Search
, , ,
1

Hi, I’m using 2 references in the links model:

toAsset: {
  type: mongoose.Types.ObjectId,
  ref: 'Assets',
  index: 'links'
}
fromAsset: {
  type: mongoose.Types.ObjectId,
  ref: 'Assets',
  index: 'links'
},
comment: {
  type: String,
  index: 'links'
}

Here, toAsset and fromAsset both reference a single object each, and the intention is to query 2 fields in the assets model:

title: {
  type: String,
  required: true,
  index: 'assets'
},
note: {
  type: String,
  required: true,
  index: 'assets'
}

I’m using the following in Node:

const search = {
  ...(args.phraseToSearchLinks.length > 0 && {
    $search: {
      index: 'links',
      // compound: { 
      //   must: [{
      //     phrase: {
      //       query: args.phraseToSearchLinks,
      //       path: 'comment',
      //       slop: 2,
      //       score: { boost: { value: 3 } }
      //     }
      //   }]
      // },
      embeddedDocument: {
        path: directionOfLink,
        operator: {
          compound: {
            should: [{
              phrase: {
                query: args.phraseToSearchLinks,
                path: `${directionOfLink}.title`,
                slop: 2,
                score: { boost: { value: 2 } }
              }
            },
            {
              phrase: {
                query: args.phraseToSearchLinks,
                path: `${directionOfLink}.note`,
                slop: 2
              }
            }]
          }
        }
      }
    }
  })
}

Here, querying the comment field returns search results, but I don’t know how to combine it with the embeddedDocument, which isn’t returning anything at the moment.

I’ve made changes to the index for links to include the referenced fields, but I’m not seeing results when I perform a search on anything in a title or note field:

{
  "analyzer": "lucene.standard",
  "searchAnalyzer": "lucene.standard",
  "mappings": {
    "dynamic": false,
    "fields": {
      "comment": {
        "analyzer": "htmlStrippingAnalyzer",
        "type": "string"
      },
      "createdAt": {
        "type": "date"
      },
      "fromAsset": {
        "fields": {
          "note": {
            "analyzer": "htmlStrippingAnalyzer",
            "searchAnalyzer": "htmlStrippingAnalyzer",
            "type": "string"
          },
          "title": {
            "analyzer": "lucene.standard",
            "multi": {
              "keywordAnalyzer": {
                "analyzer": "ngramShingler",
                "searchAnalyzer": "ngramShingler",
                "type": "string"
              }
            },
            "searchAnalyzer": "lucene.standard",
            "type": "string"
          }
        },
        "type": "embeddedDocuments"
      },
      "isActive": {
        "type": "boolean"
      },
      "toAsset": {
        "fields": {
          "note": {
            "analyzer": "htmlStrippingAnalyzer",
            "searchAnalyzer": "htmlStrippingAnalyzer",
            "type": "string"
          },
          "title": {
            "analyzer": "lucene.standard",
            "multi": {
              "keywordAnalyzer": {
                "analyzer": "ngramShingler",
                "searchAnalyzer": "ngramShingler",
                "type": "string"
              }
            },
            "searchAnalyzer": "lucene.standard",
            "type": "string"
          }
        },
        "type": "embeddedDocuments"
      },
      "updatedAt": {
        "type": "date"
      }
    }
  },
  "analyzers": [
    {
      "name": "ngramShingler",
      "tokenFilters": [
        {
          "maxShingleSize": 3,
          "minShingleSize": 2,
          "type": "shingle"
        }
      ],
      "tokenizer": {
        "maxGram": 5,
        "minGram": 2,
        "type": "nGram"
      }
    },
    {
      "charFilters": [
        {
          "ignoredTags": [
            "a",
            "div",
            "p",
            "strong",
            "em",
            "img",
            "figure",
            "figcaption",
            "ol",
            "ul",
            "li",
            "span"
          ],
          "type": "htmlStrip"
        }
      ],
      "name": "htmlStrippingAnalyzer",
      "tokenFilters": [],
      "tokenizer": {
        "type": "standard"
      }
    }
  ]
}

At this stage, I’m not certain if what I’m attempting is possible! Any advice would be excellent.

2

Looks like I’ll have to wait until some time in March for version 6, which is when I’ll have a chance to run $search inside $lookup.