Use extractField to extract document id field about minisearch HOT 6 CLOSED

Hi @mquinn-leverege , this issue is resolved by this commit, now the ID field is consistent with the other fields, and its extraction can be customized with extractField.

This change is already released as part of version v2.5.0. This should solve your case, so I will close this issue, but feel free to comment if not.

from minisearch.

Hi @noraj ,
Yes, you can use extractField to implement a “virtual field”:

const miniSearch = new MiniSearch({
  fields: […],
  extractField: (document, fieldName) => {
    if (fieldName === 'id') {
      return `#{document.field1}-#{document.field2}`
    } else {
      return document[fieldName]
    }
  }
})

In the example above, the ID field is composed by joining two other fields.

from minisearch.

PS: It could be nice to create a expo gallery in the documentation or github wiki to link projects using minisearch so people could look at real-life/production examples.

from minisearch.

@mquinn-leverege thanks for reporting this. Yes, I agree that it would be nice to be able to tread the id field like the other fields via extractField. I'll look into this, it might impact a few places.

from minisearch.

@lucaong I have a project where I have no uniq field but I could have a uniq string by combining three fields.

Could I do something like

idField: ['field1', 'field2', 'field3']

or

function join(...strings) {
  return strings.join('')
}
...
idField: join('field1', 'field2', 'field3')

or create a virtual field with extractField and pass it to idField?

from minisearch.

Thanks it works well

const miniSearch = new MiniSearch({
  fields: […],
  extractField: (document, fieldName) => {
    if (fieldName === 'id') {
      return `${document.field1}-${document.field2}`
    } else {
      return document[fieldName]
    }
  }
})

from minisearch.