PR #138 breaks nested type

[inkstak]

Hi,

The #138 breaks my app specs.
Why adding nested to TYPES_WITH_EMBEDDED_PROPERTIES ?

I use nested properties like this :

indexes :users, type: :nested do
  indexes :name_badge: 
end

to expect the following mapping :

"user": {
  "type": "nested",
  "properties": {
    "name": {
      "type": "string"
    }
  }
}

So now, properties becomes fields

According to https://www.elastic.co/guide/en/elasticsearch/reference/current/mapping-nested-type.html and https://www.elastic.co/guide/en/elasticsearch/reference/current/_multi_fields.html :

the field type multi_field has been removed. Instead, any of the core field types (excluding object and nested) now accept a fields parameter.

So, when recreating mapping, ES ignore the fields parameter, and the final mapping is now :

"user": {
  "type": "nested"
}

[eric-pigeon]

change nested to be a string instead of a symbol

indexes :users, type: "nested" do
  indexes :name_badge: 
end

[inkstak]

Thanks @xxxpigeonxxx

[karmi]

@xxxpigeonxxx, thanks for the hint!

@inkstak Sorry for the delay. We should do .to_s on the type then.

[inkstak]

But why adding nested to TYPES_WITH_EMBEDDED_PROPERTIES ?
Doesn't it go against the ES doc ?

[karmi]

@inkstak In what sense? Do you have a complete example which would demonstrate your point? Thanks!

[mmahalwy]

@karmi my example is I cannot get this:

mappings _all: { enabled: false } do
        indexes :text, {
          type: 'string',
          similarity: 'my_bm25',
          term_vector: 'with_positions_offsets_payloads',
          fields: {
            stemmed: {
              type: 'string',
              similarity: 'my_bm25',
              term_vector: 'with_positions_offsets_payloads',
              analyzer: 'standard'
            },
            # phonetic: {
            #   type: 'string',
            #   analyzer: 'dbl_metaphone'
            # }
          }
        }
      end

With this:

mappings _all: { enabled: false } do
        indexes :text, type: 'multi_field', similarity: 'my_bm25', term_vector: 'with_positions_offsets_payloads' do
          indexes :stemmed, type: 'string', similarity: 'my_bm25', term_vector: 'with_positions_offsets_payloads', analyzer: 'standard'
          indexes :phonetic, type: 'string', analyzer: 'dbl_metaphone'
        end
      end

Instead, in ES it becomes:

"text": {
"index": "no",
"type": "string",
"fields": {
"stemmed": {
"analyzer": "standard",
"similarity": "my_bm25",
"term_vector": "with_positions_offsets_payloads",
"type": "string"
},
"phonetic": {
"analyzer": "dbl_metaphone",
"type": "string"
}
}

[karmi]

@mmahalwy, sorry, don't get what you expect as the output?

[mmahalwy]

@karmi sorry, should explain further. According to the ES docs, multi_field is deprecated. Therefore, in the code of doing

mappings _all: { enabled: false } do
        indexes :text, type: 'multi_field', similarity: 'my_bm25', term_vector: 'with_positions_offsets_payloads' do
          indexes :stemmed, type: 'string', similarity: 'my_bm25', term_vector: 'with_positions_offsets_payloads', analyzer: 'standard'
...

You'd expect to index:

text: {
          type: 'string',
          similarity: 'my_bm25',
          term_vector: 'with_positions_offsets_payloads',
          fields: {
            stemmed: {
              type: 'string',
              similarity: 'my_bm25',
              term_vector: 'with_positions_offsets_payloads',
              analyzer: 'standard'
            },
...

But what I get:

{
  "text": {
    "index": "no",
    "type": "string",
    "fields": {
      "stemmed": {
        "analyzer": "standard",
        "similarity": "my_bm25",
        "term_vector": "with_positions_offsets_payloads",
        "type": "string"
      },
      "phonetic": {
        "analyzer": "dbl_metaphone",
        "type": "string"
      }
    }
  }
}

when it should be:

{
  "text": {
    "analyzer": "standard",
    "similarity": "my_bm25",
    "term_vector": "with_positions_offsets_payloads",
    "type": "string",
    "fields": {
      "stemmed": {
        "analyzer": "standard",
        "similarity": "my_bm25",
        "term_vector": "with_positions_offsets_payloads",
        "type": "string"
      },
      "phonetic": {
        "analyzer": "dbl_metaphone",
        "type": "string"
      }
    }
  }
}

It made sense that multi_field previously did not have the text property to have any options such as analyzer, etc. because you'd have to explicitly set a field with the same name and those options, but with multi_field gone, one would expect it'd just set them on that and have different fields if developer wishes to include

[inkstak]

@karmi your commit should fix the issue.
I didn't read well the #138 changes and didn't think that the problem was about the symbol/string equality. Thanks.

@mmahalwy you issue is different. You should open a new one so I can close this one.