First analysis

Hibernate Search and sorting

FullTextQuery.setSort( ) is the main entry point. Sort is the Lucene object.

DistanceSortField is used for spatial:

• Coordinates or lat, long
• fieldname
• reverse

Lucene sort

Default: RELEVANCE, INDEXORDER

Otherwise Sort is a List of SortField:

• defaults like score and docid
• field name
• field type (with sorting type - up to custom Comparator)
• reverse or not
• missingValue (for string: first or last ; for numeric ; excluded others)
• FieldComparatorSource

subclasses:

• SortedNumericSortField: (for numeric multi values, compare with the min or max value)
• SortedSetSortField: (for multi values, compare with the min or max value)

Elasticsearch sort

https://www.elastic.co/guide/en/elasticsearch/reference/current/search-request-sort.html

=> JSON structure

{
    "sort" : [
        { "post_date" : {"order" : "asc"}},
        "user",
        { "name" : "desc" },
        { "age" : "desc" },
        "_score"
    ],
    "query" : {
        "term" : { "user" : "kimchy" }
    }
}

_doc is the most efficient
sort mode for multi values: min, max, sum, avg, median

Sort on nested

"sort" : [
       {
          "offer.price" : {
             "mode" :  "avg",
             "order" : "asc",
             "nested_path" : "offer",
             "nested_filter" : {
                "term" : { "offer.color" : "blue" }
             }
          }
       }
    ]

nested_path seems a constraint that could be guessed?
nested_filter allows to filter out the nested elements filtered by

Missing value

_last, _first, custom value (used as the value used to sort when the field is missing)

Ignoring unmapped field

{
"sort" : [
{ "price" :

},
],
"query" : {
"term" :

}
}

pretends the documents have a field type of long and the field is always empty in the documents.

Geo distance sorting

distance_type: sloppy_arc, arc, plane
unit: km
point as lat, long or geohash

Multiple geo points can be passed as an array containing any geo_point format
The final distance for a document will then be min/max/avg (defined via mode) distance of all points contained in the document to all points given in the sort request.

sort by script

track score even if not sorted by score

"track_scores": true