For searches I search in, for example, the fields field2 and field3, but I need to score with the field3 of the first group. For example, sequence 1 has more value than sequence 2, and is more than 3 if it exists. There are not always 2.3 or more sequences. Sometimes there is only one,
Should I create an index with the first_group, and delete those data from the first index documents? Or is it possible to work with this type of data in a single index?
Maybe my level of English does not allow me to advance in what I want to say.
{
"field1" : "sometext",
"field2" : "sometext",
...
"first_group" : [
{
"sequence" : 1, // Integer autoincrement
"field2" : "sometext",
"field3" : "sometext",
"date" : "20170120" // Format YYYYMMDD not mapping. ES put type text
},
{
"sequence" : 2, // Integer autoincrement
"field2" : "sometext2",
"field3" : "sometext2",
"date" : "20170121" // Format YYYYMMDD not mapping. ES put type text
},
...
],
"other_group" : [
{
// Similar
}
]
"other_group2" : [
{
// Similar
}
]
}