Could anyone shed any light on the nested bool should/musts here? Is it necessary or just poorly written?
(the query is generated, depending on user-input)
Welcome!
Please don't post images of text as they are hard to read, may not display correctly for everyone, and are not searchable.
Instead, paste the text and format it with </> icon or pairs of triple backticks (```), and check the preview window to make sure it's properly formatted before posting it. This makes it more likely that your question will receive a useful answer.
That being said, for this exact request, this could be indeed simplified a lot as only one bool query is needed I think.
Thanks for the reply. Will do in future.
Would one bool query still only be needed to be able to search multiple vehicle makes and models?
I am trying to refactor it but the following returns no hits as its asking for a vehicle with a make BMW and Bentley and model 1 Series, 2 Series, etc. How do I change those to OR?
GET _search
{
"from": 0,
"size": "11",
"query": {
"bool": {
"must": [
{
"terms": {
"body_type.keyword": [
"Hatchback",
"Coupe"
]
}
},
{
"terms": {
"make.name.keyword": [
"BMW",
"Bentley"
]
}
},
{
"term": {
"make.name.keyword": "Bentley"
}
},
{
"term": {
"make.name.keyword": "BMW"
}
},
{
"term": {
"model.name.keyword": "1 Series"
}
},
{
"term": {
"model.name.keyword": "2 Series"
}
},
{
"term": {
"fuel_type.keyword": "Petrol"
}
},
{
"range": {
"price": {
"lte": "70000"
}
}
},
{
"range": {
"price": {
"gte": "0"
}
}
},
{
"exists": {
"field": "geolocation"
}
}
]
}
},
"sort": {
"date_first_registered": {
"order": "desc"
}
},
"_source": true
}
EDIT I think I just had that click moment. JSON object children of a bool.must query are joined by AND. bool.should is OR. So I can do this for multiple makes and models:
GET _search
{
"from": 0,
"size": "11",
"query": {
"bool": {
"must": [
{
"terms": {
"body_type.keyword": [
"Hatchback",
"Coupe"
]
}
},
{
"terms": {
"make.name.keyword": [
"BMW",
"Bentley"
]
}
},
{
"bool": {
"should": [
{
"term": {
"make.name.keyword": "Bentley"
}
},
{
"bool": {
"must": [
{
"term": {
"make.name.keyword": "BMW"
}
},
{
"term": {
"model.name.keyword": "1 Series"
}
}
]
}
},
{
"bool": {
"must": [
{
"term": {
"make.name.keyword": "BMW"
}
},
{
"term": {
"model.name.keyword": "2 Series"
}
}
]
}
}
]
}
},
{
"term": {
"fuel_type.keyword": "Petrol"
}
},
{
"range": {
"price": {
"lte": "70000"
}
}
},
{
"range": {
"price": {
"gte": "0"
}
}
},
{
"exists": {
"field": "geolocation"
}
}
]
}
},
"sort": {
"date_first_registered": {
"order": "desc"
}
},
"_source": true
}
I'm not sure why the original query would have used constant_score for this though. Any ideas?
Different matches generate difference scores. If you want everything to have the same weight, you use constant_score. Not trying to be a jerk, but you should really read the documentation. It's actually really good for the questions you asked.