I'm currently doing a query that's a mix of multi match and function score. The important bit of the JSON looks like this:
"function_score":{
"query":{
"query_string":{
"query":"some query",
"fields":["id","name","strippedDescription","colourSearch","sizeSearch"]
}
}
}
However, I also want to include results that don't necessarily match the query but have a particular numeric value that's greater than 0. I think a bool query would do this, but I don't know how to use a bool query with a function score query.
I understand that a multi match query is just shorthand for a bool query, and I could expand out the multi match query into its bool counter-part, however, I then don't know how I would do function score within that.
Any ideas? I'm on version 1.1.0 by the way.
Figured it out! I was missing the fact that you can nest multi field queries within bool queries! My final solution looks like this:
{
"query":{
"function_score":{
"query":{
"bool":{
"should": [
{
"range": {
"allBoost": {
"gt": 0
}
}
},{
"multi_match":{
"query":"some search query",
"fields":[
"id",
"name",
"description",
"category"
]
}
}
]
}
},
"functions":[
{
"filter":{
"range": {
"allBoost": {
"gt": 0
}
}
},
"script_score":{
"script":"doc['allBoost'].value"
}
},
{
"filter":{
"range": {
"allBoost": {
"lte": 0
}
}
},
"script_score":{
"script":"_score"
}
}
],
"boost_mode": "replace"
}
}
}
Related
what's difference between the following es filter query？
1. filter context for multi query conditions:
{
"query": {
"bool": {
"filter": [
{ "term": { "status": "published" }},
{ "range": { "publish_date": { "gte": "2015-01-01" }}}
]
}
}
}
must in filter context:
{
"query": {
"bool": {
"filter": [
{
"bool": {
"must": [
{ "term": { "status": "published" }},
{ "range": { "publish_date": { "gte": "2015-01-01" }}}
]
}
}
]
}
}
}
The first query is used in scenarios where you just want to filter using AND operator on different fields. By default if you write filter query in this way, it would be executed as AND operation.
The second query, in your case/scenario, does exactly as the first query (no difference, just two ways of doing same thing), however the reason we can "also" do that is to implement/cover more complex filter use-cases that uses many different AND and OR combinations.
Note that in Elasticsearch AND is represented by must while OR is represented by should clauses.
Let's say I would want to filter a scenario like I want all documents having
sales from department 101 or
sales from department 101B along with price > 150.
You probably would have to end up writing query in the below way:
POST sometestindex/_search
{
"query":{
"bool":{
"filter":[
{
"bool":{
"should":[
{
"term":{
"dept.keyword":"101"
}
},
{
"bool":{
"must":[
{
"term":{
"dept.keyword":"101B"
}
},
{
"range":{
"price":{
"gte":150
}
}
}
]
}
}
],
"minimum_should_match": 1
}
}
]
}
}
}
In short, for your scenario, first query is just a short-hand way of writing the second-query, however if you have much more complex filter logic, then you need to leverage the Bool query inside your filter as you've mentioned in your second query, as I've mentioned in the sample example.
Hope that clarifies!
I am having trouble applying a secondary filter to my elasticsearch query below. Only the first filter is matching. I want both filters to apply to the query.
"query": {
"bool": {
"must": [
{
"bool": {
"filter": {
"range": {
"#timestamp": {
"gte": "2019-03-12",
"lte": "2019-03-13"
}
}
}
}
},
{
"bool": {
"filter": {
"bool": {
"must": {
"match": {
"msg_text": "foo AND bar"
}
}
}
}
}
}
]
}
}
Well I've mentioned two solutions, first one makes use of Match Query while the second one makes use of Query String.
Also I'm assuming msg_text field is of type text.
Difference is that, query_string uses a parser, that would parse the text you mention based on the operators like AND, OR.
While match query would read the text, analyse the text and based on it constructs a bool query. In the sense you don't need to mention operators and it won't work
You can read more about them in the links I've mentioned.
1. Using Match Query
POST <your_index_name>/_search
{
"query":{
"bool":{
"filter":{
"bool":{
"must":[
{
"range":{
"#timestamp":{
"gte":"2019-03-12",
"lte":"2019-03-13"
}
}
},
{
"match":{
"msg_text":"foo bar"
}
}
]
}
}
}
}
}
2. Using Query String
POST <your_index_name>/_search
{
"query":{
"bool":{
"filter":{
"bool":{
"must":[
{
"range":{
"#timestamp":{
"gte":"2019-03-12",
"lte":"2019-03-13"
}
}
},
{
"query_string":{
"fields": ["msg_text"], <----- You can add more fields here using comma as delimiter
"query":"foo AND bar"
}
}
]
}
}
}
}
}
Technically nothing is wrong with your solution, in the sense, it would work, but I hope my answers clear, simplifies the query and helps you understand what you are trying to do.
Let me know if it helps!
Struggling with inner-hits on elasticsearch. Would appreciate any help.
I have two child types: childA and childB.
I am querying parents of childA like this
"query":{
"bool": {
"should": {
"has_child": {
"type": "ChildA",
"query": {
"match": {
"name": {
"query": "a"
}
}
}
}
}
}
}
My problem is how to include in the results all child docs of type childB as well without affecting results from the above query.
I was thinking to use inner-hits on a has_child query(type childB) for that but my query doesn't depend on childB type.
Anyone has an idea?
Thanks in advance
I found a way to include childB type docs.
I combine the following query with above query (has_child on childA type) in a filter query to get childB docs also. I am not really sure if its a good way though(thinking about performance)
{
"query":{
"bool":{
"should":[
{
"bool":{
"must_not":[
{
"has_child":{
"type":"ChildB",
"query":{
"match_all":{}
},
"inner_hits":{}
}
}
]
}
},
{
"has_child":{
"type":"ChildB",
"query":{
"match_all":{}
},
"inner_hits":{}
}
}
]
}
}
}
I am running into some serious Problems with a custom Search. All i want is a Wildcard Search in three Fields and the Result should to filtered by another field. In Elastica it results in this Query:
{
"bool":{
"should":[
{
"wildcard":{
"ean":"*180g*"
}
},
{
"wildcard":{
"titel":"*180g*"
}
},
{
"wildcard":{
"interpret":"*180g*"
}
}
],
"filter":[
{
"term":{
"genre":{
"value":"Rock",
"boost":1
}
}
}
]
}
}
Actually i can't find an error, but Elasticsearch does not give me Filtered Results. What happens? Elasticsearch returns ALL Items with the Filtered Term, either if the Boolean Shoulds are True or False. When i add the Filter as "Must" i am getting the same results? What is wrong here !?
You need to add "minimum_should_match": 1 in your bool query.
{
"bool":{
"minimum_should_match": 1,
"should":[
{
"wildcard":{
"ean":"*180g*"
}
},
{
"wildcard":{
"titel":"*180g*"
}
},
{
"wildcard":{
"interpret":"*180g*"
}
}
],
"filter":[
{
"term":{
"genre":{
"value":"Rock",
"boost":1
}
}
}
]
}
}
Say that I query:
POST /story/story/_search
{
"query":{
"bool":{
"should":[
{
"match":{
"termVariations":{
"query":"not driving",
"type":"boolean",
"operator":"AND"
}
}
},
{
"match":{
"termVariations":{
"query":"driving",
"type":"boolean",
"operator":"AND"
}
}
}
]
}
}
}
This query returned by one analyzer or another 3 documents.
How do I tell which should clause was matched? Can Elasticsearch return the matched phrase along with the result?
Thanks!
The best option here would be named queries.
You can name your query and the name of the queries that matched would be provided per document.
{
"query": {
"bool": {
"should": [
{
"match": {
"name.first": {
"query": "qbox",
"_name": "first"
}
}
},
{
"match": {
"name.last": {
"query": "search",
"_name": "last"
}
}
}
]
}
}
}
Thanks #keety! highlight was exactly what I was looking for!! :-)