doc.crossroads.names.name refers to this data:
"crossroads": {
"names": [
{
"name": "Travessa do Salão", ...
},
{
"name": "Rua do Alquebe", ...
},
{
"name": "Travessa do Sarraipo", ...
},
{ ... }
]
}
What your SEARCH expression does is look for the three tokens [ "rua",
"do", "sarraipo" ]
in *all* of the name tokens [ "travessa", "do", "salao", "rua", "do",
"alqueb", "travessa", "do", "sarraipo", ... ]
and it does find all three of them *somewhere*.
There is no way to express that you want the tokens to appear in a single
object yet, but you can use your current expression to pre-filter documents
and add a FILTER to ignore the undesired matches, e.g. like:
LET tokens = TOKENS("Rua do Sarraipo", "text_en")
FOR doc IN v_roads
SEARCH ANALYZER(tokens ALL IN doc.crossroads.names.name, "text_en")
FILTER LENGTH(doc.crossroads.names[* FILTER tokens ALL IN
TOKENS(CURRENT.name, "text_en")]) > 0
...
On Wednesday, April 13, 2022 at 9:45:23 AM UTC+2 [email protected] wrote:
> Hello,
>
> Arango DB documentation contains token search example where search for
> movies with both dinosaur and park in the description (
> https://www.arangodb.com/docs/stable/arangosearch-fulltext-token-search.html
> )
>
> My case:
> v_roads view :
> {
> "cleanupIntervalStep": 2,
> "commitIntervalMsec": 1000,
> "consolidationIntervalMsec": 1000,
> "globallyUniqueId": "h7AEB0046C8EF/1716517141",
> "links": {
> "roads": {
> "analyzers": [
> "text_en"
> ],
> "fields": {
> "country": {
> "fields": {
> "names": {
> "fields": {
> "name": {}
> }
> }
> }
> },
> "road": {
> "fields": {
> "names": {
> "fields": {
> "name": {}
> }
> }
> }
> },
> "location": {
> "analyzers": [
> "hfts_geo_json"
> ]
> },
> "houseNumber": {
> "fields": {
> "names": {
> "fields": {
> "name": {}
> }
> }
> }
> },
> "district": {
> "fields": {
> "names": {
> "fields": {
> "name": {}
> }
> }
> }
> },
> "crossroads": {
> "fields": {
> "names": {
> "fields": {
> "name": {}
> }
> }
> }
> },
> "place": {
> "fields": {
> "names": {
> "fields": {
> "name": {}
> }
> }
> }
> },
> "postalCode": {
> "fields": {
> "names": {
> "fields": {
> "name": {}
> }
> }
> }
> }
> },
> "includeAllFields": false,
> "storeValues": "none",
> "trackListPositions": false
> }
> },
> "consolidationPolicy": {
> "type": "tier",
> "segmentsBytesFloor": 2097152,
> "segmentsBytesMax": 5368709120,
> "segmentsMax": 10,
> "segmentsMin": 1,
> "minScore": 0
> },
> "id": "1716517141",
> "primarySort": [],
> "writebufferActive": 0,
> "primarySortCompression": "lz4",
> "storedValues": [],
> "type": "arangosearch",
> "writebufferIdle": 64,
> "writebufferSizeMax": 33554432
> }
>
> My query :
> FOR doc IN v_roads SEARCH ANALYZER(TOKENS("Rua do Sarraipo", "text_en")
> ALL IN doc.crossroads.names.name, "text_en") SORT BM25(doc) DESC,
> doc.rank DESC LIMIT 10 RETURN {"road" : doc.road.names, "crossroad" :
> doc.crossroads.names, "roadId" : doc._key}
>
> This query returns unexpected result (roadId = 13445), this item does not
> contain all tokens in crossroads.names.name attributes, see result_1.txt,
> result_2.txt in attachments
>
> Could you explain is this expected behaviour or a bug?
>
--
You received this message because you are subscribed to the Google Groups
"ArangoDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/arangodb/908a45c8-cfa0-475e-82af-7c88d58b5c47n%40googlegroups.com.
