MongoDB (text search with the relevant field) aggregation problem - mongodb

Question

I have the MongoDB aggregation query
db.data.aggregate([{ "$match" : { "$text" : { "$search" : "STORAGE TYPE" } } },
{ "$group" :
{ "_id" :{"doc_type": "$doc_type" ,"title" : "$title", "player_name" : "$player_name", "player_type" : "INSTITUTION", "country_code" :"$country_code" },
"number_records" : { "$sum" : 1}
}
},
{"$match" : {"doc_type": "PATENT"} },
{"$sort":{"number_records" : -1}},
{"$limit" : 10}],
{"allowDiskuse" : true}
)
When I tried to execute the above code, it keeps on buffering for a long time, I am not getting any output. Can anyone help me?
When I used command explain(), it shows the following code:
{
"stages" : [
{
"$cursor" : {
"query" : {
"$and" : [
{
"$text" : {
"$search" : "STORAGE TYPE"
}
},
{
"doc_type" : "PATENT"
}
]
},
"fields" : {
"country_code" : 1,
"doc_type" : 1,
"player_name" : 1,
"title" : 1,
"_id" : 0
},
"queryPlanner" : {
"plannerVersion" : 1,
"namespace" : "datadocuments.data",
"indexFilterSet" : false,
"parsedQuery" : {
"$and" : [
{
"doc_type" : {
"$eq" : "PATENT"
}
},
{
"$text" : {
"$search" : "STORAGE TYPE",
"$language" : "english",
"$caseSensitive" : false,
"$diacriticSensitive" : false
}
}
]
},
"winningPlan" : {
"stage" : "FETCH",
"filter" : {
"doc_type" : {
"$eq" : "PATENT"
}
},
"inputStage" : {
"stage" : "TEXT",
"indexPrefix" : {
},
"indexName" : "title",
"parsedTextQuery" : {
"terms" : [
"storag",
"type"
],
"negatedTerms" : [ ],
"phrases" : [ ],
"negatedPhrases" : [ ]
},
"textIndexVersion" : 3,
"inputStage" : {
"stage" : "TEXT_MATCH",
"inputStage" : {
"stage" : "FETCH",
"inputStage" : {
"stage" : "OR",
"inputStages" : [
{
"stage" : "IXSCAN",
"keyPattern" : {
"_fts" : "text",
"_ftsx" : 1
},
"indexName" : "title",
"isMultiKey" : true,
"isUnique" : false,
"isSparse" : false,
"isPartial" : false,
"indexVersion" : 2,
"direction" : "backward",
"indexBounds" : {
}
},
{
"stage" : "IXSCAN",
"keyPattern" : {
"_fts" : "text",
"_ftsx" : 1
},
"indexName" : "title",
"isMultiKey" : true,
"isUnique" : false,
"isSparse" : false,
"isPartial" : false,
"indexVersion" : 2,
"direction" : "backward",
"indexBounds" : {
}
}
]
}
}
}
}
},
"rejectedPlans" : [ ]
}
}
},
{
"$group" : {
"_id" : {
"doc_type" : "$doc_type",
"title" : "$title",
"player_name" : "$player_name",
"player_type" : {
"$const" : "INSTITUTION"
},
"country_code" : "$country_code"
},
"number_records" : {
"$sum" : {
"$const" : 1
}
}
}
},
{
"$sort" : {
"sortKey" : {
"number_records" : -1
},
"limit" : NumberLong("10")
}
}
],
"ok" : 1
}
I couldn't figure out the mistake; is there any problem in aggregation, if not, how to increase the performance?

Answer 1

Your error comes from your second match stage : at this point, doc_type doesn't exist, but _id.doc_type instead. But you bettermerge this stage with the first one, to improve performance by reducing number of documents passed to the $group stage.
Your improved query will be :
db.data.aggregate([
{"$match" : { "$text" : { "$search" : "STORAGE TYPE" `},"doc_type": "PATENT" } },`
{ "$group" :
{ "_id" :{"doc_type": "$doc_type" ,"title" : "$title", "player_name" : "$player_name", "player_type" : "INSTITUTION", "country_code" :"$country_code" },
"number_records" : { "$sum" : 1}
}
},
{"$sort":{"number_records" : -1}},
{"$limit" : 10}],
{"allowDiskuse" : true}
)