使用Node中的mongodb驱动程序进行不区分大小写的“ like”查询

Question

I'm looking to make a query like mentioned in the title with an array of data.

I have this line working..

collection.find({"name": {"$in":[/papa JOhn's/i,"Taco Bell"]}}).toArray(function(err, items) {

However, the data is going to be dynamic and fed into the query as an array

restaurant_data = [rest1, rest2 .... restn]

I want to essentially have something like

collection.find({"name": {"$in": /restaurant_data/i }}).toArray(function(err, items) {

My Attempt..

let test_data = ['papa john', 'taco bell']
//Get an array of the restaurant names here

collection.find({"name": {"$in": test_data.map(element => {

  /element/i

})}}).toArray(function(err, items) {

My working attempt!

I got it doing this... Not sure if it's the most efficient

let test_data = ['papa john', 'taco bell']

collection.find({"name": {"$in": test_data.map((element) => {
  var regex = new RegExp(".*" + element + ".*");
  return regex
})}}).toArray(function(err, items) {

Answer 1

collection.find({"name": {"$in":
  restaurant_data.map(e => new RegExp(e, 'i'))
}}).toArray

has better chance to work. Note that /bla/i will match blabla so

collection.find({"name": {"$in": [/bla/i] } })

will match documents with name blabla Not sure this is what you want or not.

Answer 2

If you're using mongodb 3.2+ I'd recommending you using the property Collation as using Regex might make the queries quite slower, you can use it in both creating indexes as well as querying:

db.peeps.createIndex({UserName:-1}, { collation: {locale:'en', caseLevel:false, caseFirst: "off"} )
db.peeps.find({UserName:'bob'}).collation({locale:'en', caseLevel:false, caseFirst: "off"}) // 3 results!

However, as I've faced this problem as well when I tried using collation it made my queries at least 2 times slower so I ended up sticking with inserting lowerCase data in mongodb and then validating and using toLowerCase in the payload fields instead of using Collation.