Error: Invalid operator: $regex

[thatshailesh]

Facing issue while doing something like

await test.find(
        { 
            selector: { name: {"$regex": /cat/}}
        }
    )

Expected Behavior

it should return names containing cat

Current Behavior

Current error stack

 Error: Invalid operator: $regex
    at Request._callback (/node_modules/nano/lib/nano.js:154:15)
    at Request.self.callback (node_modules/request/request.js:185:22)
    at Request.emit (events.js:198:13)
    at Request.<anonymous> (/node_modules/request/request.js:1161:10)
    at Request.emit (events.js:198:13)
    at IncomingMessage.<anonymous> (/node_modules/request/request.js:1083:12)
    at Object.onceWrapper (events.js:286:20)
    at IncomingMessage.emit (events.js:203:15)
    at endReadableNT (_stream_readable.js:1129:12)

Possible Solution

Steps to Reproduce (for bugs)

1.Just use $regex, it will throw error

Context

I am trying to do wildcard search based on name in CouchDB

Your Environment

• Version used: 8.0.1
• Browser Name and version: Chrome
• Operating System and version (desktop or mobile): OSX

[glynnbird]

Try expressing your regular expression as a string. Nano doesn't understand JavaScript regular expressions - it passes a regular expression string to CouchDB for processing e.g.

selector: { name: {"$regex": "^cat" }}

should find names beginning with "cat".

[thatshailesh]

Thank you very much, it worked :)

Can you please help me with indexing, I am working on search functionality where I will be doing a fuzzy search over name of an entity so i need to search for occurrences

I tried adding the index and i keep getting this error Reason: (required index service unavailable) text
And I am able to create an index only for json type but the field is text type
Using $regex operator and it seems very slow

What would be the best practice to implement a search in CouchDB ?
Should I use views? but its range based only

[glynnbird]

First of all the regex operator is very slow and an index can't be used to speed it up. It will inevitable cause CouchDB to do a full database scan (looking through each document body in turn) to get your answer. If you are looking for "all documents containing 'cat'" then you need to index each word in the string in question, which is what type: 'text' indexes are for. Unfortunately they are not baked into CouchDB today (stay tuned for CouchDB 3.0 which should have free-text search baked in).

Without the Lucene-based free-text search index, you're a bit stuck. You could roll your own index with a MapReduce index like so?:

function(doc) { 
  // break "str" into words
  var words = doc.str.split(/\W/); 
  for(var i in words) { 
     if (words[i]) { 
        // create an index with the lower-cased word as the key
       emit(words[i].toLowerCase(), null)
      }
   }
}

[thatshailesh]

Thanks for sharing

So after adding this view I can use find directly ?

[ghost]

How to pass regex flags then? Suppose I need case insensitive search.

[EddyTheDove]

@Cherviakov you can do it like this:

{ "firstName" : { "$regex": `${ RegExp(firstName.toLowerCase(), "i") }` }}

[ghost]

Thanks @EddyTheDove