Please enable JavaScript.

Coggle requires JavaScript to display documents.

MongoFiltering data - Coggle Diagram

- - - - the way of saying which field is more important in an index
      - db.rent.createIndex( {"name":"text", "description":"text"}, {"name":"5", "description", "1"}, name: "my_index" )
    - - 1) The text in string field that is put into index is tokenized, i.e. cut into words, removing all whitespaces, dashes, hyphens, quotes, pattern syntax, terminal punctuations etc. example:
        Mid-gadgets present safe punctuation as.? Right! ---> ['Mid', 'Gadgets', 'present', 'safe', 'punctuation', 'as', 'Right']
      - 2) Mongo will check the language of the index to retrieve the stop words and perform tokenization without stop words, as they don't provide real value to the utterance so e.g.
        ['Is', 'this', 'a' , 'dagger', 'which', 'I', 'see', 'before', 'me'] --> ['dagger', see']
        This makes an index drastically smaller so it is crucial to have the right language of an index.
      - 3) Stemming words - i.e. getting to the root of the token, convert all the conjunctions of the word to its root. example :
        [dagger, see] --> [dag, see']
      - 4) Sensitiveness- if the index is not case-sensitive it will make everything lower-case. Also if the index is not diacritic-sensitive it will also remap all the diacritic symbols to the root symbol.
  - - - $search
        
        db.crew.find({ $text : { $search : "technical Anna"}}) - will search for documents which contain any of the terms specified
      - $language
      - $caseSensitive
      - $diacriticSensitive
    - - AND
        
        to search for documents containing all specific words, exactly one after another (in the same field), we have to put them between \" \" like this:
        db.crew.find( {$text: {$search: "\"technical Anna\""}}, {score :{$meta: "textScore"}}) .sort( {score: {$meta: "textScore}}} )
        if we don't need exact order, i.e. we also want to look for "technical and nice Anna" then we can put them in e.g. {$search: "\"technical\" Anna"}}
      - OR
        
        by default text search makes an or for search phrase, so searching for "nice smile" will return documents in which field having text index contains either "nice" or " smile" or both
      - NOT
        
        put - before the word
        db.crew.find( {$text: {$search: "technical -Anna"}}, {score :{$meta: "textScore"}}) .sort( {score: {$meta: "textScore}}} )
        this way we get only documents containing word "technical" but not "Anna"
  - - - instead of doing e.g.
        db.rent.find({$text: {$search: "\"tribeca\" loft"} })
        we can do
        db.rent.find({"neighbourhood_cleansed": "Tribeca" , $text: {$search: "loft"} })
- - - - matches arrays that contain all the elements in a provided list regardless of the order
      - db.crew.find({ skills : { $all: "technical"} })
    - - returns arrays where an element matches multiple conditions
      - db.crew.find({ skills : { $elemMatch: { name: "flying", lvl: {$gt: 7 } } } })
    - - db.crew.find({ skills : { $size: 1} })
  - - - limits the number of items in an array that a query returns
      - db.crew.find({}, { skills : { $slice: [1,3]} }) - skip 1 and show 3
      - db.crew.find({}, { skills : { $slice: 1} })
    - - Given the fact that you have an array field in the query this operator limits the array content to the first N elements that matches the query condition
      - db.crew.find({skills: "management}, { "skills.$" : 1 })
    - - display the first element to match the projection condition
- - - - filters documents where a location field is between a min and max value in meters from a specified geometry
      - find({ field: { $near: { $geometry: { type : "Point", coordinates:[lon,lat] }, $minDistance: value_in_meters, $maxDistance: value_in_meters } }, projection })