DevOps Classroom Series – 21/Sept/2021

GET API

• The Get API is used for retrieving a document where you already know ID

GET /catalog/_doc/<doc-id>
GET /<index>/<type>/<id>

UPDATE API

• The Update API is useful for updating existing document with ID

POST /<index>/_update/<id>
{
    "doc" : {
        "tags": ["technical", "eduacation" ] 
    }
}

• Update the document if present or create a new document doc_as_upsert:true does this trick

• Update the value of the document by its existing fields using script

DELETE API

• DELETE API lets you delete the document by ID

DELETE /<index>/<type>/<id>

Dealing with multiple indexes

• Operations such as search and aggregation can run against multiple indexes in same query
• The following query matches all documents

GET /_search

• Searching all documents in one index

GET /<index>/_search

• Searching all documents in multiple indexes

GET /<index-1>,<index-2>..<index-n>/_search

Searching – What is Relavent

• Text Analysis: All the fields that are of text type are analyzed by what is known as analyzer
  • The main task of analyzer is to take the value of field and break it down to terms
  • Analyzer performs the process of breaking up into terms
    • at the time of indexing
    • at the time of searching
  • Analyzer has the following components
    • Character filter: Zero or more
    • Tokenizer: Exactly one
    • Token filters: Zero or more
  • Elastic search ships with few built-in character filters which we can use or create our own analyzer. Elastic search ships with Mapping Char Filter
  • For example if you are indexing converstations (chats/emails) etc and then if you want to transfer emoticons into some text
    • 🙂 should be translated to _smile_
    • 🙁 should be translate to _sad_
  • This can be acheived through the character filter

  "char_filter": {
      "my_char_filter" :{
          "type": "mapping",
          "mappings": [
              ":) => _smile_",
              ":( => _sad_"
          ]
      }
  }
  

  • Refer Here
  • The responsibility of a tokenizer is to recieve a stream of characters and generate tokens. These tokens are used to build an inverted index (Token is roughly equivalent to word) Refer Here for official docs
  • Refer Here for token filters