# Note: '?v' add headers to the table of results.
> curl --request GET http://localhost:9200/_cat/health?v  

# Passing JSON into the curl
> curl --request GET http://localhost:9200/_search -H 'Content-Type: application/json' -d '{ <json> }'

# Index details
GET _cat/<indices|shards|segments>/<index name (wildcards allowed)>?v

# Force Merge segments
POST /<index name>/_forcemerge?[max_num_segments=<num>]&[only_expunge_deletes=<true|false>]

# List running tasks / cancel a task (find a reindex task and cancel it).
GET _tasks?actions=*reindex*&detailed
POST _tasks/oTUltX4IQMOUUVeiohTt8A:12345/_cancel

PUT <index-name>
{
    "settings" : {
        "index" : {
            "number_of_shards" : 3, 
            "number_of_replicas" : 2 
        }
    }
}

POST _reindex
{
  "source": {
    "index": "<current index name>"
  },
  "dest": {
    "index": "<new index name>"
  }
}

POST /<index>/<type use:default>
{
   "name" : "martin",
   "age" : 21
}

POST /<index>/<type use:default>/_bulk
{ "index" : { } }
{ "name" : "bill", "age" : 43 }
{ "index" : { } }
{ "name" : "carl", "age" : 12 }
etc...

GET <index>/_search 
{ <json> }

{ "_source" : [ "<path>.<field>" ], "query" : { ... } }

{ "from" : 0, "size" : 10000 }

{ "query" : { "wildcard" : { "<path>.<field>" : "<search_string>*" } } }

{ "query" : { "term" : { "<path>.<field>" : "<search_string>" } } }
{ "query" : { "terms" : { "<path>.<field>" : ["<search_string1>", "<search_string2>"] } } }

{ "query" : 
   { "dis_max" : 
      { "queries" : 
         [  { "term": { "<path>.<field>" : "<search_string>" } }, 
            { "term": { "<path>.<field>" : "<search_string>" } } 
         ]
      } 
   } 
}

{ "query": {
    "bool": {
      "must": [
        {
          "term": {"shape": "round"}
        },
        {
          "bool": {
            "should": [
              {"match": {"color": "red"}},
              {"match": {"color": "blue"}
            ]
          }
        }
      ]
    }
  }
}

GET <index>/_msearch 
{ "index" : "<index>" }
{ "query" : { "terms" : { "<path>.<field>" : ["<search_str1>", "<search_str2>"] } } }
{ "index" : "<index>" }
{ "from" : 0, "size" : 10000, "query" : { "terms" : { "<path>.<field>" : ["<srch_str3>", "<srch_str4>"] } } }
etc...

# 1. Create an index
PUT /mjn/default/_bulk
{"index":{}}
{"word":"one"}
{"index":{}}
{"word":"two"}
{"index":{}}
{"word":"three"}
{"index":{}}
{"word":"four"}
{"index":{}}
{"word":"five"}

# 2. Search the index, requesting the scroll_id and specifying the size of the dataset to be returned.
#    The search results will include a _scroll_id field.
#    (scroll=1m requests that the search context be kept alive for 1 minute).
GET /mjn/_search?scroll=1m
{"size":2}

# 3. Call the scroll API to get the next batch of results.  Can call this until no more hits returned.
POST /_search/scroll
{ "scroll":"1m",
  "scroll_id": "<scroll id>" }