Elasticsearch Mapping, Data Types, and Document Operations Guide

Elasticsearch mappings describe how each field in a document should be indexed; they can be generated automatically or defined manually using the _mapping API. Example to retrieve the current mapping:

curl '172.16.1.127:9200/get-together/_doc/_mapping?pretty'

Automatic mapping creates fields on the fly, while manual mapping lets you specify types such as text, keyword, date, long, boolean, etc. Example of creating an index with a document:

curl -XPUT '172.16.1.127:9200/my_index/_doc/1?pretty' -H 'Content-Type: application/json' -d '{"name":"Late Night with Elasticsearch","date":"2013-10-25T19:00"}'

Basic data types covered include:

String – use text for analyzed fields and keyword for exact matches.

Number – choose byte, short, int, long, float, or double depending on range and precision.

Date – stored as date type, default ISO‑8601 format or custom formats via the format option.

Boolean – stored as boolean values true/false.

Array – any basic type can be an array without extra mapping.

Multi‑fields allow the same field to be indexed in different ways, for example adding a non‑indexed sub‑field:

curl -XPUT '172.16.1.127:9200/blog/_mapping/posts?pretty' -H 'Content-Type: application/json' -d '{"posts":{"properties":{"tags":{"type":"text","index":true,"fields":{"verbatim":{"type":"text","index":false}}}}}}'

Predefined fields such as _source, _all, _id, _type, and _index have special roles: _source stores the original JSON, _all (now deprecated) concatenated searchable text, and the others identify documents within an index.

Document updates can be performed by re‑indexing a new document, using the update API with a partial doc, an upsert clause, or a painless script. Example of a scripted price increase:

curl -XPOST '172.16.1.127:9200/online-shop/_doc/1/_update?pretty' -H 'Content-Type: application/json' -d '{"script":{"source":"ctx._source.price += params.price_diff","params":{"price_diff":10}}}'

Version numbers enable optimistic concurrency control. Supplying ?version=2 ensures the operation only succeeds if the document is at that version, otherwise a version_conflict_engine_exception is returned. The retry_on_conflict parameter can automatically retry updates.

Deletion can be done per‑document, by query, or at the index level. Deleting a document marks it as removed; the actual space is reclaimed later. Example:

curl -XDELETE '172.16.1.127:9200/online-shop/_doc/1?pretty'

Indexes can be deleted, closed, or opened. Closing an index frees resources while keeping the data on disk for later use.