I need to destructurate my index mapping.
My index has the following mapping
"A": {
"properties": {
"B": {
"properties": {
-c
-d
-e
}
}
}
}
What I need is to delete "one level" in order to have a mapping like this:
"A": {
"properties": {
-c
-d
-e
}
}
Is it possible to obtain this result without reindexing all my data?
Short answer, No.
Longer answer, also No. This question has been asked so many times. The answer will always be no and this is why :
You can only find that which is stored in your index. In order to make your data searchable, your database needs to know what type of data each field contains and how it should be indexed. If you switch a field type from e.g. a string to a date, all of the data for that field that you already have indexed becomes useless. One way or another, you need to reindex that field.
This applies not just to Elasticsearch, but to any database that uses indices for searching. And if it isn't using indices then it is sacrificing speed for flexibility.
Elasticsearch (and Lucene) stores its indices in immutable segments — each segment is a “mini" inverted index. These segments are never updated in place. Updating a document actually creates a new document and marks the old document as deleted. As you add more documents (or update existing documents), new segments are created. A merge process runs in the background merging several smaller segments into a new big segment, after which the old segments are removed entirely.
Typically, an index in Elasticsearch will contain documents of different types. Each _type has its own schema or mapping. A single segment may contain documents of any type. So, if you want to change the field definition for a single field in a single type, you have little option but to reindex all of the documents in your index.
If you are interested with more info, you can read the rest of the excerpt here by Clinton Gormley.
I also suggest the following readings :
You have to create a new index with the updated (one level deleted) mapping. You cannot updated the same mapping to achieve what you want.
