Index data to elasticsearch from SQL Server

Elastic Stack Elasticsearch
1

I have a database in my SQL Server which has multiple tables and the size of the database is around 25 GB.
I want to index this data to Elasticsearch 5.6 (As it supports mapping of multiple types). However the column names in some of the tables have invalid name such as a [.] in between words, etc.
I'd like to convert all the tables into a single json file and then fix the field names and then index the data to elasticsearch. I'm having a hard time figuring out how to do this. I've read some links such as this and this but I need to fix the column names before indexing.

How do I convert the SQL tables (multiple) into json and then after fixing the names, index it to elasticsearch? Are there any other ways to fix this?

2

Hello,

we can load data by writing the query, and the names which are not proper can be replaced by giving alias names,
Ex
select id_value as id, name.keyword as name from sample

3

there are hundreds of such columns. I was thinking of doing a search and replace after converting the tables into JSON as that feels easier

4

yup, as per my knowledge,

create a new table with proper data and columns and load the data or else using alias names

5

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.