查询具有多语言列的表

Question

In table object_addresses I have 3 columns (for 3 languages) name_en, name_it, name_sk. In the future columns for other languages will be added.

Is it possible somehow to make queries SELECT name_{en, it, sk} FROM object_addresses to such a table depending on the required language, without creating 3 queries for every language and choosing proper queries in switch-case statement?

Concatenation name_ + en doesn't work.

I use spring-boot-starter-data-jpa version 2.4.0, PostgreSQL JDBC Driver (ver. 42.2.5, JDBC4.2)

Answer 1

There are two ways to solve this problem:

1. create your SQL dynamically. This can be as simple as using string concatenation and executing it using a JdbcTemplate . Beware though that it is easy to create SQL injection vulnerabilities this way. While you didn't mention it in your question, you tagged it with spring-data-jpa . Spring Data JPA doesn't offer any special mechanism for this. You'll have to write a custom method.

2. The proper way to fix this is to fix your schema. When you have a situation where "future columns [like this] will be added" in 99% of the cases your schema design is broken. Fix it, by extracting those texts in a separate table, referencing the original one, plus an additional key field demarcating the language. Or alternatively the original table contains just a text_key which then is referenced in the new table, this way the new table can hold texts from various sources. You probably don't want to navigate to these texts using entity references, but a dedicated query which will probably benefit from some serious caching.