scalar-schemaを使ってパーティションキーあり、クラスタリーキーなし、セカンダリインデックスありの条件のテーブルをAmazon DynamoDBに登録したい about scalardb HOT 5 CLOSED

@scalar-kato Thank you for reporting this issue!

This error happened when a secondary index would be registered without clustering keys.
It was caused because the condition to make Attribute Definitions was wrong.
I've fixed it at #189.

By the way, I'm not sure why you want to use "BOOLEAN" as a secondary index type.
The query with the index would be too slow because it goes through all records to check deleted records.
For example, you can create a deleted table and insert the keys of records that you want to delete asynchronously.

from scalardb.

@scalar-kato

    "clustering-key": [
    ],

これを足してもだめですか？
（ "clustering-key"のキーは必須です）

from scalardb.

@feeblefakie
提案いただいたjson形式で再度実行しましたが同様のエラーが発生しました。

from scalardb.

@scalar-kato I agree with @yito88.
I feel the schema needs to be re-designed.
Using a secondary index for low cardinality attribute is a bad practice in any databases since it will retrieve mos records. (For example, we have 1 million records and 300k records are marked as deleted then the secondary index access will retrieve the 300k records.)

from scalardb.

@yito88 @feeblefakie
Thank you for the response!
I now understand that secondary indexes should not be used for columns with low cardinality, and I will fix the DB schema in the future.

from scalardb.