Missing the key question-to-table name mapping file

[dsgv]

For the Bird dataset, there already exists a database schema file named bird_test_data.json (mapping file from questions to table names) and schema.csv under the directory CLSR\schema_retrieval\evaluation\data\bird, while for the Spider dataset, there is no mapping file from questions to table names or database schema file. So, where does the dataset used in the table selection experiment on the SpiderUnion dataset in the paper come from? I would be extremely grateful if the authors could provide it!