BigQuery until now only supported
column names that were alphanumeric and underscore characters. With this new release, users have more flexibility in naming your column names, including unicode column names in various international languages, connector punctuation characters, special ASCII characters, etc. You can find the detailed documentation here: link
Today if you are loading CSV files with auto-detect or PARQUET files, BigQuery will automatically normalize your column names using underscore to obey this naming convention. With this new release, previously invalid names will now be supported. However, this also means it might affect your downstream data pipelines. That is why we wanted to create this allowlist to ensure you are aware of the implications and encourage you to make corresponding changes to your data schemas, if needed, before using the load job with the flexible column names.
Once you submit this request, you can expect your project to be activated within 7 business days. Thanks!
