Hyper for BigQuery

What Hyper reads

Hyper does not keep a copy of your BigQuery data. It reads BigQuery each time you ask. It can see every Google Cloud project where your Google account has a role. It can list projects, datasets, tables, models, and saved functions, read the columns and rows of a table, and read past queries and other jobs. A query that only reads data, with one SELECT statement, runs without asking and returns up to 200 rows at a time. By default, a query stops before it scans more than 1 GB.

What Hyper can do

Run any other SQL, such as SQL that adds, changes, or deletes rows, or that creates or drops tables. Run load, copy, and export jobs. Hyper asks you before each of these, and you cannot choose to always allow them. Create, change, and delete datasets and tables, add rows, change who can see a dataset or table, and manage saved functions, machine learning models, and rules that limit which rows people can see. Hyper asks you before each of these, unless you chose to always allow that kind of action. Hyper cannot upload files from your computer.

Before you connect

Hyper can read every table in every project where you have a role. Each query is billed to the Google Cloud project that runs it. After the first 1 TiB each month, an on-demand query costs $6.25 for each TiB (about 1,100 GB) that it scans. SQL that changes or deletes rows takes effect at once. You can restore a deleted dataset or table for up to 7 days, and after that it is gone for good. A change to who can see a dataset or table can make it public. Anyone who can write data into your tables can add text that hides instructions that try to trick an assistant, so check each change before you approve it.