issue_comments: 513246124

This data as json

html_url issue_url id node_id user created_at updated_at author_association body reactions issue performed_via_github_app
https://github.com/simonw/sqlite-utils/issues/42#issuecomment-513246124 https://api.github.com/repos/simonw/sqlite-utils/issues/42 513246124 MDEyOklzc3VlQ29tbWVudDUxMzI0NjEyNA== 9599 2019-07-19T14:18:35Z 2019-07-19T14:19:40Z OWNER

How about the Python version? That should be easier to design.

python db["dea_sales"].extract( columns=["company_name", "company_address"], to_table="companies" ) If we want to transform the extracted data (e.g. rename those columns) maybe support a transform= argument?

python db["dea_sales"].extract( columns=["company_name", "company_address"], to_table="companies", transform = lambda extracted: { "name": extracted["company_name"], "address": extracted["company_address"], } ) This would create a new "companies" table with three columns: id, name and address.

Would also be nice if there was a syntax for saying "... and use the value from this column as the primary key column in the newly created table".

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
 470345929