issue_comments: 1312556044

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/datasette/issues/1882#issuecomment-1312556044 https://api.github.com/repos/simonw/datasette/issues/1882 1312556044 IC_kwDOBm6k_c5OPAQM 9599 2022-11-12T19:29:11Z 2022-11-12T19:29:11Z OWNER

Thought of an edge-case: with sqlite-utils one feature I really like is that I can pipe data into it without caring if the table already exists or not:

cat data.json | sqlite-utils insert my.db mytable -

How could this new API support that?

I thought about adding a "create": true option to /db/table/-/insert which creates the table if it doesn't already exist, but if I do that I'll need to start adding other options to that endpoint - to set the primary key, add foreign keys and suchlike - which would be ignored except for the cases where the table was being created from scratch.

This doesn't feel right to me - I want to keep those options here, on /db/-/create.

One idea I had was to implement it such that you can call /db/-/create multiple times for the same table, but only if you are using the "rows" option. If so, and if the rows can be safely inserted, it would let you do that.

But instead, I'm going to outsource this to the CLI tool I plan to write that feeds data into this API. I'm already planning to use that tool for CSV inserts (so the API doesn't need to accept CSV directly). I think it's a good place for other usability enhancements like "insert this, creating the table if it does not exist" as well.

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
 1435294468