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/271#issuecomment-860142489,https://api.github.com/repos/simonw/sqlite-utils/issues/271,860142489,MDEyOklzc3VlQ29tbWVudDg2MDE0MjQ4OQ==,9599,2021-06-13T02:53:06Z,2021-06-13T02:53:06Z,OWNER,"Looks like this is the problem: https://github.com/simonw/sqlite-utils/blob/b0f9d1e494c9891ce407e27b0f5c6deeea361d30/sqlite_utils/db.py#L1724-L1742

Note how `set_cols = [col for col in all_columns if col not in pks] ` can potentially return an empty list if ALL of the columns are primary keys - but the next line of code that assigns `sql2` continues regardless, when it should instead be skipped if there are no columns in `set_cols`.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",919702451,
https://github.com/simonw/datasette/issues/1376#issuecomment-860229226,https://api.github.com/repos/simonw/datasette/issues/1376,860229226,MDEyOklzc3VlQ29tbWVudDg2MDIyOTIyNg==,9599,2021-06-13T15:29:45Z,2021-06-13T15:29:45Z,OWNER,"Oh good catch - this is a SQLite version issue.

The `fixtures.db` file used on https://latest.datasette.io/ includes a generated column (for testing purposes) which is a feature added in SQLite 3.31.0 on 2020-01-22.

https://latest.datasette.io/-/versions

But... it looks like the packaged Datasette Docker container doesn't have that SQLite version!

I should fix that. I'm renaming this issue.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",919822817,
https://github.com/simonw/datasette/issues/1376#issuecomment-860229397,https://api.github.com/repos/simonw/datasette/issues/1376,860229397,MDEyOklzc3VlQ29tbWVudDg2MDIyOTM5Nw==,9599,2021-06-13T15:31:02Z,2021-06-13T15:31:02Z,OWNER,Alternative fix would be to update that section of the documentation - if the container upgrade proves tricky I can fall back on that.,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",919822817,
https://github.com/simonw/datasette/issues/1375#issuecomment-860230385,https://api.github.com/repos/simonw/datasette/issues/1375,860230385,MDEyOklzc3VlQ29tbWVudDg2MDIzMDM4NQ==,9599,2021-06-13T15:37:49Z,2021-06-13T15:37:49Z,OWNER,"There is a feature for this at the moment, but it's a little bit hidden: you can use `?_json=col` to tell
Datasette that you would like a specific column to be exported as nested JSON: https://docs.datasette.io/en/stable/json_api.html#special-json-arguments

I considered trying to make this automatic - so it detects columns that appear to contain valid JSON and outputs them as nested objects - but the problem with that is that it can lead to inconsistent results - you might hit the API and find that not every column contains valid JSON (compared to the previous day) resulting in the API retuning  string instead of the expected dictionary and breaking your code.","{""total_count"": 1, ""+1"": 1, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",919508498,
https://github.com/simonw/datasette/issues/1376#issuecomment-860230663,https://api.github.com/repos/simonw/datasette/issues/1376,860230663,MDEyOklzc3VlQ29tbWVudDg2MDIzMDY2Mw==,9599,2021-06-13T15:39:37Z,2021-06-13T15:39:37Z,OWNER,Actually it looks like there is a PR open already that addresses this: #1296 ,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",919822817,