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/406#issuecomment-1040974519,https://api.github.com/repos/simonw/sqlite-utils/issues/406,1040974519,IC_kwDOCGYnMM4-DAK3,9599,2022-02-16T01:08:17Z,2022-02-16T01:08:17Z,OWNER,"I had no idea this was possible! I guess SQLite will allow any text string as the column type, defaulting to `TEXT` as the underlying default representation if it doesn't recognize the type.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",1128466114,
https://github.com/simonw/sqlite-utils/issues/406#issuecomment-1040978032,https://api.github.com/repos/simonw/sqlite-utils/issues/406,1040978032,IC_kwDOCGYnMM4-DBBw,9599,2022-02-16T01:10:31Z,2022-02-16T01:10:31Z,OWNER,"Allowing custom strings in the `create()` method, as you suggest in your example, feels like a reasonable way to support this.
```python
db[""dummy""].create({
    ""title"": str,
    ""vector"": ""array"",
})
```
I'm slightly nervous about that just because people might accidentally use this without realizig what they are doing - passing `""column-name"": ""string""` for example when they should have used `""column-name"": str` in order to get a `TEXT` column.

Alternatively, this could work:

```python
db[""dummy""].create({
    ""title"": str,
    ""vector"": CustomColumnType(""array"")
})
```
This would play better with `mypy` too I think.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",1128466114,
https://github.com/simonw/sqlite-utils/issues/406#issuecomment-1041313679,https://api.github.com/repos/simonw/sqlite-utils/issues/406,1041313679,IC_kwDOCGYnMM4-ES-P,82988,2022-02-16T09:59:51Z,2022-02-16T10:00:10Z,NONE,"The `CustomColumnType()` approach looks good. This pushes you into the mindspace that you are defining and working with a custom column type.

When creating the table, you could then error, or at least warn, if someone wasn't setting a column on a `type` or a custom column type, which I guess is where `mypy` comes in?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",1128466114,
https://github.com/simonw/sqlite-utils/issues/406#issuecomment-1041363433,https://api.github.com/repos/simonw/sqlite-utils/issues/406,1041363433,IC_kwDOCGYnMM4-EfHp,82988,2022-02-16T10:57:03Z,2022-02-16T10:57:19Z,NONE,"Wondering if this actually relates to https://github.com/simonw/sqlite-utils/issues/402 ?

I also wonder if this would be a sensible approach for eg registering `pint` based quantity conversions into and out of the db, perhaps storing the quantity as a serialised `magnitude measurement` single column string?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",1128466114,
https://github.com/simonw/sqlite-utils/issues/406#issuecomment-1248440137,https://api.github.com/repos/simonw/sqlite-utils/issues/406,1248440137,IC_kwDOCGYnMM5Kaa9J,82988,2022-09-15T18:13:50Z,2022-09-15T18:13:50Z,NONE,"I was wondering if you have any more thoughts on this? I have a tangible use case now: adding a ""vector"" column to a database to support semantic search using doc2vec embeddings ([example](https://psychemedia.github.io/storynotes/Lang_Doc2Vec.html); note that the `vtfunc` package may no longer be reliable...).","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",1128466114,