Fork me on GitHub
#sql
<
2019-08-14
>
Jakub Holý (HolyJak)08:08:08

Thanks! I solved it by using JDBC directly, it was far easier. ReadableColumn doesn't help because it is called after (.getObject rs ...) while I want to replace the .getObject. I reached my goal - discovering that the value is stored as a string (not a long as I assumed) and seeing the raw data before they are parsed into an object.

seancorfield13:08:43

Ah, so you could write a custom result set builder to intercept that.

👍 4
Jakub Holý (HolyJak)06:08:53

I see, I could make a custom fn to return my RowSetBuilder and send it via :builder-fn. But it also has to implement RowBuilder , right?

seancorfield06:08:54

Yes. It's not a huge amount of code but it's probably another area I could make things easier to extend.

Jakub Holý (HolyJak)07:08:40

Thank you! Improvement is a never ending effort 🙂

seancorfield21:08:48

@U0522TWDA See https://github.com/seancorfield/next-jdbc/commit/57f146ee41c949e74a06b3f27a24b65fffa9fab8 for a new as-maps-adapter function that makes it easier to override .getObject in a custom MapResultSetBuilder. I'll add one for the as-*-arrays variants too.

❤️ 4
ikitommi08:08:09

^:--- got the porsas async client finally working correctly. It’s bit faster than the synchronous version, but uses much smaller pool, 16 vs 256. Not using the async client in real life yet, anyone interested in?

🎉 4