Closed caicancai closed 9 months ago
I have a question that I would like to ask in this PR. I would like to ask if this situation is considered a bug.
I feel that the display results of risinglight's select can be improved, and I am trying this part of the work
Here is the output of DuckDB:
D select cast('8145308033243873281' as double) ;
┌───────────────────────────────────────┐
│ CAST('8145308033243873281' AS DOUBLE) │
│ double │
├───────────────────────────────────────┤
│ 8.145308033243873e+18 │
└───────────────────────────────────────┘
So yes, I also think the current output of RisingLight is not correct.
Here is the output of DuckDB:
D select cast('8145308033243873281' as double) ; ┌───────────────────────────────────────┐ │ CAST('8145308033243873281' AS DOUBLE) │ │ double │ ├───────────────────────────────────────┤ │ 8.145308033243873e+18 │ └───────────────────────────────────────┘
So yes, I also think the current output of RisingLight is not correct.
Thank you for your reply and patient answer. As far as I know, tidb and mysql also have similar bugs. I will try to solve this problem in risinglight. Finally, I would like to thank you for the risinglight project. As a beginner, I have learned a lot. Thank you very much for your contribution.
Attempt to divide by zero can be used as a separate test. I refer to calcite to do the same.