sl255051 commented on PR #10678:
URL: https://github.com/apache/iceberg/pull/10678#issuecomment-2225993288
> @sl255051 appreciate you are taking the stub for the PR.
>
> But I am wondering why do you think column name case insensitivity is the
right behavior when building PartitionSpec? I think in iceberg schema we can
have both column named `data` and `DATA` with each different field id assigned,
like below
>
> ```
> table {
> 1: id: required int
> 2: data: required string
> 3: DATA: required string
> }
> ```
>
> Would this change introduce additional ambiguity when resolve a column
name in a case insensitive way?
Thanks for taking the time to review my PR. I did notice that the Schema
object uses a simple Map<String, Integer> for column names which means the
schema is case sensitive. But I wonder if that is a bug too. I believe
partition columns should be case-insensitive based on this issue
https://github.com/apache/iceberg/issues/83. That issue says to make Iceberg
case-insensitive. I can see lots of work was done to enable case-insensitivity
in Iceberg. Several objects even have multiple methods to enable
case-insensitivity. Take the Schema object as an example. If case-insensitivity
is not a feature of Iceberg why would that class have both methods, `findField`
and `caseInsensitiveFindField`?
In summary, I believe case-insensitivity is the correct path forward. I can
accept that I may not have implemented in the best way. If that is the case I
would appreciate some pointers on how best to implement case-insensitivity.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
