RussellSpitzer commented on issue #4621: URL: https://github.com/apache/iceberg/issues/4621#issuecomment-1108706179
As @openinx says, Iceberg doesn't inherently have an ordering in its generic readers since these are not meant to be a full featured reader. In general, most engines like Spark or Trino are implementing that sort of sorting and ordering because in a generic case such ordering requires a full shuffling of the records as written. While you could write a table with a column that specified the insert ordering, reading with that ordering would be the responsibility of the reading implementation. For a specific table only written and read by pulsar you could probably guarantee a certain ordering for all reading but this wouldn't be possible for a generic table. A generic Iceberg table only sees ordering at a snapshot level as @flyrain stated and within that snapshot all rows are considered to have come into existence at the same point in time. -- 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]
