mattmartin14 commented on PR #1534: URL: https://github.com/apache/iceberg-python/pull/1534#issuecomment-2634539038
> Hey @mattmartin14 Yes, I think we can merge this soon, but I would love to drop data fusion if possible. I've been doing some incremental reviews to make sure we can do it step by step. Regarding the `rows_to_insert`, I don't think we need a join for that. Joins are often pretty expensive since it involve sorting the data. > > > > Some pseudocode to explain what I was thinking of: > > ```python > > # From before > > pred: BooleanExpression = ... > > arrow_expr = expression_to_pyarrow(pred) > > > > # What we already have > > df = tbl.scan(row_filter=pred).to_arrow() > > > > rows_to_overwrite = df.filter(arrow_expr) > > rows_to_append = df.filter(~arrow_expr) # ~ is the not expression in Arrow > > ``` Ok, let me keep plugging away at this and see if we can get rid of data fusion. On generating my test sets of data though, can I use duckdb for that (which appears you already support as a dependency)? - this would just be for my test cases to pre-generate a source and target dataframe/iceberg table -- 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: issues-unsubscr...@iceberg.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org For additional commands, e-mail: issues-h...@iceberg.apache.org
