sohailshk opened a new pull request, #4305: URL: https://github.com/apache/iceberg-python/pull/4305
Closes #4304 # Rationale for this change `validate_partition_name` returned early when the partition name was not also a schema column, so its empty-name and uniqueness checks only ran for names that matched a column. `update_spec` could then commit a spec with two fields of the same name (for example through `rename_field`), and `create_table` did not check uniqueness at all because `assign_fresh_partition_spec_ids` passed a fresh set of names for every field. This runs the empty-name and uniqueness checks first, and tracks the names seen so far in `assign_fresh_partition_spec_ids`. ## Are these changes tested? Yes: - `test_update_spec_rejects_duplicate_partition_name` in `tests/table/test_init.py`: renaming a field to an existing name, adding a field with a name that is being renamed away, and an empty name now raise. - `test_new_table_metadata_with_duplicate_partition_name_fails` in `tests/table/test_metadata.py`: new table metadata with two partition fields of the same name now raises. Both fail on `main` and pass with this change. `tests/table` and `tests/test_schema.py`: 890 passed. `ruff check` and `ruff format --check` (0.14.3) pass on the changed files. ## Are there any user-facing changes? Partition specs with duplicate field names, which were accepted before, now raise `ValueError: Partition name has to be unique: <name>`. -- 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]
