Fokko commented on code in PR #848: URL: https://github.com/apache/iceberg-python/pull/848#discussion_r1666586779
########## mkdocs/docs/configuration.md: ########## @@ -299,4 +299,8 @@ PyIceberg uses multiple threads to parallelize operations. The number of workers # Backward Compatibility -Previous versions of Java (`<1.4.0`) implementations incorrectly assume the optional attribute `current-snapshot-id` to be a required attribute in TableMetadata. This means that if `current-snapshot-id` is missing in the metadata file (e.g. on table creation), the application will throw an exception without being able to load the table. This assumption has been corrected in more recent Iceberg versions. However, it is possible to force PyIceberg to create a table with a metadata file that will be compatible with previous versions. This can be configured by setting the `legacy-current-snapshot-id` entry as "True" in the configuration file, or by setting the `PYICEBERG_LEGACY_CURRENT_SNAPSHOT_ID` environment variable. Refer to the [PR discussion](https://github.com/apache/iceberg-python/pull/473) for more details on the issue +Previous versions of Java (`<1.4.0`) implementations incorrectly assume the optional attribute `current-snapshot-id` to be a required attribute in TableMetadata. This means that if `current-snapshot-id` is missing in the metadata file (e.g. on table creation), the application will throw an exception without being able to load the table. This assumption has been corrected in more recent Iceberg versions. However, it is possible to force PyIceberg to create a table with a metadata file that will be compatible with previous versions. This can be configured by setting the `legacy-current-snapshot-id` property as "True" in the configuration file, or by setting the `PYICEBERG_LEGACY_CURRENT_SNAPSHOT_ID` environment variable. Refer to the [PR discussion](https://github.com/apache/iceberg-python/pull/473) for more details on the issue + +# Nanoseconds Support + +PyIceberg currently only supports upto microsecond precision in its TimestampType. PyArrow timestamp types in 's' and 'ms' will be upcast automatically to 'us' precision timestamps on write. Timestamps in 'ns' precision can also be downcast automatically on write if desired. This can be configured by setting the `downcast-ns-timestamp-on-write` property as "True" in the configuration file, or by setting the `PYICEBERG_DOWNCAST_NS_TIMESTAMP_ON_WRITE` environment variable. Refer to the [nanoseconds timestamp proposal document](https://docs.google.com/document/d/1bE1DcEGNzZAMiVJSZ0X1wElKLNkT9kRkk0hDlfkXzvU/edit#heading=h.ibflcctc9i1d) for more details on the long term roadmap for nanoseconds support Review Comment: ```suggestion PyIceberg currently only supports microsecond precision in its TimestampType. PyArrow timestamp types in 's' and 'ms' will be upcast automatically to 'us' precision timestamps on write. Timestamps in 'ns' precision can also be downcast automatically on write if desired. This can be configured by setting the `downcast-ns-timestamp-on-write` property as "True" in the configuration file, or by setting the `PYICEBERG_DOWNCAST_NS_TIMESTAMP_ON_WRITE` environment variable. Refer to the [nanoseconds timestamp proposal document](https://docs.google.com/document/d/1bE1DcEGNzZAMiVJSZ0X1wElKLNkT9kRkk0hDlfkXzvU/edit#heading=h.ibflcctc9i1d) for more details on the long term roadmap for nanoseconds support ``` -- 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
