ColinLeeo commented on issue #1069: URL: https://github.com/apache/tsfile/issues/1069#issuecomment-6905835178
Investigated the current state on both implementations: - **Java**: `TSEncoding.TYPE_SUPPORTED_ENCODINGS` lists `DICTIONARY` only for `TEXT`/`STRING`, and `TSEncodingBuilder.Dictionary.getEncoder()` throws `UnSupportedDataTypeException` for any other type. `DictionaryEncoder` itself is string-only (`HashMap<Binary, Integer>` + `List<Binary>`). - **C++**: the decoder/encoder factories accept `DICTIONARY` only for `STRING`/`TEXT` as well. So numeric dictionary encoding would be a new feature in both implementations, and it has to be introduced together (shared on-disk format, cross-language read compatibility, schema-layer type restrictions, tests). For that reason this is deferred from the 2.5.1 milestone — it can be picked up later as a combined introduction in Java and C++. Removed from the milestone and marked `deferred`. -- 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]
