Re: [PR] [Python] Support tree-model TsFiles in TsFileDataFrame [tsfile]

["ColinLeeo (via GitHub)" <gi...@apache.org>]

ColinLeeo commented on code in PR #816:
URL: https://github.com/apache/tsfile/pull/816#discussion_r3498805949


##########
python/tsfile/dataset/reader.py:
##########
@@ -230,6 +245,105 @@ def _cache_metadata_table_model(self):
             )
             sys.stderr.flush()
 
+    def _cache_metadata_tree_model(self):
+        """Build the in-memory catalog by synthesizing one virtual table.
+
+        Tree TsFiles have no schema, so we materialize a single
+        ``TableEntry``: table name = the shared root segment, tag columns
+        = ``_col_1..._col_{N_max}`` (one per remaining path segment),
+        fields = union of measurements across devices. Per-device
+        ownership is preserved by registering only the
+        ``(device_id, field_idx)`` pairs actually written on disk in
+        ``series_stats_by_ref``.
+        """
+        metadata_groups = self._reader.get_timeseries_metadata(None)

Review Comment:
   In table mode, non-numeric types are filtered out, but the tree path doesn't do this, so the two modes are inconsistent. We should keep the logic consistent — or revisit why table mode
   filtered non-numeric types in the first place.



-- 
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: notifications-unsubscribe@tsfile.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org