For most of ingested time-series data formats, TIMESTAMP column gets synthesized automatically. However, if a user includes a column with such name in the dataset to begin with, then that currently results in a duplicate column, which breaks the Python library. This should not be permitted, and such datasets should be rejected during upload (and tryIngest). This change should not affect the data formats that were designed to include TIMESTAMP column from the data producers. SITES data have examples of that.
For most of ingested time-series data formats, TIMESTAMP column gets synthesized automatically. However, if a user includes a column with such name in the dataset to begin with, then that currently results in a duplicate column, which breaks the Python library. This should not be permitted, and such datasets should be rejected during upload (and tryIngest). This change should not affect the data formats that were designed to include TIMESTAMP column from the data producers. SITES data have examples of that.