Wrap the new PartitionedTable interface #2446
Conversation
py/server/deephaven/table.py
Outdated
| """A partitioned table is a table with one column containing like-defined constituent tables, | ||
| optionally with key columns defined to allow binary operation based transformation or joins with other like-keyed | ||
| partitioned tables.""" |
There was a problem hiding this comment.
I don't think users are going to find this clear.
|
|
||
| @property | ||
| def table(self) -> Table: | ||
| """The underlying Table.""" |
There was a problem hiding this comment.
Users will probably not find this clear. Is this the source table the PartitionedTable was created from?
There was a problem hiding this comment.
No, in the partition_by case, it is the table created with agg_by on the source table.
There was a problem hiding this comment.
This is the partitioned table. The PartitionedTable is just a wrapper around it.
| def constituent_tables(self) -> List[Table]: | ||
| """Returns all the current constituent tables.""" |
There was a problem hiding this comment.
Again, I don't think the nomenclature is clear. e.g. a name like subtables may be more apparent.
There was a problem hiding this comment.
I very intentionally didn't call this sub-tables.
py/server/deephaven/table.py
Outdated
|
|
||
| @property | ||
| def unique_keys(self) -> bool: | ||
| """Whether the keys in the underlying table are unique.""" |
There was a problem hiding this comment.
I still don't understand this method. Is this indicating if a key can be repeated in the underlying table? It isn't clear to me how a key could be repeated when methods like constituent_by_keys indicate a 1:1 mapping.
Co-authored-by: Chip Kent <5250374+chipkent@users.noreply.github.com>
Phase 1 wrapping:
Fixes #2432
Related to #2434