Closed marnixhoh closed 2 years ago
For each row, all of the column data is buffered and then the event is emitted for each row. This behavior is the same no matter how many columns are in your data set. The SQL server only sends the column selected over the network, and this module will always emit to the code all the data received over the network; no additional filtering of the data is performed by this library.
@dougwilson Thank you for clarifying! I misinterpreted the docs. The docs are talking about it not being possible to stream individual fields. But selecting specific columns is of course possible.
The documentation says that rows will always be buffered entirely, whether individual columns are selected or not.
My question is, how should this be interpreted?
Will selecting individual columns like below, result in the entire stream to be buffered first? (I.e. making the streaming mechanism useless). Or does it mean that even though individual columns are selected, the entire row will always be included in the stream? (Some unnecessary bandwidth, but not the end of the world)
Please let me know if I should clarify anything :)
Thanks!