[Connection cell: conn1]
[Connection cell: conn2]
[SQL cell: using conn2]
Let's say we restart the runtime and evaluate the first connection cell. Currently, the SQL cell would pick up conn1 and even starting the second connection wouldn't roll that back.
This is a regression from #2. The corresponding change was made to update the selected database type, in case the user changes the connection cell to use different database and assigns into the same variable name.
Scenario:
Let's say we restart the runtime and evaluate the first connection cell. Currently, the SQL cell would pick up
conn1
and even starting the second connection wouldn't roll that back.This is a regression from #2. The corresponding change was made to update the selected database type, in case the user changes the connection cell to use different database and assigns into the same variable name.
I added tests for both #2 and the regression :)