Problem:
There's been a recent error with the ETL run: a fix that we created (reading columns as eg integers for covid files) creates a problem when doing comparisons with previous tables (operator does not exist: character varying = integer)
This error was not caught up with our runs, because the function was inside an if loop that was never called with our synthetic data.
Solution:
One solution could be making sure that all parts of the code in the transformations are being called during the tests.
_during the test-a-thon, we used this program covr::report(covr::package_coverage()) to see the test coverage, but it's valid for R packages. Not sure if there's a similar thing for python_
Problem: There's been a recent error with the ETL run: a fix that we created (reading columns as eg integers for covid files) creates a problem when doing comparisons with previous tables (
operator does not exist: character varying = integer
)This error was not caught up with our runs, because the function was inside an
if
loop that was never called with our synthetic data.Solution: One solution could be making sure that all parts of the code in the transformations are being called during the tests.
_during the test-a-thon, we used this program
covr::report(covr::package_coverage())
to see the test coverage, but it's valid for R packages. Not sure if there's a similar thing for python_