> I chose to modify the tests that were expecting a clean stderr (or
> some expected stderr) because they were only a few of them.
This makes me a little nervous, in that it adds a new undocumented
rule about what you need to do if you care about stderr when writing a
test (and since pretty much anybody writing a test probably HAS
extensions, they'll never notice it either).
> I chose to modify the tests that were expecting a clean stderr (or
> some expected stderr) because they were only a few of them.
This makes me a little nervous, in that it adds a new undocumented
rule about what you need to do if you care about stderr when writing a
test (and since pretty much anybody writing a test probably HAS
extensions, they'll never notice it either).