On 2/27/20 2:56 AM, Hannes Wallnöfer wrote:
Changing RawHtml to StringContent is a significant behavioral change. It's not
explicitly stated in the doc comment spec whether the text may contain HTML,
but it is reasonable to infer from the descriptions of {@code} and {@literal}
that it may be HTML content. Is this change necessary?
You are right, that change was mostly driven by personal preference, which is
not a good guideline.
I reversed it in the new webrev.
I'm pleased that reversing it did not break the underlying fix.
We should maybe (separately) clarify the spec, so that you don't have to
infer the behavior.
We should maybe (separately) have tests for the behavior (explicit or
implicit), so that we know if/when the behavior changes.
-- Jon
