- Issue created by @jwag956
- 🇦🇹Austria drunken monkey Vienna, Austria
drunken monkey → made their first commit to this issue’s fork.
- 🇦🇹Austria drunken monkey Vienna, Austria
Hi @dhruv.mittal, and thanks a lot for reporting this issue!
However, I’m afraid I cannot reproduce this problem. I tried writing a regression test, but everything I tried still passed without your fix. (Also, I think it wouldn’t be correct to replace non-breaking spaces with normal ones, as your suggested code would do. I think if we use theu
modifier we’d have to explicitly list the “normal” whitespace to replace to prevent this. Also just for the sake of backwards compatibility, in case this breaks anyone’s code. The resulting string should then still be valid UTF-8. See 63042a3c.)Could you maybe add a failing test case to
HtmlFilterTest::testBasicFieldValuesProcessing()
and then revert f10c2ad7 to demonstrate that your fix actually works.