note
ikegami
<p>You actually had the opposite problem: You had UTF-8, but the regex engine expects a string of Unicode Code Points<sup>[1]</sup>. <c>utf8::decode</c> provides the latter from the former.
<hr>
<ol>
<li><p>More specifically, it's <c>\w</c>, <c>\b</c>, <c>\d</c>, etc that are defined in terms of UCP.
</ol>
1190836
1190946