Clarify documentation of case-insensitive matching (fixes #1134)

l3kernel/l3regex.dtx

@@ -198,7 +198,7 @@
 %
 % \subsection{Characters classes}
 %
-% Character types.
+% Character properties.
 % \begin{l3regex-syntax}
 %   \item[.] A single period matches any token.
 %   \item[\\d] Any decimal digit.
@@ -413,17 +413,19 @@
 %     value $1$.
 % \end{l3regex-syntax}
 %
-% The option |(?i)| makes the match case insensitive (identifying
-% \texttt{A}--\texttt{Z} with \texttt{a}--\texttt{z}; no Unicode support
-% yet). This applies until the end of the group in which it appears, and
+% The option |(?i)| makes the match case insensitive (treating
+% \texttt{A}--\texttt{Z} and \texttt{a}--\texttt{z} as equivalent, with
+% no support yet for Unicode case changing). This
+% applies until the end of the group in which it appears, and
 % can be reverted using |(?-i)|. For instance, in
 % \verb"(?i)(a(?-i)b|c)d", the letters |a| and |d| are affected by the
 % |i| option. Characters within ranges and classes are affected
-% individually: |(?i)[Y-\\]| is equivalent to |[YZ\[\\yz]|, and
-% |(?i)[^aeiou]| matches any character which is not a vowel. Neither
-% character properties, nor |\c{...}| nor |\u{...}| are affected by the
-% |i| option.
-% ^^A \]
+% individually: |(?i)[\?-B]| is equivalent to |[\?@ABab]|
+% (and differs from the much larger class |[\?-b]|), and
+% |(?i)[^aeiou]| matches any character which is not a vowel.
+% The |i|~option has no effect on |\c{...}|, on |\u{...}|, on character
+% properties, or on character classes, for instance it has no effect at
+% all in |(?i)\u{l_foo_tl}\d\d[[:lower:]]|.
 %
 % \section{Syntax of the replacement text}
 %