detect UTF-16 surrogates as bad, use red for overlong

1 files changed, 11 insertions, 7 deletions

diff --git a/uxd.1 b/uxd.1
index ca39178..90b23a3 100644
--- a/uxd.1
+++ b/uxd.1
@@ -27,7 +27,7 @@ level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
 .\" new: \\n[rst2man-indent\\n[rst2man-indent-level]]
 .in \\n[rst2man-indent\\n[rst2man-indent-level]]u
 ..
-.TH "UXD" 1 "2024-12-17" "0.2.1" "Urchlay's Utilities"
+.TH "UXD" 1 "2024-12-18" "0.2.1" "Urchlay's Utilities"
 .SH NAME
 uxd \- UTF-8 hex dumper
 .SH SYNOPSIS
@@ -276,15 +276,11 @@ the space, ↵ for a newline.  Hopefully this is an improvement over
 the usual practice of printing these as periods, like standard hex
 dumpers do. The Unicode BOM (byte order marker, U+FEFF) is printed
 as a purple letter B.
-.sp
-Note: Overlong encodings (e.g. codepoints U+0000 to U+007F encoded
-as 2 or more bytes) are rendered as � (U+0FFD) in reverse video
-purple.
 .TP
 .B \fBred\fP
 Invalid UTF\-8 sequences. These are rendered as � (U+0FFD) with
-a red background, to make them stand out. Examples of invalid
-sequences:
+a red background, to make them stand out. Invalid
+sequences are:
 .INDENT 7.0
 .INDENT 3.5
 .INDENT 0.0
@@ -296,8 +292,16 @@ Continuation bytes that aren\(aqt preceded by a valid prefix byte.
 .IP \(bu 2
 Truncated UTF\-8 sequence at EOF.
 .IP \(bu 2
+UTF\-16 surrogates (codepoints U+D800 to U+DFFF).
+.IP \(bu 2
 Codepoints above U+10FFFF, which are disallowed by RFC 3629.
+.IP \(bu 2
+Overlong encodings (e.g. codepoints U+0000 to U+007F encoded
+as 2 or more bytes).
 .UNINDENT
+.sp
+Each occurrence of any of the above will increment the "Bad
+Sequences" count, if the \fB\-i\fP option is used.
 .UNINDENT
 .UNINDENT
 .UNINDENT