-where \fItagregexp\fP is used to match the lines that must be tagged.
-It should not match useless characters. If the match is
-such that more characters than needed are unavoidably matched by
-\fItagregexp\fP, it may be useful to add a \fInameregexp\fP, to
-narrow down the tag scope. \fBctags\fP ignores regexps without a
-\fInameregexp\fP. The syntax of regexps is the same as in emacs.
+where \fItagregexp\fP is used to match the tag. It should not match
+useless characters. If the match is such that more characters than
+needed are unavoidably matched by \fItagregexp\fP, it may be useful to
+add a \fInameregexp\fP, to narrow down the tag scope. \fBctags\fP
+ignores regexps without a \fInameregexp\fP. The syntax of regexps is
+the same as in emacs. The following character escape sequences are
+supported: \\a, \\b, \\d, \\e, \\f, \\n, \\r, \\t, \\v, which
+respectively stand for the ASCII characters BEL, BS, DEL, ESC, FF, NL,
+CR, TAB, VT.
+.br
+The \fImodifiers\fP are a sequence of 0 or more characters among
+\fIi\fP, which means to ignore case when matching; \fIm\fP, which means
+that the \fItagregexp\fP will be matched against the whole file contents
+at once, rather than line by line, and the matching sequence can match
+multiple lines; and \fIs\fP, which implies \fIm\fP and means that the
+dot character in \fItagregexp\fP matches the newline char as well.
+.br
+The separator, which is \fB/\fP in the examples, can be any character
+different from space, tab, braces and \fB@\fP. If the separator
+character is needed inside the regular expression, it must be quoted
+by preceding it with \fB\\\fP.
+.br
+The optional \fB{\fP\fIlanguage\fP\fB}\fP prefix means that the tag
+should be
+created only for files of language \fIlanguage\fP, and ignored
+otherwise. This is particularly useful when storing many predefined
+regexps in a file.
+.br
+In its second form, \fIregexfile\fP is the name of a file that contains
+a number of arguments to the \fI\-\-regex\=\fP option,
+one per line. Lines beginning with a space or tab are assumed
+to be comments, and ignored.
+