% utf8unkn.tex %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Feb. 6. 2003 Petr Olsak % % Oct. 2012: More robust macros by \long, \noexpand, \badutfinput (P.O.) % % This macro sets all others UTF-8 codes (not declared formerly or % later) as "unknown" by encTeX primitives. You need the version % Feb. 2003 of encTeX or later. % % You can re-define the \warntwobytes and \warnthreebytes macros % if you need it. The default value of these macros is that % the black box is typeset and warning is printed by \message % and the unknown code is printed without any change to \write files. \def\setunknowncodes {\bgroup \def\a ##1{\mubyte \warntwobytes ####2 ##1\endmubyte} \a ^^c2 \a ^^c3 \a ^^c4 \a ^^c5 \a ^^c6 \a ^^c7 \a ^^c8 \a ^^c9 \a ^^ca \a ^^cb \a ^^cc \a ^^cd \a ^^ce \a ^^cf \a ^^d0 \a ^^d1 \a ^^d2 \a ^^d3 \a ^^d4 \a ^^d5 \a ^^d6 \a ^^d7 \a ^^d8 \a ^^d9 \a ^^da \a ^^db \a ^^dc \a ^^dd \a ^^de \a ^^df \def\a ##1{\mubyte \warnthreebytes ####3 ##1\endmubyte} \a ^^e0 \a ^^e1 \a ^^e2 \a ^^e3 \a ^^e4 \a ^^e5 \a ^^e6 \a ^^e7 \a ^^e8 \a ^^e9 \a ^^ea \a ^^eb \a ^^ec \a ^^ed \a ^^ee \a ^^ef \egroup} \mubyte \writeparameter \relax \endmubyte \def\writeparameter{} \long\def\warntwobytes #1#2{% \ifx\writeparameter\relax % write parameter copies unknown code \expandafter\noconvert\noexpand#1% \expandafter\noconvert\noexpand#2% to output without any change \else \if\noexpand#2\relax \badutfinput \else \ifnum`#2<128 \badutfinput \else \missingutfchar{\expandafter\noconvert\noexpand#1% \expandafter\noconvert\noexpand#2}% \fi\fi\fi} \long\def\warnthreebytes #1#2#3{\ifx\writeparameter\relax \expandafter\noconvert\noexpand#1% \expandafter\noconvert\noexpand#2% \expandafter\noconvert\noexpand#3% \else \if\noexpand#2\relax \badutfinput \else \ifnum`#2<128 \badutfinput \else \missingutfchar{\expandafter\noconvert\noexpand#1% \expandafter\noconvert\noexpand#2% \expandafter\noconvert\noexpand#3}% \fi\fi\fi} \def\badutfinput{{\errorcontextlines=-1 \errhelp={Try to set mubytein=0 or input utf8off or input mixcodes.} \errmessage{UTF-8 INPUT IS CORRUPTED ! \space May be you are using another input encoding}}} \def\missingutfchar#1{{\day=0 \ifnum\day=0 % this makes me more robust in message prameter \newlinechar=`^^J \chardef\$=\xprncode255 \xprncodes=1 \message{WARNING: unknown UTF-8 code: `#1 =}% \xprncodes=0 \message{#1' (line: \the\inputlineno)}% \xprncodes=\$ \leavevmode\hbox{\kern.05em\vrule width.5em height.5em\kern.05em}% \else unknown UTF-8 code `#1'\fi}} % \xprncodes=1 or \xprncodes=0 .. set/unset printability of 128..255 chars \def\xprncodes=#1 {% \count255=128 \loop \xprncode\count255=#1\advance\count255 by 1 \ifnum \count255<256 \repeat } \ifx\unknowncodes\relax \endinput \fi % you can \let\unknowncodes\relax % if you don't want to active % the warncodes immediatelly \setunknowncodes \endinput