replace "cat -v" with pure bash implementation

1 files changed, 71 insertions, 34 deletions

diff --git a/twopane.bash b/twopane.bash
index bec96b9..39a6bfb 100755
--- a/twopane.bash
+++ b/twopane.bash
@@ -32,6 +32,7 @@ save_file()
 {
         cat > "$TWOPANE"/"${1:?$0: Error: filename cannot be empty string}"
 }
+
 save_screenrc()
 {
         save_file screenrc"${1:+.$1}"
@@ -156,7 +157,7 @@ disconnect()
 {
         if [ $# = 2 ]
         then
-                declare -n std0="$1" std1="$2"
+                declare -i -n std0="$1" std1="$2"
                 eval "exec $std0<&- $std1>&-"
                 unset std0 std1
         fi
@@ -228,12 +229,12 @@ forward()
         # the other pane's terminal.  Uppercase $STDIN/$STDOUT are the
         # real stdin/stdout of this function, connected to the lower
         # pane's terminal.  Socat here merges inputs from both sources.
+        exec {STDIN}<&0 {STDOUT}>&1 {STDERR}>&2
 
         # The input is put out raw back over the socket.  The input is
         # copied to stdout after being filtered (to display control
         # characters with carrot-encoding like '^[' etc).
-        exec {STDIN}<&0 {STDOUT}>&1 {STDERR}>&2
-        exec {TEE}> >(tee >(output_filter | soft_cursor >&$STDOUT) >&$stdout)
+        exec {TEE}> >(tee >(output_filter >&$STDOUT) >&$stdout)
         stty=cfmakeraw,opost=1,onlcr=1
         {
                 socat FD:$STDIN,$stty!!STDOUT - <&$stdin >&$TEE 2>&$STDERR &
@@ -304,34 +305,79 @@ focus()
 
 output_filter()
 {
-        exec {OUTPUT_FILTER}> >(exec cat -v)
-        while read -r -N1
+        tokenize | colorize | soft_cursor
+}
+
+# if (show_nonprinting) {
+#     while (true)
+#       {
+#         if (ch >= 32) {
+#             if (ch < 127) *bpout++ = ch;
+#             else if (ch == 127) {*bpout++ = '^'; *bpout++ = '?';}
+#             else {
+#                 *bpout++ = 'M';
+#                 *bpout++ = '-';
+#                 if (ch >= 128 + 32) {
+#                     if (ch < 128 + 127) *bpout++ = ch - 128;
+#                     else {*bpout++ = '^'; *bpout++ = '?';}}
+#                 else {*bpout++ = '^'; *bpout++ = ch - 128 + 64;}}}
+#         else if (ch == '\t' && !show_tabs) *bpout++ = '\t';
+#         else if (ch == '\n') {newlines = -1; break;}
+#         else {*bpout++ = '^'; *bpout++ = ch + 64;}
+#         ch = *bpin++;}}
+
+chr()
+{
+        declare -i n="$*"
+        printf "\\$(printf %o "$n")"
+}
+
+colorize()
+{
+        while read -r
         do
                 case "$REPLY" in
-                        # Encode literal '^' as escape.  This allows
-                        # a '^' in the output to be interpreted
-                        # unambiguously as a control sequence later
-                        # downstream, where it will be displayed in
-                        # bold.
-                        '^' )
-                                echo -n $'\e'
-                                continue
-                        ;;
-                        $'\t' )
-                                echo -n '^I'
+                        \\[0-7][0-7][0-7] ) ;;
+                        * )
+                                printf '%s\n' "$REPLY"
                                 continue
-                        ;;
-                        $'\n' )
-                                echo -n '^J'
-                                continue
-                        ;;
+                                ;;
                 esac
-                # Add unicode support to cat through bash primitives!
+                declare -i c=8#"${REPLY#?}"
+                if (( c > 128 + 127 ))
+                then
+                        :
+                elif (( c > 128 + 32 ))
+                then
+                        printf -v REPLY "M-$(chr c - 128)"
+                elif (( c > 127 ))
+                then
+                        printf -v REPLY "M-^$(chr c - 128 + 64)"
+                elif (( c < 32 ))
+                then
+                        printf -v REPLY "^$(chr c + 64)"
+                fi
+                printf $'\e[1m%s\e[m\n' "$REPLY"
+        done
+}
+
+tokenize()
+{
+        while read -r -N1
+        do
                 if [[ "$REPLY" =~ [[:print:]] ]]
                 then
-                        printf '%s' "$REPLY"
+                        # Output one printable character per line. It may be a
+                        # multibyte unicode character.
+                        printf '%s\n' "$REPLY"
+                        continue
                 else
-                        printf '%s' "$REPLY" >&${OUTPUT_FILTER}
+                        # If it is a non-printable, then we output a
+                        # multi-character line. In this case we colorize it
+                        # later so that it won't be confused with multiple
+                        # printable characters.
+                        printf '\\%.3o\n' "'$REPLY"
+                        continue
                 fi
         done
 }
@@ -342,16 +388,7 @@ soft_cursor()
         REPLY=
         while printf "$FMT" "$REPLY" $(( 101 + RANDOM % 7 ))
         do
-                read -r -N1 || break
-                case "$REPLY" in
-                        '^' )
-                                read -r -N1 || break
-                                REPLY=$'\e[1m^'$REPLY$'\e[m'
-                        ;;
-                        $'\e' )
-                                REPLY='^'
-                        ;;
-                esac
+                read -r || break
         done
 }