#!/bin/bash
## use less to show unprintable-chars - Eugene Reimer 2001-Oct
## i was happy with:  LESSCHARSET=iso8859 (aka latin1)
## until i encountered a file with blanks that werent spaces!!
## also: files containing CR's can be confusing!!
## So I made this <list> script to provide an unambiguous view for use on such troublesome files.
##
## The following (from <man less>) is a start for constructing a custom charset:
##
## LESSCHARSET	equivalentLESSCHARDEF (where .==printable b==binary c==control)
## -----------	---------------------
## ascii	8bcccbcc18b95.b
## dos		8bcccbcc12bc5b95.b.
## ebcdic	5bc6bcc7bcc41b.9b7.9b5.b..8b6.10b6.b9.7b9.8b8.17b3.3b9.7b9.8b8.6b10.b.b.b.
## iso8859	8bcccbcc18b95.33b.
## koi8-r	8bcccbcc18b95.b128.
## latin1	8bcccbcc18b95.33b.
## next		8bcccbcc18b95.bb125.bb

unset LESSCHARSET
##export LESSCHARDEF="8bcccbcc18b95.b32bb94.."		##ER: very slightly modified ISO-8859-1, treating NBSP as binary
##export LESSCHARDEF="32b95.b"				##ER: my simpler pattern for ASCII - used until 2007jan
export LESSCHARDEF="32b95.b68b.17b..4b.2b.4b.17b.5b.3b"	##ER: 2007jan: ASCII or Plautdietsch-chars or Times-symbol are printable
export LESSBINFMT="*s<%x>"
##echo "list: LESSCHARDEF=$LESSCHARDEF"			##debug

less -dfQU "$@"
	## use -c  for less to redraw from the top.
	## use -d  to prevent msgs about "dumb terminal"
	## use -f  to prevent the "may be binary" prompt.
	## use -Q  to prevent bell-ringing
	## use -U  to treat TAB and CR as binary rather than control chars

exit
2007jan:  want highlighting of chars other than "ASCII plus the following"
	Auml	c4 = 196
	Ouml	d6 = 214
	times	d7 = 215
	Uuml	dc = 220
	szlig	df = 223
	auml	e4 = 228
	ouml	f6 = 246
	uuml	fc = 252

2009-06: while working on OCRing Eldo's Plautdiesch Verbs, had 2nd thoughts about using latin1 for general less usage (eg l,m cmds);
	try switching to the Microsoft superset of latin1;
	==neither cp1252 nor windows-1252 nor windows for LESSCHARSET is legal with my version of less  <--appear to be for later versions==??==
	==SEE /etc/profile-er;  tried using custom charset via LESSCHARDEF -- but hex-80..9f chars are still disappearing:-(
	==NOTE: Konsole in Encoding:cp1252 may be the answer--??--
	==google for  LC_CTYPE  OR  locale  1252
	==for JED:  google for  site:jedsoft.org charset  OR  site:jedsoft.org encoding
	==SEE my solution described in  http://ereimer.net/programs/charsets-cp1252-utf8.htm