txt2odt.sh

#!/bin/bash
#=============================================================
# Clean up a text file for importing into a LibreOffice Word Processor
#    document (file extension = odt).
# Input: $1 (dirty text file)
# Output: $2 (LibreOffice Writer ODT file)
#
# Version 2 on 2018-06-15 by [email protected]
# Amend and/or distribute as you wish.
#
# Strategy for each file:
# 1. Change all control characters (0x00 through 0x0f) into space characters.
# 2. Delete all leading and trailing whitespace (blanks, tabs).
# 3. For file conversion: libreoffice --convert-to odt:writer8 {text file}
#=============================================================

#--- Definitions
MYNAME=`basename $0`
TEMP_PREFIX=${MYNAME}_$$
TEMP_TXT_FILE=./$TEMP_PREFIX.txt
TEMP_ODT_FILE=./$TEMP_PREFIX.odt

# Process one file (callable procedure)
proc_one_file()
{
	IN=$1
	OUT=$2
	ORIGINAL=$IN.original

	#--- Check readability of input file
	if [ ! -r $IN ]; then 
		echo '*** '$MYNAME': File '$IN' does not exist or is not readable'
		exit 86
	fi

	#--- Check writability of input file
	if [ ! -w $IN ]; then 
		echo '*** '$MYNAME': File '$IN' is not writable'
		exit 86
	fi

	#--- Temp file = input file scrubbed of control characters
	tr -s [:cntrl:] ' ' < $IN > $TEMP_TXT_FILE
	RC=$?
	if [ $RC -ne 0 ]; then 
		echo '*** '$MYNAME': tr using '$IN' failed'
		exit 86
	fi

	# Convert $IN to ODT format
	libreoffice --headless --convert-to odt:writer8 $TEMP_TXT_FILE
	RC=$?
	if [ $RC -ne 0 ]; then 
		echo '*** '$MYNAME': libreoffice conversion failed of '$IN
		exit 86
	fi
	sleep 1

	# Move ODT-version of $IN to $OUT
	cp $TEMP_ODT_FILE $OUT
	RC=$?
	if [ $RC -ne 0 ]; then 
		echo '*** '$MYNAME': mv '$IN' '$OUT' failed'
		exit 86
	fi

	# Delete files no longer needed
	rm $IN
	rm $TEMP_TXT_FILE
	rm $TEMP_ODT_FILE

	# SUCCESS
	echo $MYNAME'('$IN'): done'
}

#--- Begin main script
#--- Check command line
ARGC=$#
if [ $ARGC -ne 2 ]; then
	echo
	echo "Usage:   "$MYNAME"   {Input Text File}   {Output ODT File}"
	echo
	exit 1
fi

# Make sure that the temporary file is writable
touch $TEMP_TXT_FILE
RC=$?
if [ $RC -ne 0 ]; then 
	echo '*** '$MYNAME': touch '$TEMP_TXT_FILE' failed'
	exit 86
fi

# Convert specified text file ($1) file to an ODT file ($2)
proc_one_file $1 $2