#! /bin/sh
# $Id: cvs_Lu.sh,v 1.11 2020/05/14 12:59:06 hako Exp $
#
# cvs_Lu.sh is a filter for CVS.
#
# Convert the line break of *.csv files to LF.
# Preserve the UTF-8 BOM.
# Delete the UTF-8 BOM from ASCII csv files.
#
# by Hiroshi Hakoyama
#
# Background:
# CVS needs LF, but Excel provides CRLF for *.csv files.
# Excel needs the UTF-8 BOM to read the UTF-8 encoding and provides the BOM in a UTF-8 csv. 
# The BOM does not cause trouble for the read.csv() function in R.
# Excel can read a csv file with LF.
# Therefore, before commit UTF-8 csv files to CVS server, we should convert the line breaks to LF and preserve the UTF-8 BOM.
# 
# Solution:
# A CVS filter to change line breaks to LF for *.csv files.
#
# Usage:
# Add an alias to tcsh
# alias cvs 'cvs_Lu.sh'

if [ "$1" = "commit" -o "$1" = "import" ]; then
find . -type f -exec nkf -g {} + | grep -e UTF-8 | grep -e .csv | sed -e 's/: UTF-8//g' | tr '\n' '\0' | xargs -0 nkf --oc=UTF-8-BOM -Lu --in-place
find . -type f -exec nkf -g {} + | grep -e ASCII | grep -e .csv | sed -e 's/: ASCII//g' | tr '\n' '\0' | xargs -0 nkf -Lu --in-place
find . -type f -exec nkf -g {} + | grep -e Shift_JIS | grep -e .csv | sed -e 's/: Shift_JIS//g' | tr '\n' '\0' | xargs -0 nkf --oc=Shift_JIS -Lu --in-place
 cvs "$@"
else
 cvs "$@"
fi

exit 0