# Description
#   This script takes the raw phrase-list data that has been extracted from the
#   Web-Site http://www.intercat.gencat.es/ and attempts to reform it so that it
#   can be used by a language tutoring program such as the one at
#   http://www.geocities.com/matth3wbishop/  This language tutoring program uses an
#   array that is in the following format
#     sound-file-url||textual-equivalent-of-sound-file||translation-of-textual-equivalent||
#   In other words, each element of the array (in this case a client-side Javascript array
#   is a text string which is itself a 'two bar delimited' valued consisting of three (and 
#   possibly more) values. 
#
#   The present script is an attempt to reformat the data that has been extracted into a
#   format which will then be useable by another script to create the final array.
#   
# Notes:
#   This script is quite 'destructive' of the initial data (as obtained with the program
#   html2text from the gencat site. The reason for this is that I wish to eliminate any
#   'bad' data even if this means getting rid of some good data. I this there is enough
#   data to justify this.

# See Also:
# Author:
#   m.j.bishop

 
 if [ "$1" = "" ]
 then
   echo "usage: $0  textFileName"
   cat $0 | sed -n "/^[ ]*#/p" 
   exit 1;
 fi

 sFileName="$1"
 cat $sFileName | \
   sed -e "s/^[[:space:]]*//g" \
       -e "s/[[:space:]]*$//g" \
       -e "/\[URL=/d" \
       -e "/You can listen to the phrases/d" \
       -e "/\-/! h" -e "/\-/{G;s/\-\([^\-]\+\)\-\([^\n]\+\)\n\(.*\)[ ]\{4\}\(.*\)/\3 \1 \4 \2/g;}"  \
       -e "/\-/d"  \
       -e "s/^\(.*\) \{7\}\(.*\)$/\1|\2/g" \
       -e "s/[ ]*|/|/g"   | \
    grep "|" | \
    tr -s ' '