makecorpus.sh

  do B=`basename $F .utf-8`

     G=$D/corpus/$B.mc.utf-8

     if [ ! -s $G ]

     then sed -e 's/［/\[/g' -e 's/］/\]/g' -e 's/\[[^]]*\]//g' -e 's/（/(/g' -e 's/）/)/g' -e 's/([^)]*)//g' `ls -1t incoming/*/[th]*/$B.utf-8 | head -1` | tr '() \133\135' '\012\012\012\012\012' > $G

     then sed -e 's/ / /g' -e 's/［/\[/g' -e 's/］/\]/g' -e 's/\[[^]]*\]//g' -e 's/（/(/g' -e 's/）/)/g' -e 's/([^)]*)//g' `ls -1t incoming/*/[th]*/$B.utf-8 | head -1` | tr '() \133\135' '\012\012\012\012\012' > $G

     elif [ $D/hakubun/$B.utf-8 -nt $G ]

     then ( sed -e 's/［/\[/g' -e 's/］/\]/g' -e 's/\[[^]]*\]//g' -e 's/（/(/g' -e 's/）/)/g' -e 's/([^)]*)//g' $D/hakubun/$B.utf-8 | tr '() \133\135' '\012\012\012\012\012' ; echo '' ) | tr -s '\012' '\012' > $G.a~

     then ( sed -e 's/ / /g' -e 's/［/\[/g' -e 's/］/\]/g' -e 's/\[[^]]*\]//g' -e 's/（/(/g' -e 's/）/)/g' -e 's/([^)]*)//g' $D/hakubun/$B.utf-8 | tr '() \133\135' '\012\012\012\012\012' ; echo '' ) | tr -s '\012' '\012' > $G.a~

          sed -n -e '/	/{' -e :loop -e N -e 's/	.*\n//' -e 's/EOS//' -e '/	/bloop' -e p -e '}' $G | tr -s '\012' '\012' > $G.b~

          X=`diff $G.a~ $G.b~ | sed -n -e 's/^\([0-9]*\).*$/\1/p' -e 1q`

          if [ 0$X -gt 0 ]