3 # read files for fulltext index
4 # filelist is on stdin (e.g. by find)
5 # index entries go to stdout
8 # find /foo -name \*.html | ./fulltext >/tmp/idx 2>/tmp/mst
9 # find /usr/share/doc -type f -a \! -name \*.htm\* | ./fulltext >/tmp/idx 2>/tmp/mst
10 # sort -o /tmp/idx /tmp/idx
11 # time ./openisis -write db/test/ft -stream -fmt mfn </tmp/mst
12 # time ./openisis -db db/test/ft -ifload 0 -v i </tmp/idx
13 # time ./openisis -db db/test/ft -ifchk -v i
14 # time ./openisis -db db/test/ft -search Descriptive -ifdump
20 next unless open( F, $f );
22 print STDERR "100\t$f\n";
26 last if 255 < ++$line;
29 for $word (split /\W+/) {
31 printf "%s\t%d\t%d\t%d\t%d\n", uc($word), $fn, 800, $line, ++$w;