List the words from a handful of HTML files by frequency.

wordfrequency.sh

cat *.html | sed -e 's/<[^>]\+>//g' -e 's/[ \t]\+/\n/g' -e 's/[-\.\\(\\):,;0-9+|]//g'|sort | uniq -ci | sort -h

cat *.html | \            # all the files' contents
sed -e 's/<[^>]\+>//g' \  # without tags (assumes they don't contain line breaks)
-e 's/[ \t]\+/\n/g' \     # replace tabs and spaces with line breaks
-e 's/[^a-z]//gi' | \     # remove some non-letters (
sort | \                  # sort once to make uniq work
uniq -ci | \              # show occurrence counts, case insensitive
sort -h                   # sort by numbers