Bashscripts.org

<TD align="middle" valign="middle"><font face="Arial" size="3">100

sed 's/.*size="3">\(.\).*/\1/' report_file.htm

cat report_file.htm \
| sed "s/<\/TD>/~/g" \                       
| sed "s/size=\"3\">"/~&/g" \               
| tr "~" "\n" \                                     
| grep "size=\"3\">" \                           
| sed -e :a -e 's/<[^>]*>//g;//ba' \
| sed '1,18d' \
| cut -d">" -f2 \
| sed 's/[^[:digit:]]//g' \
| sed '/./!d' \
| sed -e :a -e '{N; s/\n/\t/g; ta}'         

# sed #1 - replacing all of the closing cell tags (</TD>) with ~
# sed #2 - replacing all of the size="3"> finds with ~
# tr - making every instance of ~ have it's own line
# grep - showing only those lines with size="3"> (the TD tag was screwing me up since multiple size="3"> were in some of those lines...hence the reason for sed #1
# sed #3 - removing the rest of the html tags
# sed #4 - removing the 1st 18 lines as they had a bunch of crap in them irrelevant to what I was needing
# cut - changing the delimiter to > and printing the 2nd field
# sed #5 - removing all remaining non-numerical junk
# sed #6 - removing the empty lines left over from sed #5
# sed #7 - changing the vertical output to horizontal and adding a tab between each entry