Shell Programming and Scripting

this obviously isnt overly important, but its a bit slow and i was wondering if i could prevent it from re-searching the file several times. i want my matches from the previous and coming three days, as well as today. this is what i have but its weighty and im having trouble finding a way to slim it down.

cd ~/work
grep -iT `date --date="3 days ago" "+%b%d"` les
grep -iT `date --date="2 days ago" "+%b%d"` les
grep -iT `date --date="1 days ago" "+%b%d"` les
grep -iT `date "+%b%d"` les
grep -iT `date --date="next day" "+%b%d"` les
grep -iT `date --date="2 days" "+%b%d"` les
grep -iT `date --date="3 days" "+%b%d"` les

just need to spit out the lines starting with mmmdd, and im hoping to save some space/time.
thanks
nomkev

show sample of your file

thats it. its just a function... i skipped the {} to save space.

edit: oops. the file. well it has names, but its basically a txt with:

may24 name name
may25 name name
may26 name name
etc...

edit again: single digits are in the 'may01' etc

This following should be quicker...

Code:

gawk '
   BEGIN {
     for (i=-3;i<=3;i++)
     {
        cmd="date --date=\"" i "days\" \"+%b%d\""
        cmd | getline
        close(cmd)
        dy[$0]++
     }
   }
   { for (s in dy) if ($0 ~ "^" s "[^0-9]") print }' les

im getting no output...
changed gawk to mawk (im using ubuntu) but no errors either.

I didn't spot that you were ignoring case in your grep commands. Changes in bold red:

Code:

gawk '
   BEGIN {
     for (i=-3;i<=3;i++)
     {
        cmd="date --date=\"" i "days\" \"+%b%d\""
        cmd | getline
        close(cmd)
        dy[tolower($0)]++
     }
   }
   { for (s in dy) if ($0 ~ "^" tolower(s) "[^0-9]") print }' les

beautiful, and much faster.
thanks alot!