Using unix/linux to do weblog Homework #2

 1) top 10 requestors
  cut -f1 -d- w4000log           pulls off first field with - delimiter
  *above * | sort | uniq -c      sort, count duplicates
  *all above* | sort | tail      sort counts, bottom has biggest counts

 2) day*hour tables -- not as easy, but can get
   same info but not as 24*7 table but list of 168 numbers
  a) successful requests
    grep ' 200 ' w4000log         pulls successful requests only
    cut -f2 -d[ w4000log          pulls off time stamp info using '['
                                  as delimiter and taking second field

    grep ' 200 ' w4000log | cut -f2 -d[ | cut -c1-2,12-14
                                 pipe together, then take off day digits
                                  and hour digits (and ':')
    *all above* | uniq -c        find and count unique records
    grep*cut*cut | sort | uniq -c > q1a
                                 pipe together, sort by day, hour and
                                  give count for each day, hour

  b) from top 10 requestors -- I don't have a clue

 Using unix/linux to do weblog Homework #2

 3) number of requests for each faculty member

  cut -f7 --delimiter=' ' h4short   use space as delimiter (not tabs) and
                                     pull 7th field
 
  sed 's/%7E|e/~/g'                 change all '%7E' or '%7e' to '~'
  *above* | cut -c3-30              take cols 3-30 from source 
  *above* | cut -f1 -d/             take off everything after /
  *above* | sort                    sort alphabetically
  *above* | uniq -c | sort g > tail count duplicates, general sort in 
                                     last ones are biggest