Week 10, Intro to Unix lab 2016

Content

• awk
  • BEGIN, END, ‘re {}
  • built-in variables IFS, OFS, NR, RS, FS, FILENAME
• nicely formatted lines with awk: seq 10 | awk '/.*/ { printf "%4d\n", $1 ; }'
• print histogram: seq 10 | awk '/.*/ { for(i=0; i< $1; i++) { printf "=" ; } printf "\n" }
• correctly matching brackets - todo rewrite example
• computing variance if we have each number on one row
• basename, dirname

Homework

• (0 point) - Recap - test command
• (1 point) Implement in shell basename without options
  • submit it in format basename_YOURNICKNAME.sh
  • Example usage. In this case basename output and $USER are identical
    • cd ~ ; echo $(basename pwd/) $USER
    • Note that the original basename discards trailing / for directories
• (2 point) Use awk to distribute space evenly in a file.
  • Distribute spaces evenly for each line
  • Do not change words on lines
  • Example input

input file
longer line with more words

      input      file      
longer line with more words

Bonus

• (2 point) - Write a script ancestor_YOURNICKNAME.sh PID1 PID2 [PID] ...
  • which takes at least two arguments PID1 and PID2
    • which must be numbers and must be running process
    • If argument is not number write an error message and exit with 1
    • If process with PID of an argument is not running write error message and exit with 2
  • and prints a numerically sorted list of pairs of PIDs together with closest common ancestor process pid
    • format per row:
      • PID_A PID_B PID_COMMON_A_B
    • Example output where processes has very distant ancestor init process
      • 1233 1234 1