Three Four Easy Scripts

1. Mere aggregation:
   Collecting output from separate utilities
   disk_info shell script (source) (OUTPUT)

   1. simple commands

      • can be on same line, separated by “;”

   2. for loop

      • loop variable

      • simple list of values for loop variable

   3. variable substitution: $d

      • shell replaces a placeholder with a value or result

   4. command substitution

      • for p in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
        

      • `seq 1 15` or $(seq 1 15)

      • utility program "seq" provides list values 1 to 15
        

      • provides a value for variable $lbl

      • often used to provide command arguments

2. Simplify an interface
   ps2booklet shell script (source)  (OUTPUT)

   1. problem: creating a booklet document (“imposition”)

   2. psutils package provides tools

      • psbook — re-order pages

• pstops — arbitrary page placement

1. commands are very simple

• psbook <input file> | pstops -p<paper size> <layout>

1. syntax of <layout> is very difficult to work with

• '2:0L(8.5in,0)+1L(8.5in,5.5in)' (DIAGRAM)

1. simple if / then

   • test, aka [

2. case

3. installation: provide several names for one script

3. Custom utility program
   tok(ens) Perl script (source)

   1. Example: extract image file names from an HTML file:

      • $ egrep '<img[^>]+src="([^"]+)"' EM-0029-01.html  (OUTPUT)

      • $ egrep -o '<img[^>]+src="([^"]+)"' EM-0029-01.html (OUTPUT)
        <img src="/Images/topskip.gif" ...

      • $ tok '<img[^>]+src="([^"]+)"' EM-0029-01.html
/Images/topskip.gif

   2. “Shebang” tells system how to execute this file (which program will interpret the text)

   3. Advantages:

      • More powerful and convenient language syntax

      • Better performance

   4. Disadvantages:

      • Less portable (requires interpreter and libraries installed)

4. Output munging
   dur Perl script (source) (OUTPUT)  (DIAGRAM)

   1. sndfile-info output is too verbose and not in convenient format.

   2. Quoting filenames
   3. Starting the child process
   4. Scanning the output
   5. Printing the results