Common Unix Concepts and Commands, part 2
-----------------------------------------

Regular expressions
  - powerful pattern-matching sublanguage
     - akin to filename patterns, but more powerful
  - used by a number of commands, e.g. grep, sed; emacs uses a version
  - basic regexp rules:
     <char>	- matches exactly itself, unless special:
     .		- matches any one character
     [...]	- matches any one character in set
     [^...]	- matches any one character *not* in set
     ^		- matches beginning of line
     $		- matches end of line
     \<char>	- matches <char>, even if special
     <re1><re2>	- sequence: matches <re1> followed by <re2>
     <re1>\|<re2> - alternation: matches <re1> or <re2>
     <re>*	- iteration: matches 0 or more occurrences of <re>
     \(<re>\)	- grouping: matches <re>
  - extended regexps allow more things; see man pages
  - vs. filename patterns:
     ? in filename pattern => . in regexp
     * in filename pattern => .* in regexp
     {<a>,<b>} in filename pattern => \(<a>\|<b>\) in regexp


Grep: command to search lines in files that match regular expressions
  grep <re> <file>...	- print out lines of <file>... that match <re>
  grep -v <re> <file>...- print out lines of <file>... that *don't* match <re>


Sed: command to edit lines in files, e.g. that match regular expressions
  sed -e <command> <file>...	- execute sed command <command> on <file>s
  many possible <command>s; one important one:
    s/<re>/<replacement>/g	- replace all occurrences matching <re>
				  with <replacement>