sort - sorts file[s].
Takes list of file-names for input. But piping and redirection also work.

Sends results to screen. So piping or redirection are required.

Do NOT use same file-name for input and output. Date will be destroyed.

sort ofile > newfile

Some Options :

-b - ignore leading blanks. space/tab

-d - dictionary. Puts all upper case words first.

-f - sorts by letter then case. Complement of -d.

-i - ignore non-printable characters.

-n - treat numbers as real numbers not a string of digits.

 
data
no options
-n

 1
 7
 23
 554
 3
 
 1
 23
 3
 554
 7
 
 1
 3
 7
 23
554




-o filename - output to specific file-name. 
  This is the case where you can use the same name as the input.
  sort data -o data

-c - check but don't sort. Report first bad line.

-C - check but don't sort. If sorted, return code 0. No output.

-k - key, sort on different 'column', Use , to list multiple -k 3,2
  sort -k 2 file
  # if file is columns of data, this will sort on the second column.
  # uses white space as field delimiter.

-t - define alternative field delimiter.
  sort -n -t ":" -k 3 /etc/passwd | less
  # uses : as the field delimiter, sorts on 3rd field, keeps numeric order. 

-u - unique. eliminate duplicates. Ignores case.
  # used with -c, lists first occurrence of duplication. Takes no other action.


uniq - eliminate consecutive duplicate lines.
# use sort -u to eliminate all duplication.

Some Options :

-c - list number of duplications for each line at start of line.

-b - print only duplicate lines.

-u - print only unique lines.



tr - translate. Takes standard input and sends to standard output>
# So redirection and piping required in most cases.

tr takes two lists, the search list and the replacement list. 
It does a one to one match on the lists. So lists should be same length.
It does support ranges.
It does support the regex symbolic ranges.

cat data | tr "12345" "ABCDE"cat data | tr "a-z" "A-Z"cat data | tr ":lower:" ":upper:".

Some options :

-d - delete rather than replace, in which case, only single list given.
cat data | tr -d "\r" > ndata # removes the carriage return for DOS text files. 

-s - squeeze. If multiple concurrent occurrences of a character listed
in the search set found, replace with single character in the replacement
set.

-c - complement. Translate any characters NOT in the search set.
   cat file | tr -c ":alpha:" "-"



ul - underlining. Translates underlines in a document to the 
appropriate character for the terminal being used.

If underling not possible, it will strip it from document.

Useful for cleaning up man-pages for printing.

man man | ul -t dumb > man.txt