Personal tools

Simple Unix tools

From HaskellWiki

(Difference between revisions)
Jump to: navigation, search
(wibble)
m (Reverted edits by LindsayMcphee (Talk); changed back to last version by Aaronmcdaid)
(30 intermediate revisions by 10 users not shown)
Line 1: Line 1:
Simple unix tools written in Haskell.  
+
Simple Unix commandline tools written in Haskell.  
  
This is intended as a beginners tutorial for learning Haskell from a  
+
This is intended as a beginner's tutorial for learning Haskell from a
"Lets just solve things already!" point of view. The examples should
+
"Let's just solve things already!" point of view. The examples should
help give a flavour of the beauty and expressiveness of Haskell
+
help give a flavor of the beauty and expressiveness of Haskell
 
programming.
 
programming.
 +
  
 
<haskell>
 
<haskell>
--
 
-- Some unix-like tools written in elegant Haskell
 
--
 
  
 +
import Control.Monad.Instances
 
import Data.List
 
import Data.List
 
import Data.Char
 
import Data.Char
import System.IO
+
import Data.Maybe
 
import Text.Printf
 
import Text.Printf
 +
import System.Environment
  
--
+
-- First, two helpers
-- First, a useful helper
+
io f = interact (unlines . f . lines)
--
+
input f = interact (unlines . f . lines)
+
  
--
+
showln  = (++ "\n") . show
-- The 'cat' program
+
--
+
cat    = input id
+
  
--
+
-- remove duplicate lines from a file (like uniq)
-- Sort a file
+
uniq    = nub  -- Warning: Unix uniq discards *consecutive* dupes. But 'nub' discards all dupes.
--
+
sort'   = input sort
+
  
--  
+
-- repeat the input file infinitely
-- Reverse a file (tac)
+
rpt     = cycle
--
+
tac     = input reverse
+
  
--
 
 
-- Return the head -10 line of a file
 
-- Return the head -10 line of a file
--
+
take'  = take 10
head'  = input $ take 10
+
  
--
 
 
-- Remove the first 10 lines of a file
 
-- Remove the first 10 lines of a file
--
+
drop'  = drop 10
drop'  = input $ drop 10
+
 
 +
-- Return the head -1 line of a file
 +
head'  = head
  
--
 
 
-- Return the tail -1 line of a file
 
-- Return the tail -1 line of a file
--
+
tail'  = last
tail'  = input $ (:[]) . last
+
  
--
+
-- return the last ten lines of a file
-- remove duplicate lines from a file (like uniq)
+
tail10  = drop =<< subtract 10 . length
--
+
 
uniq   = input nub
+
-- Reverse lines in a file (tac)
 +
tac    = reverse
 +
 
 +
-- Reverse characters on each line (rev)
 +
rev    = map reverse
 +
 
 +
-- Reverse words on each line
 +
rev_w  = map (unwords . reverse . words)
 +
 
 +
-- Count number of characters in a file (like wc -c)
 +
wc_c    = showln . length
 +
 
 +
-- Count number of lines in a file, like wc -l
 +
wc_l    = showln . length . lines
 +
 
 +
-- Count number of words in a file (like wc -w)
 +
wc_w   = showln . length . words
  
--
 
 
-- double space a file
 
-- double space a file
--
+
space  = intersperse ""
space  = input $ intersperse ""
+
  
--  
+
-- undo double space
-- repeat the input file infinitely
+
unspace = filter (not.null)
--
+
rpt    = interact cycle
+
  
--
+
-- remove the first occurrence of the line "str"
-- remove the first occurence of the line "str"
+
remove  = delete
--
+
remove  = input $ delete "str"
+
  
--
+
-- make a string all upper case
-- make a file all upper case
+
upper  = map toUpper
--
+
upper  = interact $ map toUpper
+
  
--
 
 
-- remove leading space from each line
 
-- remove leading space from each line
--
+
clean  = map (dropWhile isSpace)
clean  = input $ map (dropWhile isSpace)
+
  
--
+
-- remove trailing whitespace
-- join lines of a file
+
clean'  = map (reverse . dropWhile isSpace . reverse)
--
+
join    = input $ (:[]) . concat
+
  
--
+
-- delete leading and trailing whitespace
-- Translate the letter 'e' to '*', like tr 'e' '*' (or y// in sed)
+
clean'' = map (f . f)
--
+
     where f = reverse . dropWhile isSpace
y      = interact $ map f
+
     where f 'e' = '*'
+
          f  c  = c
+
--
+
-- Filter the letter 'e' from a file, like tr -d 'e'
+
--
+
tr      = interact $ filter (/= 'e')
+
  
--
+
-- insert blank space at beginning of each line
-- Count number of characters in a file (like wc -c)
+
blank  = map (s ++)
--
+
    where s = replicate 8 ' '
wc_c    = interact $ show . length
+
  
--
+
-- join lines of a file
-- Count number of lines in a file, like wc -l
+
join = return . concat
--
+
wc_l    = interact $ show . length . lines
+
  
--
+
-- Translate the letter 'e' to '*', like tr 'e' '*' (or y// in sed)
-- Count number of words in a file (like wc -w)
+
tr a b = interact (map f)
--
+
    where f c = if c == a then b else c
wc_w    = interact $ show . length . words
+
 
 +
-- Delete characters from a string.
 +
tr_d a = tr a ' '
  
--
 
 
-- grep lines matching "^foo" from a file
 
-- grep lines matching "^foo" from a file
--
+
grep = filter (isPrefixOf "foo")
grep   = input $ filter (isPrefixOf "foo")
+
  
--
 
 
-- grep lines that don't match "^foo" (grep -v)
 
-- grep lines that don't match "^foo" (grep -v)
--
+
grep_v  = filter (not . isPrefixOf "foo")
grep_v  = input $ filter (not . isPrefixOf "foo")
+
  
--
 
 
-- number each line of a file
 
-- number each line of a file
--
+
num = zipWith (printf "%3d %s") [(1::Int)..]
num     = input $ zipWith (printf "%3d %s") [1::Int..]
+
  
--
+
-- Compute a simple cksum of a file
-- Compute a simple hash of a file
+
cksum  = foldl' k 5381
--
+
  where k h c = h * 33 + ord c
cksum  = interact $ printf "%u\n" . foldl' k 5381
+
 
    where k h c = h * 33 + fromIntegral (ord c) :: Int
+
-- And our main wrapper
 +
main = do
 +
    who <- getProgName
 +
    maybe (return ()) id $ lookup who $
 +
        [("blank",      io blank                  )
 +
        ,("cksum",      interact (showln . cksum) )
 +
        ,("clean",      io clean''                )
 +
        ,("echo" ,      interact id              )
 +
        ,("drop",        interact drop'            )
 +
        ,("grep",        io grep                  )
 +
        ,("grep -v",    io grep_v                )
 +
        ,("head",        io (return . head')      )
 +
        ,("join",        io join                  )
 +
        ,("num",        io num                    )
 +
        ,("remove",      io (remove "str")        )
 +
        ,("revw",        io rev_w                  )
 +
        ,("reverse",    io rev                    )
 +
        ,("reverseword", io rev_w                  )
 +
        ,("rpt",        io rpt                    )
 +
        ,("sort",        interact sort            )
 +
        ,("space",      io space                  )
 +
        ,("tac",        interact tac              )
 +
        ,("take",        io take'                  )
 +
        ,("tail",        io (return . tail')      )
 +
    --  ,( "tr"  ,    interact tr)
 +
    --  ,( "tr -d",  interact (tr_d . unwords))
 +
        ,("unspace",    io unspace                )
 +
        ,("upper",      interact upper            )
 +
        ,("uniq",        interact uniq            )
 +
        ,("wc_c",        interact wc_c            )
 +
        ,("wc_l",        interact wc_l            )
 +
        ,("wc_w",        interact wc_w            )
 +
        ]
 
</haskell>
 
</haskell>
  
 +
==How to run==
 +
These functions can be executed as one liners from a shell. For example,
 +
to use the Haskell version of 'wc':
 +
 +
    $ cat file.txt | ghc -e 'wc_l' UnixTools.hs
 +
 +
Or, one could define 'main' to be a chosen tool/function (add a line to
 +
the effect that "main = wc_l") and then compile the tool with
 +
 +
    $ ghc --make UnixTools.hs
 +
 +
The given Haskell codes presents yet a third way of doing things: much
 +
like the [http://en.wikipedia.org/wiki/BusyBox BusyBox] suite of Unix
 +
tools, it is possible to compile a single monolithic binary and have it
 +
detect what name it is run by and then act appropriately. This is the
 +
approach the following code takes: you can compile it and then make
 +
symbolic links (like <code>"ln -s UnixTools echo; ln -s UnixTools cat"
 +
</code>) and then run those commands (<code>"./echo foo | ./cat"</code>
 +
would produce output of "foo").
 +
 +
 +
 +
==Where to now?==
 +
* [[Haskell|Haskell.org]]
 +
* The Haskell standard [http://www.cse.unsw.edu.au/~dons/data/List.html list library], with [http://haskell.org/ghc/docs/latest/html/libraries/base/Data-List.html docs]
 +
* Alternative [[Wc|implementations]] of the wc program
 +
* Learn how to [[Introduction to QuickCheck|test Haskell code]]
 +
* [[Example code|More]] Haskell code
 +
* Haskell for [[Applications and libraries/Operating system#Shell|shell scripting]]
 +
* Export list functions to the shell with [http://www.cse.unsw.edu.au/~dons/h4sh.html h4sh]
 +
* [[Checking for correct invocation of a command line haskell program]]
 +
* [[Poor man's here document]]
 +
* [http://andrew.bromage.org/darcs/diff Diff in 120 Lines of Haskell]
 
[[Category:Tutorials]]
 
[[Category:Tutorials]]
 +
[[Category:Code]]

Revision as of 15:06, 5 November 2011

Simple Unix commandline tools written in Haskell.

This is intended as a beginner's tutorial for learning Haskell from a "Let's just solve things already!" point of view. The examples should help give a flavor of the beauty and expressiveness of Haskell programming.


import Control.Monad.Instances
import Data.List
import Data.Char
import Data.Maybe
import Text.Printf
import System.Environment
 
-- First, two helpers
io f = interact (unlines . f . lines)
 
showln  = (++ "\n") . show
 
-- remove duplicate lines from a file (like uniq)
uniq    = nub   -- Warning: Unix uniq discards *consecutive* dupes. But 'nub' discards all dupes.
 
-- repeat the input file infinitely
rpt     = cycle
 
-- Return the head -10 line of a file
take'   = take 10
 
-- Remove the first 10 lines of a file
drop'   = drop 10
 
-- Return the head -1 line of a file
head'   = head
 
-- Return the tail -1 line of a file
tail'   = last
 
-- return the last ten lines of a file
tail10  = drop =<< subtract 10 . length
 
-- Reverse lines in a file (tac)
tac     = reverse
 
-- Reverse characters on each line (rev)
rev     = map reverse
 
-- Reverse words on each line
rev_w   = map (unwords . reverse . words)
 
-- Count number of characters in a file (like wc -c)
wc_c    = showln . length
 
-- Count number of lines in a file, like wc -l
wc_l    = showln . length . lines
 
-- Count number of words in a file (like wc -w)
wc_w    = showln . length . words
 
-- double space a file
space   = intersperse ""
 
-- undo double space
unspace = filter (not.null)
 
-- remove the first occurrence of the line "str"
remove  = delete
 
-- make a string all upper case
upper   = map toUpper
 
-- remove leading space from each line
clean   = map (dropWhile isSpace)
 
-- remove trailing whitespace
clean'  = map (reverse . dropWhile isSpace . reverse)
 
-- delete leading and trailing whitespace
clean'' = map (f . f)
    where f = reverse . dropWhile isSpace
 
-- insert blank space at beginning of each line
blank   = map (s ++)
     where s = replicate 8 ' '
 
-- join lines of a file
join = return . concat
 
-- Translate the letter 'e' to '*', like tr 'e' '*' (or y// in sed)
tr a b = interact (map f)
    where f c = if c == a then b else c
 
-- Delete characters from a string.
tr_d a = tr a ' '
 
-- grep lines matching "^foo" from a file
grep = filter (isPrefixOf "foo")
 
-- grep lines that don't match "^foo" (grep -v)
grep_v  = filter (not . isPrefixOf "foo")
 
-- number each line of a file
num  = zipWith (printf "%3d %s") [(1::Int)..]
 
-- Compute a simple cksum of a file
cksum   =  foldl' k 5381
   where k h c = h * 33 + ord c
 
-- And our main wrapper
main = do
    who <- getProgName
    maybe (return ()) id $ lookup who $
        [("blank",       io blank                  )
        ,("cksum",       interact (showln . cksum) )
        ,("clean",       io clean''                )
        ,("echo" ,       interact id               )
        ,("drop",        interact drop'            )
        ,("grep",        io grep                   )
        ,("grep -v",     io grep_v                 )
        ,("head",        io (return . head')       )
        ,("join",        io join                   )
        ,("num",         io num                    )
        ,("remove",      io (remove "str")         )
        ,("revw",        io rev_w                  )
        ,("reverse",     io rev                    )
        ,("reverseword", io rev_w                  )
        ,("rpt",         io rpt                    )
        ,("sort",        interact sort             )
        ,("space",       io space                  )
        ,("tac",         interact tac              )
        ,("take",        io take'                  )
        ,("tail",        io (return . tail')       )
    --  ,( "tr"  ,    interact tr)
    --  ,( "tr -d",   interact (tr_d . unwords))
        ,("unspace",     io unspace                )
        ,("upper",       interact upper            )
        ,("uniq",        interact uniq             )
        ,("wc_c",        interact wc_c             )
        ,("wc_l",        interact wc_l             )
        ,("wc_w",        interact wc_w             )
        ]

1 How to run

These functions can be executed as one liners from a shell. For example, to use the Haskell version of 'wc':

   $ cat file.txt | ghc -e 'wc_l' UnixTools.hs

Or, one could define 'main' to be a chosen tool/function (add a line to the effect that "main = wc_l") and then compile the tool with

   $ ghc --make UnixTools.hs

The given Haskell codes presents yet a third way of doing things: much like the BusyBox suite of Unix tools, it is possible to compile a single monolithic binary and have it detect what name it is run by and then act appropriately. This is the approach the following code takes: you can compile it and then make symbolic links (like "ln -s UnixTools echo; ln -s UnixTools cat" ) and then run those commands ("./echo foo | ./cat" would produce output of "foo").


2 Where to now?