Exercise: Palindromes and Isograms

[wtd]

Write a program which reads in a line of input, splits it up into words, then outputs (in order) words that are either palindromes or isograms.

Please note that words containing spaces may be input by surrounding them with quotes. Quotes may be input themselves by escaping them with a backslash.

Again, please identify the language you choose to use.

[Saad]

Haven't coded in haskell for a while but I decided to try it anyway

Haskell:

import Char separateWords :: String -> [String] separateWords line@(x:xs) | x == '\"' = separateWords_ "" [] xs True                           | otherwise = separateWords_ "" [] xs False separateWords_ :: String -> [String] -> String -> Bool -> [String] separateWords_ currentWord allWords (x:xs) True | x == '\"' = separateWords_ "" (allWords ++ [currentWord]) xs False                                                 | otherwise = separateWords_ (currentWord ++ [x]) allWords xs True separateWords_ "" allWords [] _ = allWords separateWords_ "" allWords (x:xs) False | x == '\"' = separateWords_ "" (allWords) xs True                                         | x == ' ' = separateWords_ "" (allWords) xs False                                         separateWords_ currentWord allWords (x:xs) False| x == '\"' = separateWords_ "" (allWords ++ [currentWord]) xs True                                                 | x == ' ' = separateWords_ "" (allWords ++ [currentWord])  xs False                                                 | otherwise = separateWords_ (currentWord ++ [x]) allWords xs False separateWords_ currentWord allWords [] _ = allWords ++ [currentWord] isPalindrome :: String -> Bool isPalindrome string = word == reverse word                 where                     word = filter isAlpha string isIsograph :: String -> Bool isIsograph string = not $ any (\letters -> length letters /= 1) $ group $ sort string                     main = getLine >>= \input -> print $ filter (\word -> isPalindrome word || isIsograph word) $ separateWords $ map (toLower) input

[md]

For those of us who are lazy (me), what's an isogram. And does it have to be a palindrome of real words?

[rizzix]

Here's mine: < see code below >

Edit: syntax tags are getting to be really annoying

[rizzix]

md @ Fri May 09, 2008 5:40 pm wrote:

For those of us who are lazy (me), what's an isogram. And does it have to be a palindrome of real words?

An isogram is-a palindrome, so just check for palindromes

[Saad]

As rizzix pointed out before, I did have errors in my code. I make a couple of errors that I should have caught (I did test it just not thoroughly enough). Error is now fixed what I thought were the errors.

code:

import Char import List separateWords :: String -> [String] separateWords line@(x:xs) | x == '\"' = separateWords_ "" [] xs True                           | otherwise = separateWords_ [x] [] xs False     where         separateWords_ :: String -> [String] -> String -> Bool -> [String]         separateWords_ currentWord allWords (x:xs) True | x == '\"' = separateWords_ "" (allWords ++ [currentWord]) xs False                                                         | otherwise = separateWords_ (currentWord ++ [x]) allWords xs True                                                                 separateWords_ "" allWords [] _ = allWords         separateWords_ "" allWords (x:xs) False | x == '\"' = separateWords_ "" (allWords) xs True                                                 | x == ' ' = separateWords_ "" (allWords) xs False                                                         separateWords_ currentWord allWords (x:xs) False| x == '\"' = separateWords_ "" (allWords ++ [currentWord]) xs True                                                         | x == ' ' = separateWords_ "" (allWords ++ [currentWord])  xs False                                                         | otherwise = separateWords_ (currentWord ++ [x]) allWords xs False         separateWords_ currentWord allWords [] _ = allWords ++ [currentWord] isPalindrome :: String -> Bool isPalindrome string = word == reverse word                 where                     word = map toLower $ filter isAlpha string isIsograph :: String -> Bool isIsograph string = all (\letters -> length letters == 1) $ group $ sort string                     main = getLine >>= \input -> mapM_ (print) $ filter (\word -> isPalindrome word || isIsograph word) $ separateWords input

And a sample output

code:

>ghc a.hs >Exit code: 0 >main Race qux foo foof Saad sad "race car" "cars" "Race" "qux" "foof" "sad" "race car" "cars" >Exit code: 0

rizzix @ Fri May 09, 2008 8:32 pm wrote:

md @ Fri May 09, 2008 5:40 pm wrote:

For those of us who are lazy (me), what's an isogram. And does it have to be a palindrome of real words?

An isogram is-a palindrome, so just check for palindromes

From the definition from wikipedia

An isogram (also known as a "nonpattern word") is a logological term for a word or phrase without a repeating letter. Link here

[richcash]

Well, on wikipedia, one of the defintions for isograms says a word with each letter appearing exactly once. Only under the latter definition that isograms are words where each letter appears the same number of times would palindromes be a subset of isograms.

So there might be some ambiguity here unless one of those definitions is implied for CS problems.

Edit : ^Saad beat me to it.

[rizzix]

Oh interesting... I thought it was `just` single letter words like aaa, bbb, etc

Saad: Shouldn't you account for single-quoted strings?

wtd: way to go, for making it so ambiguous

[wtd]

In this case, it's defined as a word with no repeating letters.

And no, you don't have to check to make sure the "words" are valid.

[rizzix]

OK, so what about strings like: 'abc def ghi'

The space character appears twice. Would that be considered an isogram?

[wtd]

No, it does not count. That would be an isogram. Isograms for the purposes of this exercise are case-insensitive.

[rizzix]

wtd @ Fri May 09, 2008 10:38 pm wrote:

Isograms for the purposes of this exercise are case-insensitive.

You know that only makes it longer, not shorter

[btiffin]

Here's one in REBOL. It does not handle escaped quotes with backslash, but caret instead.

REBOL:

#!rebol -c rebol [] line: ask "? " words: parse/all line " " probe words foreach word words [     mod: uppercase trim/all copy word     if equal? mod reverse copy mod [print ["Palindrome:" word]]     iso: copy []     foreach ch mod [         alter iso ch     ]     if equal? length? iso length? word [print ["Isogram:" word]] ]

Output:

$ ./palins.r ? "Naomi, sex at noon taxes, I moan" aba "was saw" shockingly ["Naomi, sex at noon taxes, I moan" "aba" "was saw" "shockingly"] Palindrome: Naomi, sex at noon taxes, I moan Palindrome: aba Palindrome: was saw Isogram: shockingly

[rizzix]

Here everything inclusive of the case-check:

Haskell:

import List (nub) import Char (toLower) main = getLine >>= (mapM_ putStrLn) . (filter isIsoOrPalin) . strings []     where strings rs []        = reverse rs           strings rs ('\'':xs) = let (ys, ss) = esc_single xs [] in strings (ys:rs) ss           strings rs ('"':xs)  = let (ys, ss) = esc_double xs [] in strings (ys:rs) ss           strings rs (' ':xs)  = let (ys, ss) = word xs [] in strings (ys:rs) ss           strings rs xs        = let (ys, ss) = word xs [] in strings (ys:rs) ss                     esc_single ('\\':'\'':xs) rs = esc_single xs ('\'':rs)           esc_single [] rs             = (reverse rs, [])           esc_single ('\'':xs) rs      = (reverse rs, xs)           esc_single (x:xs) rs         = esc_single xs (x:rs)                     esc_double ('\\':'"':xs) rs  = esc_double xs ('"':rs)           esc_double [] rs             = (reverse rs, [])           esc_double ('"':xs) rs       = (reverse rs, xs)           esc_double (x:xs) rs         = esc_double xs (x:rs)                     word (' ':xs) rs = (reverse rs, xs)           word (x:xs)   rs = word xs (x:rs)           word []       rs = (reverse rs, [])                     isIsoOrPalin xs = let rs = filter ((/=) ' ') (map toLower xs)                             in xs == reverse xs || nub rs == rs

btiffin: is parse a builtin function?

[btiffin]

Quote:

btiffin: is parse a builtin function?

Yep, and I'm just using the string splitting feature. In block mode it includes Icon like pattern matching with to, thru charset matching, datatypes and literals, bobloblaw, with expression evaluation on every match step.

In the case I used, the rule was simply " ", meaning split using space as a delimiter. I really should have included ^- for tabs as well. And my code would probably FAIL given that I didn't bother to properly parse for backslash escaped quotes. REBOL uses caret for escaping.

Cheers