Jump to content

Compare a list of strings

From Rosetta Code
Task
Compare a list of strings
You are encouraged to solve this task according to the task description, using any language you may know.
Task

Given a   list   of arbitrarily many strings, show how to:

  •   test if they are all lexically equal
  •   test if every string is lexically less than the one after it (i.e. whether the list is in strict ascending order)


Each of those two tests should result in a single true or false value, which could be used as the condition of an   if   statement or similar.

If the input list has less than two elements, the tests should always return true.

There is no need to provide a complete program and output.

Assume that the strings are already stored in an array/list/sequence/tuple variable (whatever is most idiomatic) with the name   strings,   and just show the expressions for performing those two tests on it (plus of course any includes and custom functions etc. that it needs),   with as little distractions as possible.

Try to write your solution in a way that does not modify the original list,   but if it does then please add a note to make that clear to readers.

If you need further guidance/clarification,   see #Perl and #Python for solutions that use implicit short-circuiting loops,   and #Raku for a solution that gets away with simply using a built-in language feature.


Other tasks related to string operations:
Metrics
Counting
Remove/replace
Anagrams/Derangements/shuffling
Find/Search/Determine
Formatting
Song lyrics/poems/Mad Libs/phrases
Tokenize
Sequences



11l

Translation of: D
L(strings_s) [‘AA AA AA AA’, ‘AA ACB BB CC’]
   V strings = strings_s.split(‘ ’)
   print(strings)
   print(all(zip(strings, strings[1..]).map(a -> a[0] == a[1])))
   print(all(zip(strings, strings[1..]).map(a -> a[0]  < a[1])))
   print()

360 Assembly

The program uses one ASSIST macro (XPRNT) to keep the code as short as possible.

*        Compare a list of strings 31/01/2017
COMPLIST CSECT
         USING  COMPLIST,R13       base register
         B      72(R15)            skip savearea
         DC     17F'0'             savearea
         STM    R14,R12,12(R13)    prolog
         ST     R13,4(R15)         " <-
         ST     R15,8(R13)         " ->
         LR     R13,R15            " addressability
         MVC    SNAME,=C'ABC'
         LA     R1,SNAME
         LA     R2,ABC
         BAL    R14,TEST           call test('ABC',abc)
         MVC    SNAME,=C'AAA'
         LA     R1,SNAME
         LA     R2,AAA
         BAL    R14,TEST           call test('AAA',aaa)
         MVC    SNAME,=C'ACB'
         LA     R1,SNAME
         LA     R2,ACB
         BAL    R14,TEST           call test('ACB',acb)     
         L      R13,4(0,R13)       epilog 
         LM     R14,R12,12(R13)    " restore
         XR     R15,R15            " rc=0
         BR     R14                exit
*------- ----   test(name,xlist) -----------------------
TEST     MVC    NAME,0(R1)         store argument #1
         MVC    XLIST(6),0(R2)     store argument #2
         MVI    ALLEQ,X'01'        alleq=true
         MVI    INCRE,X'01'        incre=true
         LA     R6,1               i=1
LOOPI    LA     R2,NXLIST          hbound(xlist)
         BCTR   R2,0               -1
         CR     R6,R2              do i to hbound(xlist)-1
         BH     ELOOPI
         MVC    XBOOL,ALLEQ
         OC     XBOOL,INCRE        or
         CLI    XBOOL,X'01'        and while alleq or incre
         BNE    ELOOPI
         LA     R2,1(R6)           i+1
         SLA    R2,1               *2
         LA     R3,XLIST-2(R2)     @xlist(i+1)
         LR     R1,R6              i
         SLA    R1,1               *2
         LA     R4,XLIST-2(R1)     @xlist(i)
         CLC    0(2,R3),0(R4)      if xlist(i+1)=xlist(i)
         BNE    SEL1B
         MVI    INCRE,X'00'        incre=false
         B      SEL1END
SEL1B    CLC    0(2,R3),0(R4)      if xlist(i+1)<xlist(i)
         BNL    SEL1OTH
         MVI    INCRE,X'00'        incre=false
         MVI    ALLEQ,X'00'        alleq=false
         B      SEL1END
SEL1OTH  MVI    ALLEQ,X'00'        alleq=false
SEL1END  LA     R6,1(R6)           i=i+1
         B      LOOPI
ELOOPI   CLI    ALLEQ,X'01'        if alleq
         BNE    SEL2B
         MVC    TXT,=CL40'all elements are equal'
         B      SEL2END
SEL2B    CLI    INCRE,X'01'        if incre
         BNE    SEL2OTH
         MVC    TXT,=CL40'elements are in increasing order'
         B      SEL2END
SEL2OTH  MVC    TXT,=CL40'neither equal nor in increasing order'
SEL2END  MVI    PG,C' '
         MVC    PG+1(79),PG        clear buffer
         MVC    PG(3),NAME
         MVC    PG+3(3),=C' : '
         MVC    PG+6(40),TXT
         XPRNT  PG,L'PG
         BR     R14                return to caller
*        ----   ----------------------------------------
SNAME    DS     CL3
ABC      DC     CL2'AA',CL2'BB',CL2'CC'
AAA      DC     CL2'AA',CL2'AA',CL2'AA'
ACB      DC     CL2'AA',CL2'CC',CL2'BB'
NAME     DS     CL3
XLIST    DS     3CL2
NXLIST   EQU    (*-XLIST)/L'XLIST
ALLEQ    DS     X
INCRE    DS     X
TXT      DS     CL40
XBOOL    DS     X
PG       DS     CL80
         YREGS
         END    COMPLIST
Output:
ABC : elements are in increasing order
AAA : all elements are equal
ACB : neither equal nor in increasing order

Action!

DEFINE PTR="CARD"

BYTE FUNC AreEqual(PTR ARRAY a BYTE len)
  INT i

  FOR i=1 TO len-1
  DO
    IF SCompare(a(0),a(i))#0 THEN
      RETURN (0)
    FI
  OD
RETURN (1)

BYTE FUNC IsAscendingOrder(PTR ARRAY a BYTE len)
  INT i

  FOR i=1 TO len-1
  DO
    IF SCompare(a(i-1),a(i))>=0 THEN
      RETURN (0)
    FI
  OD
RETURN (1)

PROC Test(PTR ARRAY a BYTE len)
  INT i

  Print("Input array: [")
  FOR i=0 TO len-1
  DO
    Print(a(i))
    IF i<len-1 THEN
      Put(32)
    FI
  OD
  PrintE("]")

  IF AreEqual(a,len) THEN
    PrintE("All strings are lexically equal.")
  ELSE
    PrintE("Not all strings are lexically equal.")
  FI

  IF IsAscendingOrder(a,len) THEN
    PrintE("The list is in strict ascending order.")
  ELSE
    PrintE("The list is not in strict ascending order.")
  FI
  PutE()
RETURN

PROC Main()
  PTR ARRAY a1(4),a2(4),a3(4),a4(1)

  a1(0)="aaa" a1(1)="aaa" a1(2)="aaa" a1(3)="aaa"
  Test(a1,4)

  a2(0)="aaa" a2(1)="aab" a2(2)="aba" a2(3)="baa"
  Test(a2,4)

  a3(0)="aaa" a3(1)="aab" a3(2)="aba" a3(3)="aba"
  Test(a3,4)

  a4(0)="aaa"
  Test(a4,1)
RETURN
Output:

Screenshot from Atari 8-bit computer

Input array: [aaa aaa aaa aaa]
All strings are lexically equal.
The list is not in strict ascending order.

Input array: [aaa aab aba baa]
Not all strings are lexically equal.
The list is in strict ascending order.

Input array: [aaa aab aba aba]
Not all strings are lexically equal.
The list is not in strict ascending order.

Input array: [aaa]
All strings are lexically equal.
The list is in strict ascending order.

Ada

We will store the "list" of strings in a vector. The vector will hold "indefinite" strings, i.e., the strings can have different lengths.

  package String_Vec is new Ada.Containers.Indefinite_Vectors
     (Index_Type => Positive, Element_Type => String);
   
   use type String_Vec.Vector;

The equality test iterates from the first to the last-but one index. For index Idx, it checks checks if Strings(Idx) and Strings(Idx+1) are different. If the answer is yes for any Idx, the function immediately returns False. If the answer is no for all Idx, the function finally returns True.

   function All_Are_The_Same(Strings: String_Vec.Vector) return Boolean is
   begin
      for Idx in Strings.First_Index .. Strings.Last_Index-1 loop
	 if Strings(Idx) /= Strings(Idx+1) then
	    return False;
	 end if;
      end loop;
      return True;
   end All_Are_The_Same;

Similarily, the strictly ascending test checks if Strings(Idx) is greater or equal Strings(Idx+1).

   function Strictly_Ascending(Strings: String_Vec.Vector) return Boolean is
   begin
      for Idx in Strings.First_Index+1 .. Strings.Last_Index loop
	 if Strings(Idx-1) >= Strings(Idx) then
	    return False;
	 end if;
      end loop;
      return True;
   end Strictly_Ascending;

If the variable Strings is of the type String_Vec.vector, one can call these two functions as usual.

Put_Line(Boolean'Image(All_Are_The_Same(Strings)) & ", " &
         Boolean'Image(Strictly_Ascending(Strings)));

If Strings holds two or more strings, the result will be either of TRUE, FALSE, or FALSE, TRUE, or FALSE, FALSE, indicating all strings are the same, or they are strictly ascending, or neither.

However, if Strings only holds zero or one string, the result will be TRUE, TRUE.

ALGOL 68

[]STRING list1 = ("AA","BB","CC");
[]STRING list2 = ("AA","AA","AA");
[]STRING list3 = ("AA","CC","BB");
[]STRING list4 = ("AA","ACB","BB","CC");
[]STRING list5 = ("single_element");

[][]STRING all lists to test = (list1, list2, list3, list4, list5);

PROC equal = ([]STRING list) BOOL:
   BEGIN
      BOOL ok := TRUE;
      FOR i TO UPB list - 1 WHILE ok DO
         ok := list[i] = list[i+1]
      OD;
      ok
   END;

PROC less than = ([]STRING list) BOOL:
   BEGIN
      BOOL ok := TRUE;
      FOR i TO UPB list - 1 WHILE ok DO
         ok := list[i] < list[i + 1]
      OD;
      ok
   END;

FOR i TO UPB all lists to test DO
   []STRING list = all lists to test[i];
   print (("list:", (STRING s; FOR i TO UPB list DO s +:= " " + list[i] OD; s), new line));
   IF equal (list) THEN
      print (("...is lexically equal", new line))
   ELSE
      print (("...is not lexically equal", new line))
   FI;
   IF less than (list) THEN
      print (("...is in strict ascending order", new line))
   ELSE
      print (("...is not in strict ascending order", new line))
   FI
OD
Output:
list: AA BB CC
...is not lexically equal
...is in strict ascending order
list: AA AA AA
...is lexically equal
...is not in strict ascending order
list: AA CC BB
...is not lexically equal
...is not in strict ascending order
list: AA ACB BB CC
...is not lexically equal
...is in strict ascending order
list: single_element
...is lexically equal
...is in strict ascending order

ALGOL W

    % returns true if all elements of the string array a are equal, false otherwise %
    % As Algol W procedures cannot determine the bounds of an array, the bounds     %
    % must be specified in lo and hi                                                %
    logical procedure allStringsEqual ( string(256) array a ( * )
                                      ; integer     value lo, hi
                                      ) ;
    begin
        logical same;
        integer listPos;
        same    := true;
        listPos := lo + 1;
        while same and listPos <= hi do begin
            same    := a( lo ) = a( listPos );
            listPos := listPos + 1
        end;
        same
    end allStringsEqual ;

    % returns true if the elements of the string array a are in ascending order,    %
    % false otherwise                                                               %
    % As Algol W procedures cannot determine the bounds of an array, the bounds     %
    % must be specified in lo and hi                                                %
    logical procedure ascendingOrder  ( string(256) array a ( * )
                                      ; integer     value lo, hi
                                      ) ;
    begin
        logical ordered;
        integer listPos;
        ordered := true;
        listPos := lo + 1;
        while ordered and listPos <= hi do begin
            ordered := a( listPos - 1 ) < a( listPos );
            listPos := listPos + 1
        end;
        ordered
    end ascendingOrder ;

AppleScript

Translation of: JavaScript

(ES6 Functional example)


-- allEqual :: [String] -> Bool
on allEqual(xs)
    _and(zipWith(my _equal, xs, rest of xs))
end allEqual

-- azSorted :: [String] -> Bool
on azSorted(xs)
    _and(zipWith(my azBeforeOrSame, xs, rest of xs))
end azSorted

-- _equal :: a -> a -> Bool
on _equal(a, b)
    a = b
end _equal

-- azBefore :: String -> String -> Bool
on azBeforeOrSame(a, b)
    a  b
end azBeforeOrSame

-- _and :: [a] -> Bool
on _and(xs)
    foldr(_equal, true, xs)
end _and


-- TEST
on run
    set lstA to ["isiZulu", "isiXhosa", "isiNdebele", "Xitsonga", "Tshivenda", ¬
        "Setswana", "Sesotho sa Leboa", "Sesotho", "English", "Afrikaans"]
    
    set lstB to ["Afrikaans", "English", "Sesotho", "Sesotho sa Leboa", "Setswana", ¬
        "Tshivenda", "Xitsonga", "isiNdebele", "isiXhosa", "isiZulu"]
    
    set lstC to ["alpha", "alpha", "alpha", "alpha", "alpha", "alpha", "alpha", ¬
        "alpha", "alpha", "alpha"]
    
    
    {allEqual:map(allEqual, [lstA, lstB, lstC]), azSorted:map(azSorted, [lstA, lstB, lstC])}
    
    -- > {allEqual:{false, false, true}, azSorted:{false, true, true}}
end run



-- GENERIC FUNCTIONS

-- foldr :: (a -> b -> a) -> a -> [b] -> a
on foldr(f, startValue, xs)
    tell mReturn(f)
        set v to startValue
        set lng to length of xs
        repeat with i from lng to 1 by -1
            set v to lambda(v, item i of xs, i, xs)
        end repeat
        return v
    end tell
end foldr

-- map :: (a -> b) -> [a] -> [b]
on map(f, xs)
    tell mReturn(f)
        set lng to length of xs
        set lst to {}
        repeat with i from 1 to lng
            set end of lst to lambda(item i of xs, i, xs)
        end repeat
        return lst
    end tell
end map

-- zipWith :: (a -> b -> c) -> [a] -> [b] -> [c]
on zipWith(f, xs, ys)
    set nx to length of xs
    set ny to length of ys
    if nx < 1 or ny < 1 then
        {}
    else
        set lng to cond(nx < ny, nx, ny)
        set lst to {}
        tell mReturn(f)
            repeat with i from 1 to lng
                set end of lst to lambda(item i of xs, item i of ys)
            end repeat
            return lst
        end tell
    end if
end zipWith

-- cond :: Bool -> (a -> b) -> (a -> b) -> (a -> b)
on cond(bool, f, g)
    if bool then
        f
    else
        g
    end if
end cond

-- Lift 2nd class handler function into 1st class script wrapper 
-- mReturn :: Handler -> Script
on mReturn(f)
    if class of f is script then
        f
    else
        script
            property lambda : f
        end script
    end if
end mReturn
Output:
{allEqual:{false, false, true}, azSorted:{false, true, true}}

Arturo

allEqual?: function [lst] -> 1 = size unique lst
ascending?: function [lst] -> lst = sort lst

lists: [
    ["abc" "abc" "abc"]
    ["abc" "abd" "abc"]
    ["abc" "abd" "abe"]
    ["abc" "abe" "abd"]
]

loop lists 'l [
    print ["list:" l]
    print ["allEqual?" allEqual? l]
    print ["ascending?" ascending? l "\n"]
]
Output:
list: [abc abc abc] 
allEqual? true 
ascending? true 
 
list: [abc abd abc] 
allEqual? false 
ascending? false 
 
list: [abc abd abe] 
allEqual? false 
ascending? true 
 
list: [abc abe abd] 
allEqual? false 
ascending? false

AWK

# syntax: GAWK -f COMPARE_A_LIST_OF_STRINGS.AWK
BEGIN {
    main("AA,BB,CC")
    main("AA,AA,AA")
    main("AA,CC,BB")
    main("AA,ACB,BB,CC")
    main("single_element")
    exit(0)
}
function main(list,  arr,i,n,test1,test2) {
    test1 = 1 # elements are identical
    test2 = 1 # elements are in ascending order
    n = split(list,arr,",")
    printf("\nlist:")
    for (i=1; i<=n; i++) {
      printf(" %s",arr[i])
      if (i > 1) {
        if (arr[i-1] != arr[i]) {
          test1 = 0 # elements are not identical
        }
        if (arr[i-1] >= arr[i]) {
          test2 = 0 # elements are not in ascending order
        }
      }
    }
    printf("\n%d\n%d\n",test1,test2)
}
Output:
list: AA BB CC
0
1

list: AA AA AA
1
0

list: AA CC BB
0
0

list: AA ACB BB CC
0
1

list: single_element
1
1

BQN

If grade up matches grade down, then all elements are equal.

If all are not equal and the list is invariant under sorting, then it is in ascending order.

AllEq  ⍋≡⍒
Asc  ¬AllEq∧∧≡⊢

•Show AllEq "AA", "AA", "AA", "AA"
•Show Asc "AA", "AA", "AA", "AA"

•Show AllEq "AA", "ACB", "BB", "CC"
•Show Asc "AA", "ACB", "BB", "CC"
1
0
0
1

Try It!

Bracmat

Some explanation of the tests:

test1 and test2 are functions that return their input, but, more importantly, either succeed or fail.

first and x are local variables in test1 and test2, respectively.

The bodies of the two functions consist of pattern matching operations that either succeed or fail. The pattern matching operator is the colon :. This operator, like all Bracmat's operators, is binary. The operand on the left hand side is the subject, the operand on the right hand side is the pattern.

The symbols ?, !, %, @, >, and ~ are prefixes.

? when prefixed to a symbol like first or x, makes the symbol a variable that receives the value of the subject or of part of te subject, without constraining what can be received. When prefixed to a zero length symbol (the empty string), it matches anything, like a wildcard.

! when prefixed to a symbol like first or x, evaluates to the value that was bound to the symbol. So it complements the ? prefix. A symbol is a variable if and only if it is prefixed with ? or !.

% is a prefix that modifies a pattern component such that it can match one or more elements from the subject. So it is more restrictive than ?.

@ is a prefix that modifies a pattern component such that it can match zero or one elements from the subject. So it is (much) more restrictive than ?. The combination %@ means: this subpattern can only match exactly one element.

> is a prefix that modifies a pattern component to only match values that are greater than the value of the pattern component.

~ is a prefix used to negate what comes after it. In test1, the first ~ negates the outcome of a pattern matching operation. In the subpattern ~!first it says: match anything that is not the value of !first. In ~>!x it is negates the prefix >. Together, ~> means: "not greater than" or, equivalently, "less than or equal to".

If a pattern match operator occurs inside a pattern as in %@:~>!x, then both operands are patterns. So this expression is to be read as:"match exactly one element of the subject and require that it is less than or equal to the value of x".

In words, the tests do the following: test1 assigns the first element of the argument to the "first" and then looks for another element that is not equal to "first". If the search succeeds, test1 fails and if the search fails, test1 succeeds. Test2 searches for two consecutive elements where the second element is not greater than the first elemnt. If the search succeeds, test2 fails and if the search fails, test2 succeeds.


  (test1=first.~(!arg:%@?first ? (%@:~!first) ?))
& (test2=x.~(!arg:? %@?x (%@:~>!x) ?))

Demonstration

(     ( lstA
      .   isiZulu
          isiXhosa
          isiNdebele
          Xitsonga
          Tshivenda
          Setswana
          "Sesotho sa Leboa"
          Sesotho
          English
          Afrikaans
      )
      ( lstB
      .   Afrikaans
          English
          Sesotho
          "Sesotho sa Leboa"
          Setswana
          Tshivenda
          Xitsonga
          isiNdebele
          isiXhosa
          isiZulu
      )
      ( lstC
      .   alpha
          alpha
          alpha
          alpha
          alpha
          alpha
          alpha
          alpha
          alpha
          alpha
      )
  : ?lists
&   map
  $ ( (
      =   name list
        .   !arg:(?name.?list)
          &   out
            $ ( test1
                !name
                (test1$!list&succeeds|fails)
              )
          &   out
            $ ( test2
                !name
                (test2$!list&succeeds|fails)
              )
      )
    . !lists
    )
)

Output

test1 lstA fails
test2 lstA fails
test1 lstB fails
test2 lstB succeeds
test1 lstC succeeds
test2 lstC fails

Bruijn

Translation of: Haskell
:import std/String .

all-eq? [land? (zip-with eq? 0 (tail 0))]

all-gt? [land? (zip-with lt? 0 (tail 0))]

# --- tests ---

list-a "abc" : ("abc" : {}("abc"))

list-b "abc" : ("def" : {}("ghi"))

:test (all-eq? list-a) ([[1]])
:test (all-eq? list-b) ([[0]])
:test (all-gt? list-a) ([[0]])
:test (all-gt? list-b) ([[1]])

C

#include <stdbool.h>
#include <string.h>

static bool
strings_are_equal(const char **strings, size_t nstrings)
{
    for (size_t i = 1; i < nstrings; i++)
        if (strcmp(strings[0], strings[i]) != 0)
            return false;
    return true;
}

static bool
strings_are_in_ascending_order(const char **strings, size_t nstrings)
{
    for (size_t i = 1; i < nstrings; i++)
        if (strcmp(strings[i - 1], strings[i]) >= 0)
            return false;
    return true;
}

C#

Works with: C sharp version 7
public static (bool lexicallyEqual, bool strictlyAscending) CompareAListOfStrings(List<string> strings) =>
    strings.Count < 2 ? (true, true) :
    (
        strings.Distinct().Count() < 2,
        Enumerable.Range(1, strings.Count - 1).All(i => string.Compare(strings[i-1], strings[i]) < 0)
    );

C++

Assuming that the strings variable is of type T<std::string> where T is an ordered STL container such as std::vector:

Works with: C++ version 11
#include <algorithm>
#include <string>

// Bug: calling operator++ on an empty collection invokes undefined behavior.
std::all_of( ++(strings.begin()), strings.end(),
             [&](std::string a){ return a == strings.front(); } )  // All equal

std::is_sorted( strings.begin(), strings.end(),
                [](std::string a, std::string b){ return !(b < a); }) )  // Strictly ascending

Clojure

Used similar approach as the Python solution

;; Checks if all items in strings list are equal (returns true if list is empty)
(every?	(fn [[a nexta]] (= a nexta)) (map vector strings (rest strings))))

;; Checks strings list is in ascending order (returns true if list is empty)
(every?	(fn [[a nexta]] (<= (compare a nexta) 0)) (map vector strings (rest strings))))

COBOL

Works with: GnuCOBOL
       identification division.
       program-id. CompareLists.

       data division.
       working-storage section.
       78  MAX-ITEMS              value 3.
       77  i                      pic 9(2).
       01  the-list.
           05 list-items occurs MAX-ITEMS.
              10 list-item        pic x(3).
       01  results.
           05 filler              pic 9(1).
              88 equal-strings    value 1 when set to false is 0.
           05 filler              pic 9(1).
              88 ordered-strings  value 1 when set to false is 0.

       procedure division.
       main.
           move "AA BB CC" to the-list
           perform check-list
           move "AA AA AA" to the-list
           perform check-list
           move "AA CC BB" to the-list
           perform check-list
           move "AA ACBBB CC" to the-list
           perform check-list
           move "AA" to the-list
           perform check-list
           stop run
           .
       check-list.
           display "list:"
           set equal-strings to true
           set ordered-strings to true
           perform varying i from 1 by 1 until i > MAX-ITEMS
              if list-item(i) <> spaces
                 display function trim(list-item(i)), " " no advancing
                 if i < MAX-ITEMS and list-item(i + 1) <> spaces
                    if list-item(i + 1) <> list-item(i)
                       set equal-strings to false            
                    end-if
                    if list-item(i + 1) <= list-item(i) 
                       set ordered-strings to false            
                    end-if
                 end-if
              end-if
           end-perform
           display " "
           if equal-strings
              display "... is lexically equal"
           else
              display "... is not lexically equal"
           end-if
           if ordered-strings
              display "... is in strict ascending order"
           else
              display "... is not in strict ascending order"
           end-if
           display " "
           .
Output:
list:
AA BB CC  
... is not lexically equal
... is in strict ascending order
 
list:
AA AA AA  
... is lexically equal
... is not in strict ascending order
 
list:
AA CC BB  
... is not lexically equal
... is not in strict ascending order
 
list:
AA ACB BB  
... is not lexically equal
... is in strict ascending order
 
list:
AA  
... is lexically equal
... is in strict ascending order

Common Lisp

(defun strings-equal-p (strings)
  (null (remove (first strings) (rest strings) :test #'string=)))

(defun strings-ascending-p (strings)
  (loop for string1 = (first strings) then string2
        for string2 in (rest strings)
        always (string-lessp string1 string2)))

D

void main() {
    import std.stdio, std.algorithm, std.range, std.string;

    foreach (const strings; ["AA AA AA AA", "AA ACB BB CC"].map!split) {
        strings.writeln;
        strings.zip(strings.dropOne).all!(ab => ab[0] == ab[1]).writeln;
        strings.zip(strings.dropOne).all!(ab => ab[0] < ab[1]).writeln;
        writeln;
    }
}
Output:
["AA", "AA", "AA", "AA"]
true
false

["AA", "ACB", "BB", "CC"]
false
true

DuckDB

Works with: DuckDB version V1.0

This entry covers lists as values and as columns in a table.

Lists as columns

Let's suppose the list of strings is in column s of a table, t.

Then to check whether all the strings are the same, one could run the query:

select true = all (select coalesce( (lag(s) over ()) = s, true)
                   from t);

And to check whether the strings are monotonic non-decreasing:

select true = all (select coalesce(lag(s) over () <= s, true)
                   from t);
Output:

These queries can easily be modified to take into account different requirements, e.g. regarding collation.

Lists as values

The above solutions are easily adapted to DuckDB lists, for which we can readily define DuckDB functions:

create or replace function all_equal(lst) as (
  with l as (select unnest(lst) as s)
  select true = all (select coalesce( (lag(s) over ()) = s, true)
                     from l)
);
  
create or replace function monotonic_non_decreasing(lst) as (
  with l as (select unnest(lst) as s)
  select true = all (select coalesce(lag(s) over () <= s, true)
                     from l)
);

# Examples

select l, all_equal(l), monotonic_non_decreasing(l)
from (select ['a','b'] as l union all select [] as l);
Output:
┌───────────┬──────────────┬─────────────────────────────┐
│     l     │ all_equal(l) │ monotonic_non_decreasing(l) │
│ varchar[] │   boolean    │           boolean           │
├───────────┼──────────────┼─────────────────────────────┤
│ [a, b]    │ false        │ true                        │
│ []        │ true         │ true                        │
└───────────┴──────────────┴─────────────────────────────┘

Delphi

Translation of: Go
program Compare_a_list_of_strings;

{$APPTYPE CONSOLE}

uses
  System.SysUtils;

type
  // generic alias for use helper. The "TArray<string>" will be work too
  TListString = TArray<string>;

  TListStringHelper = record helper for TListString
    function AllEqual: boolean;
    function AllLessThan: boolean;
    function ToString: string;
  end;

{ TListStringHelper }

function TListStringHelper.AllEqual: boolean;
begin
  Result := True;
  if Length(self) < 2 then
    exit;

  var first := self[0];
  for var i := 1 to High(self) do
    if self[i] <> first then
      exit(False);
end;

function TListStringHelper.AllLessThan: boolean;
begin
  Result := True;
  if Length(self) < 2 then
    exit;

  var last := self[0];
  for var i := 1 to High(self) do
  begin
    if not (last < self[i]) then
      exit(False);
    last := self[i];
  end;
end;

function TListStringHelper.ToString: string;
begin
  Result := '[';
  Result := Result + string.join(', ', self);
  Result := Result + ']';
end;

var
  lists: TArray<TArray<string>>;

begin
  lists := [['a'], ['a', 'a'], ['a', 'b']];

  for var list in lists do
  begin
    writeln(list.ToString);
    writeln('Is AllEqual: ', list.AllEqual);
    writeln('Is AllLessThan: ', list.AllLessThan, #10);
  end;

  readln;
end.
Output:
[a]
Is AllEqual: TRUE
Is AllLessThan: TRUE

[a, a]
Is AllEqual: TRUE
Is AllLessThan: FALSE

[a, b]
Is AllEqual: FALSE
Is AllLessThan: TRUE

Dyalect

func isSorted(xs) {
    var prev
    for x in xs {
        if prev && !(x > prev) {
            return false
        }
        prev = x
    }
    true
}

func isEqual(xs) {
    var prev
    for x in xs {
        if prev && x != prev {
            return false
        }
        prev = x
    }
    true
}

EasyLang

Translation of: AWK
proc test s$[] . .
   ident = 1
   ascend = 1
   for i = 2 to len s$[]
      h = strcmp s$[i] s$[i - 1]
      if h <> 0
         ident = 0
      .
      if h <= 0
         ascend = 0
      .
   .
   print s$[]
   if ident = 1
      print "all equal"
   .
   if ascend = 1
      print "ascending"
   .
   print ""
.
test [ "AA" "BB" "CC" ]
test [ "AA" "AA" "AA" ]
test [ "AA" "CC" "BB" ]
test [ "AA" "ACB" "BB" "CC" ]
test [ "single_element" ]

Elena

ELENA 6.x :

import system'collections;
import system'routines;
import extensions;

extension helper
{
    isEqual()
        = nil == self.seekEach(self.FirstMember, (n,m => m != n));
        
    isAscending()
    {
        var former := self.enumerator();
        var later := self.enumerator();
        
        later.next();
        
        ^ nil == former.zipBy(later, (prev,next => next <= prev )).seekEach::(b => b)
    }
}

testCases
    = new string[][]{
         new string[]{"AA","BB","CC"},
         new string[]{"AA","AA","AA"},
         new string[]{"AA","CC","BB"},
         new string[]{"AA","ACB","BB","CC"},
         new string[]{"single_element"}};

public program()
{
    testCases.forEach::(list)
        {
            console.printLine(list.asEnumerable()," all equal - ",list.isEqual());
            console.printLine(list.asEnumerable()," ascending - ",list.isAscending())
        };
        
    console.readChar()
}
Output:
AA,BB,CC all equal - false
AA,BB,CC ascending - true
AA,AA,AA all equal - true
AA,AA,AA ascending - true
AA,CC,BB all equal - false
AA,CC,BB ascending - false
AA,ACB,BB,CC all equal - false
AA,ACB,BB,CC ascending - true
single_element all equal - true
single_element ascending - true

Elixir

defmodule RC do
  def compare_strings(strings) do
    {length(Enum.uniq(strings))<=1, strict_ascending(strings)}
  end
  
  defp strict_ascending(strings) when length(strings) <= 1, do: true
  defp strict_ascending([first, second | _]) when first >= second, do: false
  defp strict_ascending([_, second | rest]), do: strict_ascending([second | rest])
end

lists = [ ~w(AA AA AA AA), ~w(AA ACB BB CC), ~w(AA CC BB), [], ["XYZ"] ]
Enum.each(lists, fn list ->
  IO.puts "#{inspect RC.compare_strings(list)}\t<= #{inspect list} "
end)
Output:
{true, false}   <= ["AA", "AA", "AA", "AA"]
{false, true}   <= ["AA", "ACB", "BB", "CC"]
{false, false}  <= ["AA", "CC", "BB"]
{true, true}    <= []
{true, true}    <= ["XYZ"]

Erlang

Translation of: Haskell
-module(compare_strings).

-export([all_equal/1,all_incr/1]).

all_equal(Strings) ->
	all_fulfill(fun(S1,S2) -> S1 == S2 end,Strings).

all_incr(Strings) ->
	all_fulfill(fun(S1,S2) -> S1 < S2 end,Strings).

all_fulfill(Fun,Strings) ->
	lists:all(fun(X) -> X end,lists:zipwith(Fun, lists:droplast(Strings), tl(Strings)) ).

F#

let allEqual strings = Seq.isEmpty strings || Seq.forall (fun x -> x = Seq.head strings) (Seq.tail strings)
let ascending strings = Seq.isEmpty strings || Seq.forall2 (fun x y -> x < y) strings (Seq.tail strings)

Actually allEqual is a shortcut and ascending is a general pattern. We can make a function out of it which constructs a new function from a comparision function

let (!) f s = Seq.isEmpty s || Seq.forall2 f s (Seq.tail s)

and define the 2 task functions that way

let allEqual = !(=)
let ascending = !(<)

getting something similar as the builtin in Raku

Factor

Assuming the list is on top of the data stack, testing for lexical equality:

USE: grouping
all-equal?

Testing for ascending order:

USING: grouping math.order ;
[ before? ] monotonic?

Forth

Raw Forth

Note: This will work under some ANS-Forth systems. It assumes that WORD stores its string at HERE --- this isn't guaranteed by ANS-Forth.

Raw Forth is a very low level language and has no Native lists so we have to build from scratch. Remarkably by concatenating these low level operations and using the simple Forth parser we can build the linked lists of strings and the list operators quite simply. The operators and lists that we create become extensions to the language.

\ linked list of strings creators
: ,"       ( -- )  [CHAR] " WORD  c@ 1+ ALLOT  ;             \ Parse input stream until " and write into next available memory
: [[       ( -- )  0 C, ;                                    \ begin a list. write a 0 into next memory byte (null string)
: ]]       ( -- )  [[ ;                                      \ end list with same null string

: nth      ( n list -- addr) swap 0 do count + loop ;        \ return address of the Nth item in a list

: items    ( list -- n )                                     \ return the number of items in a list
          0 >R
          BEGIN
            COUNT + DUP
            R> 1+ >R
          0= UNTIL
          DROP
          R> 1- ;

: compare$ ( $1 $2 -- -n|0|n )  count rot count compare ;    \ compare is an ANS Forth word. returns 0 if $1=$2

: compare[]   ( list n1 n2 -- flag)                          \ compare items n1 and n2 in list
            ROT dup >R nth ( -- $1)
            swap r> nth    ( -- $1 $2)
            compare$ ;

\ create our lexical operators
: LEX=     ( list -- flag)
           0                                                 \ place holder for the flag
           over items 1
           DO
              over I  I 1+ compare[] +                       \ we sum the comparison results on the stack
           LOOP
           nip 0= ;

: LEX<     ( list -- flag)
           0                                                 \ place holder for the flag
           over items 1
           DO
              over I  I 1+ compare[] 0< NOT +
           LOOP
           nip 0= ;

\ make some lists
create strings  [[ ," ENTRY 4" ," ENTRY 3" ," ENTRY 2" ," ENTRY 1" ]]
create strings2 [[ ," the same" ," the same" ," the same" ]]
create strings3 [[ ," AAA" ," BBB" ," CCC" ," DDD" ]]

Test at the Forth console (-1 is the result for TRUE)

Output:
STRINGS  lex= . 0 ok
STRINGS2 lex= . -1 ok
STRINGS3 lex= . 0 ok 
STRINGS  lex< . 0 ok
STRINGS2 lex< . 0 ok
STRINGS3 lex< . -1 ok

novice-package

This depends upon having the novice-package available --- the novice-package is ANS-Forth, as is this code.

I don't think it is a good idea to write "Raw Forth" as described above. Application code is hard to write and hard to read when low-level code is mixed in with application code. It is better to hide low-level code in general-purpose code-libraries so that the application code can be simple. Here is my solution using LIST.4TH from my novice-package: https://www.forth2020.org/beginners-to-forth/a-novice-package

: test-equality ( string node -- new-string bad? ) 
    over count                          \ -- string node adr cnt 
    rot .line @ count   compare ; 

: test-ascending ( string node -- new-string bad? ) 
    .line @ >r 
    count  r@ count     compare -1 <>   \ -- bad? 
    r> swap ; 

: test-seq { seq 'test -- flag }        \ 'TEST picture: string node -- new-string bad? 
     seq length 2 < if  true exit then 
     seq .line @  seq 2nd  'test  find-node 
     nip  0= ;

Here is a test of the above code:

Output:
(( c" aaa" new-seq >> c" aaa" new-seq >> c" aaa" new-seq )) drop  ok-1 
dup  ' test-equality test-seq . -1  ok-1 
kill-seq  ok 
(( c" aaa" new-seq >> c" bbb" new-seq >> c" aaa" new-seq )) drop  ok-1 
dup  ' test-equality test-seq . 0  ok-1 
kill-seq  ok 
(( c" aaa" new-seq >> c" bbb" new-seq >> c" ccc" new-seq )) drop  ok-1 
dup  ' test-ascending test-seq . -1  ok-1 
kill-seq  ok 
(( c" aaa" new-seq >> c" bbb" new-seq >> c" aaa" new-seq )) drop  ok-1 
dup  ' test-ascending test-seq . 0  ok-1 
kill-seq  ok 

Fortran

Fortran does not offer a "string" item, which is to say, a sequence of items plus the length as one entity as in Pascal, among others. It does offer a CHARACTER variable, having some specified number of characters so the usual approach is to choose a length that is "long enough". In character comparisons, trailing spaces are ignored so that "xx" and "xx " are deemed equal. Similarly, it does not offer a list-of-thingies item, so again the usual approach is to provide an array of a size "long enough". One could develop a scheme with auxiliary counters stating how many elements are in use and so forth, but for this example, the parameterisation will do. Inspection of such arrays of character entities requires explicit DO-loops and IF-statements, and functions ALLINORDER and ALLEQUAL could be devised. Earlier Fortrans (prior to 77) lack a CHARACTER type, and so one must struggle with integer arrays.

Later Fortran (90 et seq) offers the special function ALL (and its associate, ANY) for testing multiple logical expressions, and also syntax allowing multiple elements of an array to be specified, as in A(3:7) to access elements 3, 4, 5, 6, 7 of array A. The ALL function has the special feature that if no logical expressions exist, then they, er, ... all ... are true and the result of ALL(nothing) is true. Well, none of them are false... Whatever the rationalisations this delivers the required result when the list has but one element and so there are no pairs to produce logical expressions, so, none of them are false, so the result is true, as specified.

On the other hand a function such as ALLINORDER would show the sound of one hand clapping. It would also reveal the order in which comparisons were made, and whether the loop would quit on the first failure or blockheadedly slog on through the lot regardless. Alas, on these questions the documentation for ALL is suspiciously silent.

      INTEGER MANY,LONG
      PARAMETER (LONG = 6,MANY = 4)	!Adjust to suit.
      CHARACTER*(LONG) STRINGS(MANY)	!A list of text strings.
      STRINGS(1) = "Fee"
      STRINGS(2) = "Fie"
      STRINGS(3) = "Foe"
      STRINGS(4) = "Fum"
      IF (ALL(STRINGS(1:MANY - 1) .LT. STRINGS(2:MANY))) THEN
        WRITE (6,*) MANY," strings: strictly increasing in order."
       ELSE
        WRITE (6,*) MANY," strings: not strictly increasing in order."
      END IF
      IF (ALL(STRINGS(1:MANY - 1) .EQ. STRINGS(2:MANY))) THEN
        WRITE (6,*) MANY," strings: all equal."
       ELSE
        WRITE (6,*) MANY," strings: not all equal."
      END IF
      END

And yes, if MANY is set to one and the extra texts are commented out, the results are both true, and ungrammatical statements are made. Honest. Possibly, another special function, as in COUNT(STRINGS(1:MANY - 1) .LT. STRINGS(2:MANY))) would involve less one-hand-clapping when there are no comparisons to make, but the production of a report that would use it is not in the specification.

F2003-F2008

F2008 standard ([ISO 2010], 4.4.3) defines the character variable of the character type as a set of values composed of character strings and a character string is a sequence of characters, numbered from left to right 1, 2, 3, ... up to the number of characters in the string. The number of characters in the string is called the length of the string. The length is a type parameter; its kind is processor dependent and its value is greater than or equal to zero. I.e in declaration

 character (len=12) :: surname

keyword len is NOT a size of array, it is an intrinsic parameter of character type, and character type is in fortran a first-class type: they can be assigned as objects or passed as parameters to a subroutine.

In summary, the character data type in Fortran is a real, first class data type. Fortran character strings are not hacked-up arrays!

program    compare_char_list
   implicit none
   character(len=6), allocatable, dimension(:) :: ss
   integer :: many
   ss = ["Fee","Fie","Foe","Fum"]
   many = size(ss)
   if (all(ss(1:many - 1) .lt. ss(2:many))) then
      write (*,*) many," strings: strictly increasing in order."
   else
      write (*,*) many," strings: not strictly increasing in order."
   end if
   if (all(ss(1:many - 1) .eq. ss(2:many))) then
      write (*,*) many," strings: all equal."
   else
      write (*,*) many," strings: not all equal."
   end if
end program compare_char_list

FreeBASIC

' FB 1.05.0 Win64

Function AllEqual(strings() As String) As Boolean
   Dim length As Integer = UBound(strings) - LBound(strings) + 1
   If length < 2 Then Return False
   For i As Integer = LBound(strings) + 1 To UBound(strings)
     If strings(i - 1) <> strings(i) Then Return False
   Next
   Return True
End Function

Function AllAscending(strings() As String) As Boolean
   Dim length As Integer = UBound(strings) - LBound(strings) + 1
   If length < 2 Then Return False
   For i As Integer = LBound(strings) + 1 To UBound(strings)
     If strings(i - 1) >= strings(i) Then Return False
   Next
   Return True
End Function

Fōrmulæ

Fōrmulæ programs are not textual, visualization/edition of programs is done showing/manipulating structures but not text. Moreover, there can be multiple visual representations of the same program. Even though it is possible to have textual representation —i.e. XML, JSON— they are intended for storage and transfer purposes more than visualization and edition.

Programs in Fōrmulæ are created/edited online in its website.

In this page you can see and run the program(s) related to this task and their results. You can also change either the programs or the parameters they are called with, for experimentation, but remember that these programs were created with the main purpose of showing a clear solution of the task, and they generally lack any kind of validation.


FutureBasic

local fn ListObjectsAreIdentical( array as CFArrayRef ) as BOOL
  BOOL result = NO
  
  CFSetRef set = fn SetWithArray( array )
  result = ( fn SetCount( set ) <= 1 )
end fn = result

local fn ListIsInLexicalOrder( array as CFArrayRef ) as BOOL
  BOOL result = NO
  
  CFArrayRef sortedArray = fn ArraySortedArrayUsingSelector( array, @"compare:" )
  result = fn ArrayIsEqual( array, sortedArray )
end fn = result

void local fn ListTest
  long i
  
  CFArrayRef listA = @[@"aaa", @"aaa", @"aaa", @"aaa"]
  CFArrayRef listB = @[@"aaa", @"aab", @"aba", @"baa"]
  CFArrayRef listC = @[@"caa", @"aab", @"aca", @"abc"]
  CFArrayRef lists = @[listA, listB, listC]
  
  for i = 0 to 2
    CFArrayRef temp = lists[i]
    printf @"Input array elements: %@ %@ %@ %@", temp[0], temp[1], temp[2], temp[3]
    if ( fn ListObjectsAreIdentical( temp ) )
      printf @"List elements are lexically equal."
    else
      printf @"List elements not lexically equal."
    end if
    if ( fn ListIsInLexicalOrder( temp ) == YES )
      printf @"List elements are in ascending order."
    else
      printf @"List elements not in ascending order."
    end if
    CFArrayRef sorted = fn ArraySortedArrayUsingSelector( temp, @"compare:" )
    printf @"List elements sorted in ascending order: %@ %@ %@ %@", sorted[0], sorted[1], sorted[2], sorted[3]
    print
  next
end fn

fn ListTest

HandleEvents
Output:
Input array elements: aaa aaa aaa aaa
List elements are lexically equal.
List elements are in ascending order.
List elements sorted in ascending order: aaa aaa aaa aaa

Input array elements: aaa aab aba baa
List elements not lexically equal.
List elements are in ascending order.
List elements sorted in ascending order: aaa aab aba baa

Input array elements: caa aab aca abc
List elements not lexically equal.
List elements not in ascending order.
List elements sorted in ascending order: aab abc aca caa

Go

package cmp

func AllEqual(strings []string) bool {
	for _, s := range strings {
		if s != strings[0] {
			return false
		}
	}
	return true
}

func AllLessThan(strings []string) bool {
	for i := 1; i < len(strings); i++ {
		if !(strings[i - 1] < s) {
			return false
		}
	}
	return true
}

See Compare_a_list_of_strings/GoTests for validation tests.

Note also there is the function sort.StringsAreSorted in the Go standard library. This function tests that the list is ordered by less than or equal to, but not strictly less than.

Gosu

var list = {"a", "b", "c", "d"}

var isHomogeneous = list.toSet().Count < 2 
var isOrderedSet = list.toSet().order().toList() == list

Haskell

allEqual :: Eq a => [a] -> Bool
allEqual xs = and $ zipWith (==) xs (tail xs)
 
allIncr :: Ord a => [a] -> Bool
allIncr xs = and $ zipWith (<) xs (tail xs)


Alternatively, using folds:

allEqual
  :: Eq a
  => [a] -> Bool
allEqual [] = True
allEqual (h:t) = foldl (\a x -> a && x == h) True t

allIncreasing
  :: Ord a
  => [a] -> Bool
allIncreasing [] = True
allIncreasing (h:t) = fst $ foldl (\(a, x) y -> (a && x < y, y)) (True, h) t

or seeking earlier exit (from longer lists) with until, but in fact, perhaps due to lazy execution, the zipWith at the top performs best.

allEq
  :: Eq a
  => [a] -> Bool
allEq [] = True
allEq (h:t) =
  null . snd $ 
  until 
    (\(x, xs) -> null xs || x /= head xs)
    (\(_, x:xs) -> (x, xs)) 
    (h, t)

allInc
  :: Ord a
  => [a] -> Bool
allInc [] = True
allInc (h:t) =
  null . snd $
  until
    (\(x, xs) -> null xs || x >= head xs)
    (\(_, x:xs) -> (x, xs))
    (h, t)

Icon and Unicon

Icon and Unicon expressions either succeed and return a value (which may be &null) or fail.

#
# list-compare.icn
#
link fullimag

procedure main()
   L1 := ["aa"]
   L2 := ["aa", "aa", "aa"]
   L3 := ["", "aa", "ab", "ac"]
   L4 := ["aa", "bb", "cc"]
   L5 := ["cc", "bb", "aa"]

   every L := (L1 | L2 | L3 | L4 | L5) do {
      writes(fullimage(L))
      writes(": equal ")
      writes(if allequal(L) then "true" else "false")
      writes(", ascending ")
      write(if ascending(L) then "true" else "false")
   }
end

# test for all identical
procedure allequal(L)
   if *L < 2 then return
   a := L[1]
   every b := L[2 to *L] do {
      if a ~== b then fail
      a := b
   }
   return
end

# test for strictly ascending
procedure ascending(L)
   if *L < 2 then return
   a := L[1]
   every b := L[2 to *L] do {
      if a >>= b then fail
      a := b
   }
   return
end
Output:
prompt$ unicon -s list-compare.icn -x
["aa"]: equal true, ascending true
["aa","aa","aa"]: equal true, ascending false
["","aa","ab","ac"]: equal false, ascending true
["aa","bb","cc"]: equal false, ascending true
["cc","bb","aa"]: equal false, ascending false

J

Solution (equality test):

   allEq =: 1 = +/@~:     NB. or 1 = #@:~. or -: 1&|. or }.-:}:

Solution (order test):

   asc =: /: -: i.@#      NB. or -: (/:~) etc.

Notes: asc indicates whether y is monotonically increasing, but not necessarily strictly monotonically increasing (in other words, it allows equal elements if they are adjacent to each other).

Java

This is a fairly basic procedure in Java, using for-loops, String.equals, and String.compareTo.

boolean allEqual(String[] strings) {
    String stringA = strings[0];
    for (String string : strings) {
        if (!string.equals(stringA))
            return false;
    }
    return true;
}
boolean isAscending(String[] strings) {
    String previous = strings[0];
    int index = 0;
    for (String string : strings) {
        if (index++ == 0)
            continue;
        if (string.compareTo(previous) < 0)
            return false;
        previous = string;
    }
    return true;
}


Alternately,

Works with: Java version 8
import java.util.Arrays;

public class CompareListOfStrings {

    public static void main(String[] args) {
        String[][] arr = {{"AA", "AA", "AA", "AA"}, {"AA", "ACB", "BB", "CC"}};
        for (String[] a : arr) {
            System.out.println(Arrays.toString(a));
            System.out.println(Arrays.stream(a).distinct().count() < 2);
            System.out.println(Arrays.equals(Arrays.stream(a).distinct().sorted().toArray(), a));
        }
    }
}
Output:
[AA, AA, AA, AA]
true
false
[AA, ACB, BB, CC]
false
true

JavaScript

ES5

Iterative

function allEqual(a) {
  var out = true, i = 0;
  while (++i<a.length) {
    out = out && (a[i-1] === a[i]);
  } return out;
}

function azSorted(a) {
  var out = true, i = 0;
  while (++i<a.length) {
    out = out && (a[i-1] < a[i]);
  } return out;
}

var e = ['AA', 'AA', 'AA', 'AA'], s = ['AA', 'ACB', 'BB', 'CC'], empty = [], single = ['AA'];
console.log(allEqual(e)); // true
console.log(allEqual(s)); // false
console.log(allEqual(empty)); // true
console.log(allEqual(single)); // true
console.log(azSorted(e)); // false
console.log(azSorted(s)); // true
console.log(azSorted(empty)); // true
console.log(azSorted(single)); // true

ES6

Functional

Using a generic zipWith, and functionally composed predicates:

(() => {
    'use strict';

    // allEqual :: [String] -> Bool
    let allEqual = xs => and(zipWith(equal, xs, xs.slice(1))),

        // azSorted :: [String] -> Bool
        azSorted = xs => and(zipWith(azBefore, xs, xs.slice(1))),

        // equal :: a -> a -> Bool
        equal = (a, b) => a === b,

        // azBefore :: String -> String -> Bool
        azBefore = (a, b) => a.toLowerCase() <= b.toLowerCase();


    // GENERIC

    // and :: [Bool] -> Bool
    let and = xs => xs.reduceRight((a, x) => a && x, true),

        // zipWith :: (a -> b -> c) -> [a] -> [b] -> [c]
        zipWith = (f, xs, ys) => {
            let ny = ys.length;
            return (xs.length <= ny ? xs : xs.slice(0, ny))
                .map((x, i) => f(x, ys[i]));
        };


    // TEST

    let lists = [
        ['isiZulu', 'isiXhosa', 'isiNdebele', 'Xitsonga',
            'Tshivenda', 'Setswana', 'Sesotho sa Leboa', 'Sesotho',
            'English', 'Afrikaans'
        ],
        ['Afrikaans', 'English', 'isiNdebele', 'isiXhosa',
            'isiZulu', 'Sesotho', 'Sesotho sa Leboa', 'Setswana',
            'Tshivenda', 'Xitsonga',
        ],
        ['alpha', 'alpha', 'alpha', 'alpha', 'alpha', 'alpha',
            'alpha', 'alpha', 'alpha', 'alpha', 'alpha', 'alpha'
        ]
    ];

    return {
        allEqual: lists.map(allEqual),
        azSorted: lists.map(azSorted)
    };

})();
Output:
{
  "allEqual": [
    false,
    false,
    true
  ],
  "azSorted": [
    false,
    true,
    true
  ]
}

jq

Works with: jq

Also works with gojq and jaq, the Go and Rust implementations of jq

For both the following functions, the input is assumed to be a (possibly empty) array. The elements may be of any JSON type.

# Are the strings all equal?
def lexically_equal:
  if length <= 1 then true
  else . as $in
  | all( range(0;length-1); $in[0] == $in[. + 1])
  end;

# Are the elements in strictly ascending order?
def lexically_ascending:
  . as $in
  | all( range(0;length-1); $in[.] < $in[. + 1]);

Examples:

[] | lexically_equal #=> true
["a", "ab"] | lexically_ascending #=> true

Jsish

Code from Javascript, ES5.

/* Compare list of strings, in Jsish */
function allEqual(a) {
  var out = true, i = 0;
  while (++i<a.length) {
    out = out && (a[i-1] === a[i]);
  } return out;
}

function allAscending(a) {
  var out = true, i = 0;
  while (++i<a.length) {
    out = out && (a[i-1] < a[i]);
  } return out;
}

if (allEqual(strings)) puts("strings array all equal");
else puts("strings array not all equal");

if (allAscending(strings)) puts("strings array in strict ascending order");
else puts("strings array not in strict ascending order");
Output:

None, task requirement asks for an assumed preloaded strings array, no full program, and little other distractions.

Julia

Works with: Julia version 0.6
allequal(arr::AbstractArray) = isempty(arr) || all(x -> x == first(arr), arr)

test = [["RC", "RC", "RC"], ["RC", "RC", "Rc"], ["RA", "RB", "RC"],
       ["RC"], String[], ones(Int64, 4), 1:4]

for v in test
    println("\n# Testing $v:")
    println("The elements are $("not " ^ !allequal(v))all equal.")
    println("The elements are $("not " ^ !issorted(v))strictly increasing.")
end
Output:
# Testing String["RC", "RC", "RC"]:
The elements are all equal.
The elements are strictly increasing.

# Testing String["RC", "RC", "Rc"]:
The elements are not all equal.
The elements are strictly increasing.

# Testing String["RA", "RB", "RC"]:
The elements are not all equal.
The elements are strictly increasing.

# Testing String["RC"]:
The elements are all equal.
The elements are strictly increasing.

# Testing String[]:
The elements are all equal.
The elements are strictly increasing.

# Testing [1, 1, 1, 1]:
The elements are all equal.
The elements are strictly increasing.

# Testing 1:4:
The elements are not all equal.
The elements are strictly increasing.

Klong

    {:[2>#x;1;&/=:'x]}:(["test" "test" "test"])
1
    {:[2>#x;1;&/<:'x]}:(["bar" "baz" "foo"])
1

Kotlin

// version 1.0.6

fun areEqual(strings: Array<String>): Boolean {
    if (strings.size < 2) return true
    return (1 until strings.size).all { strings[it] == strings[it - 1] }
}

fun areAscending(strings: Array<String>): Boolean {
    if (strings.size < 2) return true
    return (1 until strings.size).all { strings[it] > strings[it - 1] }
}

// The strings are given in the command line arguments

fun main(args: Array<String>) {
    println("The strings are : ${args.joinToString()}")
    if (areEqual(args)) println("They are all equal")
    else if (areAscending(args)) println("They are in strictly ascending order")
    else println("They are neither equal nor in ascending order")
}

Sample input/output:

Output:
The strings are : first, second, third
They are in strictly ascending order

Lambdatalk

{def allsame
 {def allsame.r
  {lambda {:s :n :i}
   {if {= :i :n}
    then true
    else {if {not {W.equal? {A.get :i :s} {A.get 0 :s}}}
    then false
    else {allsame.r :s :n {+ :i 1}} }}}}
 {lambda {:s}
  {allsame.r :s {- {A.length :s} 1} 0} }}
-> allsame

{def strict_order
 {def strict_order.r
  {lambda {:s :n :i}
   {if {= :i :n}
    then true
    else {if {W.inforequal? {A.get :i :s} {A.get {- :i 1} :s}}
    then false
    else {strict_order.r :s :n {+ :i 1}}}} }}
 {lambda {:s}
  {if {= {A.length :s} 1}
   then true
   else {strict_order.r :s {A.length :s} 1} }}}
-> strict_order

{S.map allsame 
       {A.new AA BB CC}
       {A.new AA AA AA}
       {A.new AA CC BB}
       {A.new AA ACB BB CC}
       {A.new single}
} -> false true false false true

{S.map strict_order 
       {A.new AA BB CC}
       {A.new AA AA AA}
       {A.new AA CC BB}
       {A.new AA ACB BB CC}
       {A.new single}
} -> true false false true true

Lua

function identical(t_str)
    _, fst = next(t_str)
    if fst then
        for _, i in pairs(t_str) do
            if i ~= fst then return false end
        end
    end
    return true
end

function ascending(t_str)
    prev = false
    for _, i in ipairs(t_str) do
        if prev and prev >= i then return false end
        prev = i
    end
    return true
end

function check(str)
    t_str = {}
    for i in string.gmatch(str, "[%a_]+") do
        table.insert(t_str, i)
    end
    str = str .. ": "
    if not identical(t_str) then str = str .. "not " end
    str = str .. "identical and "
    if not ascending(t_str) then str = str .. "not " end
    print(str .. "ascending.")
end

check("ayu dab dog gar panda tui yak")
check("oy oy oy oy oy oy oy oy oy oy")
check("somehow   somewhere  sometime")
check("Hoosiers")
check("AA,BB,CC")
check("AA,AA,AA")
check("AA,CC,BB")
check("AA,ACB,BB,CC")
check("single_element")
Output:
ayu dab dog gar panda tui yak: not identical and ascending.
oy oy oy oy oy oy oy oy oy oy: identical and not ascending.
somehow   somewhere  sometim: not identical and not ascending.
Hoosiers: identical and ascending.
AA,BB,CC: not identical and ascending.
AA,AA,AA: identical and not ascending.
AA,CC,BB: not identical and not ascending.
AA,ACB,BB,CC: not identical and ascending.
single_element: identical and ascending.

M2000 Interpreter

Module CheckIt {
      Function Equal(Strings){
            k=Each(Strings, 2, -1)
            a$=Array$(Strings, 0)
            =True
            While k {
                  =False
                  if a$<>array$(k) then exit
                  =True
            }
      }
      Function LessThan(Strings){
            =True
            if len(Strings)<2 then exit
            k=Each(Strings, 2)
            a$=Array$(Strings, 0)
            While k {
                  =False
                  if a$>=array$(k) then exit
                  a$=array$(k)
                  =True
            }
      }
      
      Print Equal(("alfa","alfa","alfa", "alfa"))=True
      Print Equal(("alfa",))=True
      Dim A$(10)="alfa"
      Print Equal(A$())=True
      Print Equal(("alfa1","alfa2","alfa3", "alfa4"))=False
      
      Print LessThan(("alfa1","alfa2","alfa3", "alfa4"))=True
      Print LessThan(("alfa1",))=true
      alfa$=Lambda$ k=1 ->{=String$("*", k) : k++}
      Dim A$(20)<<alfa$()
      Print LessThan(A$())=True
      A$(5)=""
      Print LessThan(A$())=False
}
Checkit

Maple

lexEqual := proc(lst)
	local i:
	for i from 2 to numelems(lst) do
		if lst[i-1] <> lst[i] then return false: fi:
	od:
	return true:
end proc:
lexAscending := proc(lst)
	local i:
	for i from 2 to numelems(lst) do
		if StringTools:-Compare(lst[i],lst[i-1]) then return false: fi:
	od:
	return true:
end proc:
tst := ["abc","abc","abc","abc","abc"]:
lexEqual(tst):
lexAscending(tst):
Examples:
true
false

Mathcad

Mathcad is a non-text-based programming environment. The expressions below are an approximations of the way that they are entered (and) displayed on a Mathcad worksheet. The worksheet is available at xxx_tbd_xxx

This particular version of "Compare a list of strings" was created in Mathcad Prime Express 7.0, a free version of Mathcad Prime 7.0 with restrictions (such as no programming or symbolics). All Mathcad numbers are complex doubles. There is a recursion depth limit of about 4,500. Strings are a distinct data and are not conceptually a list of integers.

-- define list of list of strings (nested vector of vectors of strings)
-- Mathcad vectors are single column arrays.
-- The following notation is for convenience in writing arrays in text form.
-- Mathcad array input is normally via Mathcad's array operator or via one of the
--   array-builder functions, such as stack or augment.
-- "," between vector elements indicates a new row.
-- " " between vector elements indicates a new column.

list:=["AA","AA","AA"],["AA","BB","CC"],["AA","CC","BB"],["CC","BB","AA"],["AA","ACB","BB","CC"],["AA"]]

-- define functions head and rest that return the first value in a list (vector)
-- and the list excluding the first element, respectively.

head(v):=if(IsArray(v),v[0,v)
rest(v):=if(rows(v)>1,submatrix(v,1,rows(v)-1,0,0),0)

-- define a function oprel that iterates through a list (vector) applying a comparison operator op to each pair of elements at the top of the list. 
-- Returns immediately with false (0) if a comparison fails.

oprel(op,lst,val):=if(op(val,head(lst)),if(rows(lst)>1,oprel(op,rest(lst),head(lst)),1),0)

oprel(op,lst):=if(rows(lst)>1,oprel(op,rest(lst),head(lst)),1)

-- define a set of boolean comparison functions
-- transpose represents Mathcad's transpose operator
-- vectorize represents Mathcad's vectorize operator

eq(a,b):=a=b    (transpose(vectorize(oprel,list))) = [1 0 0 0 0 1]  -- equal
lt(a,b):=a<b    (transpose(vectorize(oprel,list))) = [0 1 0 0 1 1]  -- ascending

-- oprel, eq and lt also work with numeric values

list:=[11,11,11],[11,22,33],[11,33,22],[33,22,11],[11,132,22,33],[11]]

Mathematica /Wolfram Language

data1 = {"aaa", "aaa", "aab"};
Apply[Equal, data]
OrderedQ[data]
Output:
False
True

MATLAB / Octave

Only the first task is implemented.

alist = {'aa', 'aa', 'aa'}
all(strcmp(alist,alist{1}))

Nanoquery

// a function to test if a list of strings are equal
def stringsEqual(stringList)
	// if the list is empty, return true
	if (len(stringList) = 0)
		return true
	end

	// otherwise get the first value and check for equality
	toCompare = stringList[0]
	equal = true
	for (i = 1) (equal && (i < len(stringList))) (i = i + 1)
		equal = (toCompare = stringList[i])
	end for


	// return whether the strings are equal or not
	return equal
end

// a function to test if a list of strings are are less than each other
def stringsLessThan(stringList)
	// if the list is empty, return true
	if (len(stringList) = 0)
		return true
	end

	// otherwise get the first value and check for less than
	toCompare = stringList[0]
	lessThan = true
	for (i = 1) (lessThan && (i < len(stringList))) (i = i + 1)
		lessThan = (toCompare < stringList[i])
		toCompare = stringList[i]
	end for

	// return whether the string were less than each other or not
	return lessThan
end

NetRexx

/* NetRexx */
options replace format comments java crossref symbols nobinary

runSample(arg)
return

-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
method isEqual(list = Rexx[]) public static binary returns boolean
  state = boolean (1 == 1) -- default to true
  loop ix = 1 while ix < list.length
    state = list[ix - 1] == list[ix]
    if \state then leave ix
    end ix
  return state

-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
method isAscending(list = Rexx[]) public static binary returns boolean
  state = boolean (1 == 1) -- default to true
  loop ix = 1 while ix < list.length
    state = list[ix - 1] << list[ix]
    if \state then leave ix
    end ix
  return state

-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
method runSample(arg) private static

  samples = [ -
      ['AA', 'BB', 'CC'] -
    , ['AA', 'AA', 'AA'] -
    , ['AA', 'CC', 'BB'] -
    , ['single_element'] -
    ]

  loop ix = 0 while ix < samples.length
    sample = samples[ix]
    if isEqual(sample)     then eq  = 'elements are identical'
                           else eq  = 'elements are not identical'
    if isAscending(sample) then asc = 'elements are in ascending order'
                           else asc = 'elements are not in ascending order'
    say 'List:' Arrays.toString(sample)
    say '  'eq
    say '  'asc
    end ix
  return
Output:
List: [AA, BB, CC]
  elements are not identical
  elements are in ascending order
List: [AA, AA, AA]
  elements are identical
  elements are not in ascending order
List: [AA, CC, BB]
  elements are not identical
  elements are not in ascending order
List: [single_element]
  elements are identical
  elements are in ascending order

newLISP

It's trivial.

(= "a" "a" "a")
true

(< "a" "a" "a")
nil

(= "a" "b" "c" "d" "e" "f" "g")
nil

(< "a" "b" "c" "d" "e" "f" "g")
true

However, the strings must be in lists.

(apply = '("a" "a" "a"))
true

(apply = '("a" "b" "c" "d" "e" "f" "g"))
nil

(apply < '("aaa" "b" "c" "d" "e" "f" "gggg"))
true

Nim

This is the obvious (and more efficient way) to compare strings in Nim:

func allEqual(s: openArray[string]): bool =

  for i in 1..s.high:
    if s[i] != s[0]:
      return false
  result = true

func ascending(s: openArray[string]): bool =

  for i in 1..s.high:
    if s[i] <= s[i - 1]:
      return false
  result = true

doAssert allEqual(["abc", "abc", "abc"])
doAssert not allEqual(["abc", "abd", "abc"])

doAssert ascending(["abc", "abd", "abe"])
doAssert not ascending(["abc", "abe", "abd"])

doAssert allEqual(["abc"])
doAssert ascending(["abc"])

For “allEqual”, there is another simple way using template “allIt” from standard module “sequtils”:

import sequtils

func allEqual(s: openArray[string]): bool =
  allIt(s, it == s[0])

doAssert allEqual(["abc", "abc", "abc"])
doAssert not allEqual(["abc", "abd", "abc"])
doAssert allEqual(["abc"])

There are other less obvious and less efficient ways, using hash sets, sorting or “map” and “zip”.

OCaml

open List;;

let analyze cmp l  =
  let rec analyze' l prevs =
    match l with
    [] -> true
    | [s] -> cmp prevs s
    | s::rest -> (cmp prevs s) && (analyze' rest s)
  in analyze' (List.tl l) (List.hd l)
;;

let isEqual     = analyze (=) ;;
let isAscending = analyze (<) ;;

let test sample =
   List.iter print_endline sample;
   if (isEqual sample)
       then (print_endline "elements are identical")
       else (print_endline "elements are not identical");
   if (isAscending sample)
	     then print_endline "elements are in ascending order"
         else print_endline "elements are not in ascending order";;


let lasc =   ["AA";"BB";"CC";"EE"];;
let leq =    ["AA";"AA";"AA";"AA"];;
let lnoasc = ["AA";"BB";"EE";"CC"];;

List.iter test [lasc;leq;lnoasc];;
Output:
AA
BB
CC
EE
elements are not identical
elements are in ascending order
AA
AA
AA
AA
elements are identical
elements are not in ascending order
AA
BB
EE
CC
elements are not identical
elements are not in ascending order

Oforth

: lexEqual   asSet size 1 <= ;
: lexCmp(l)  l l right( l size 1- ) zipWith(#<) and ;

ooRexx

/* REXX ---------------------------------------------------------------
* 28.06.2014 Walter Pachl
*--------------------------------------------------------------------*/
Call test 'ABC',.list~of('AA','BB','CC')
Call test 'AAA',.list~of('AA','AA','AA')
Call test 'ACB',.list~of('AA','CC','BB')
Exit

test: Procedure
Use Arg name,list
all_equal=1
increasing=1
Do i=0 To list~items-2
  i1=i+1
  Select
    When list[i1]==list[i] Then increasing=0
    When list[i1]<<list[i] Then Do
                                all_equal=0
                                increasing=0
                                End
    When list[i1]>>list[i] Then all_equal=0
    End
  End
Select
  When all_equal Then
    Say 'List' name': all elements are equal'
  When increasing Then
    Say 'List' name': elements are in increasing order'
  Otherwise
    Say 'List' name': neither equal nor in increasing order'
  End
Return
Output:
List ABC: elements are in increasing order
List AAA: all elements are equal
List ACB: neither equal nor in increasing order

PARI/GP

Easiest is to use Set():

allEqual(strings)=#Set(strings)<2
inOrder(strings)=Set(strings)==strings

More efficient:

allEqual(strings)=for(i=2,#strings,if(strings[i]!=strings[i-1], return(0))); 1
inOrder(strings)=for(i=2,#strings,if(strings[i]>strings[i-1], return(0))); 1

PascalABC.NET

function AllEqual(lst: sequence of string) 
  := lst.All(x -> lst.First = x);

function IsStrictAscending(lst: sequence of string): boolean
  := lst.Pairwise.All(x -> x[0] < x[1]);

begin
  var strings := |'abc','abc','abc','abc'|;
  Print(AllEqual(strings));
  var strings1 := |'abc','abd','ade','aef'|;
  Print(IsStrictAscending(strings1));
end.

Perl

use List::Util 1.33 qw(all);

all { $strings[0] eq $strings[$_] } 1..$#strings  # All equal
all { $strings[$_-1] lt $strings[$_] } 1..$#strings  # Strictly ascending

Alternatively, if you can guarantee that the input strings don't contain null bytes, the equality test can be performed by a regex like this:

join("\0", @strings) =~ /^ ( [^\0]*+ ) (?: \0 \1 )* $/x  # All equal

Phix

with javascript_semantics
function allsame(sequence s)
    for i=2 to length(s) do
        if s[i]!=s[1] then return false end if
    end for
    return true
end function
 
function strict_order(sequence s)
    for i=2 to length(s) do
        if s[i]<=s[i-1] then return false end if
    end for
    return true
end function
 
procedure test(sequence s)
    printf(1,"%-22V allsame:%5t, strict_order:%5t\n",{s,allsame(s),strict_order(s)})
end procedure
 
test({"AA","BB","CC"})
test({"AA","AA","AA"})
test({"AA","CC","BB"})
test({"AA","ACB","BB","CC"})
test({"single_element"})
Output:
{"AA","BB","CC"}       allsame:false, strict_order: true
{"AA","AA","AA"}       allsame: true, strict_order:false
{"AA","CC","BB"}       allsame:false, strict_order:false
{"AA","ACB","BB","CC"} allsame:false, strict_order: true
{"single_element"}     allsame: true, strict_order: true

Phixmonti

include ..\Utilitys.pmt

( "alpha" "beta" "gamma" "delta" "epsilon" "zeta"
      "eta" "theta" "iota" "kappa" "lambda" "mu" )

dup dup sort ==	/# put 0 (false) in the pile, indicating that they are not in ascending order #/

drop	/# discard the result #/

dup len swap 1 get rot repeat == /# put 0 (false) in the pile, indicating that they are not repeated strings #/

Picat

main =>
  Lists = [["AA","BB","CC"],
           ["AA","AA","AA"],
           ["AA","CC","BB"],
           ["AA","ACB","BB","CC"],
           ["single_element"],
           []],
  foreach(L in Lists)
    Same = all_same(L).cond(true,false),
    Sorted = sorted(L).cond(true,false),   
    printf("%-18w all_same:%-5w  sorted:%-5w\n",L,Same,Sorted)
  end.

all_same([]).
all_same([_]).
all_same([A,B|Rest]) :-
  A == B,
  all_same([B|Rest]).
Output:
[AA,BB,CC]         all_same:false  sorted:true 
[AA,AA,AA]         all_same:true   sorted:true 
[AA,CC,BB]         all_same:false  sorted:false
[AA,ACB,BB,CC]     all_same:false  sorted:true 
[single_element]   all_same:true   sorted:true 
[]                 all_same:true   sorted:true 

PicoLisp

PicoLisp has the native operators =, > and < these can take an infinite number of arguments and are also able to compare Transient symbols (the Strings of PicoLisp).

(= "AA" "AA" "AA")
-> T
(= "AA" "AA" "Aa")
-> NIL
(< "AA" "AA")
-> NIL
(< "AA" "Aa")
-> T
(< "1" "A" "B" "Z" "c" )
-> T
(> "A" "B" "Z" "C")
-> NIL

If you want a function which takes one list here are some straight-forward implementation:

(de same (List)
  (apply = List))

(de sorted (List)
  (apply < List))

(de sorted-backwards (List)
  (apply > List))

(same '("AA" "AA" "AA"))
-> T

This would of course also work with <= and >= without any hassle.

PL/I

*process source xref attributes or(!);
 /*--------------------------------------------------------------------
 * 01.07.2014 Walter Pachl
 *-------------------------------------------------------------------*/
 clist: Proc Options(main);
 Dcl (hbound) Builtin;
 Dcl sysprint Print;
 Dcl abc(3) Char(2) Init('AA','BB','CC');
 Dcl aaa(3) Char(2) Init('AA','AA','AA');
 Dcl acb(3) Char(2) Init('AA','CC','BB');
 Call test('ABC',ABC);
 Call test('AAA',AAA);
 Call test('ACB',ACB);

 test: Procedure(name,x);
 Dcl name Char(*);
 Dcl x(*) Char(*);
 Dcl (all_equal,increasing) Bit(1) Init('1'b);
 Dcl (i,i1) Bin Fixed(31);
 Dcl txt Char(50) Var;
 Do i=1 To hbound(x)-1 While(all_equal ! increasing);
  i1=i+1;
  Select;
    When(x(i1)=x(i)) increasing='0'b;
    When(x(i1)<x(i)) Do;
                     increasing='0'b;
                     all_equal='0'b;
                     End;
    Otherwise /* x(i1)>x(i) */
                     all_equal='0'b;
    End;
  End;
  Select;
    When(all_equal)  txt='all elements are equal';
    When(increasing) txt='elements are in increasing order';
    Otherwise        txt='neither equal nor in increasing order';
    End;
  Put Skip List(name!!': '!!txt);
  End;
  End;
Output:
ABC: elements are in increasing order
AAA: all elements are equal
ACB: neither equal nor in increasing order

Plain English

To decide if some string things are lexically equal:
If the string things are empty, say yes.
Get a string thing from the string things.
Put the string thing's string into a canonical string.
Loop.
If the string thing is nil, say yes.
If the string thing's string is not the canonical string, say no.
Put the string thing's next into the string thing.
Repeat.

To decide if some string things are in ascending order:
If the string things' count is less than 2, say yes.
Get a string thing from the string things.
Put the string thing's next into the string thing.
Loop.
If the string thing is nil, say yes.
If the string thing's string is less than the string thing's previous' string, say no.
Put the string thing's next into the string thing.
Repeat.

PowerShell

Works with: PowerShell version 4.0
function IsAscending ( [string[]]$Array ) { ( 0..( $Array.Count - 2 ) ).Where{ $Array[$_] -le $Array[$_+1] }.Count -eq $Array.Count - 1 }
function IsEqual     ( [string[]]$Array ) { ( 0..( $Array.Count - 2 ) ).Where{ $Array[$_] -eq $Array[$_+1] }.Count -eq $Array.Count - 1 }
 
IsAscending 'A', 'B', 'B', 'C'
IsAscending 'A', 'C', 'B', 'C'
IsAscending 'A', 'A', 'A', 'A'
 
IsEqual     'A', 'B', 'B', 'C'
IsEqual     'A', 'C', 'B', 'C'
IsEqual     'A', 'A', 'A', 'A'
Output:
True
False
True
False
False
True


Prolog

los(["AA","BB","CC"]).
los(["AA","AA","AA"]).
los(["AA","CC","BB"]).
los(["AA","ACB","BB","CC"]).
los(["single_element"]).

lexically_equal(S,S,S).
in_order(G,L,G) :- compare(<,L,G).

test_list(List) :- 
    List = [L|T],
    write('for list '), write(List), nl,
    (foldl(lexically_equal, T, L, _) 
        -> writeln('The items in the list ARE lexically equal') 
        ; writeln('The items in the list are NOT lexically equal')),
    (foldl(in_order, T, L, _) 
        -> writeln('The items in the list ARE in ascending order')
        ; writeln('The items in the list are NOT in ascending order')),
    nl.

test :- forall(los(List), test_list(List)).
Output:
?- test.
for list [AA,BB,CC]
The items in the list are NOT lexically equal
The items in the list ARE in ascending order

for list [AA,AA,AA]
The items in the list ARE lexically equal
The items in the list are NOT in ascending order

for list [AA,CC,BB]
The items in the list are NOT lexically equal
The items in the list are NOT in ascending order

for list [AA,ACB,BB,CC]
The items in the list are NOT lexically equal
The items in the list ARE in ascending order

for list [single_element]
The items in the list ARE lexically equal
The items in the list ARE in ascending order

true.

PureBasic

EnableExplicit
DataSection
  Data.s ~"AA\tAA\tAA\nAA\tBB\tCC\nAA\tCC\tBB\nAA\tACB\tBB\tCC\nsingel_element"
EndDataSection

Macro PassFail(PF)
  If PF : PrintN("Pass") : Else : PrintN("Fail") : EndIf
EndMacro

Macro ProcRec(Proc)
  Define tf1$,tf2$ : Static chk.b : chk=#True
  tf1$=StringField(s$,c,tz$) : tf2$=StringField(s$,c+1,tz$)
  If Len(tf2$) : Proc(s$,tz$,c+1) : EndIf  
EndMacro

Procedure.b IsStringsEqual(s$,tz$=~"\t",c.i=1)
  ProcRec(IsStringsEqual)
  chk & Bool(tf1$=tf2$ Or tf2$="")
  ProcedureReturn chk
EndProcedure

Procedure.b IsStringsAscending(s$,tz$=~"\t",c.i=1)
  ProcRec(IsStringsAscending)
  chk & Bool(tf1$<tf2$ Or tf2$="")  
  ProcedureReturn chk  
EndProcedure

Define t$,sf$,c.i,i.i,PF.b
Read.s t$ : c=CountString(t$,~"\n")
OpenConsole("Compare a list of Strings")
For i=1 To c+1
  sf$=StringField(t$,i,~"\n")
  PrintN("List : "+sf$)
  Print("Lexical test   : ") : PassFail(IsStringsEqual(sf$))
  Print("Ascending test : ") : PassFail(IsStringsAscending(sf$))
  PrintN("")
Next
Input()
Output:
List : AA       AA      AA
Lexical test   : Pass
Ascending test : Fail

List : AA       BB      CC
Lexical test   : Fail
Ascending test : Pass

List : AA       CC      BB
Lexical test   : Fail
Ascending test : Fail

List : AA       ACB     BB      CC
Lexical test   : Fail
Ascending test : Pass

List : singel_element
Lexical test   : Pass
Ascending test : Pass

Python

A useful pattern is that when you need some function of an item in a list with its next item over possibly all items in the list then f(a, nexta) for a, nexta in zip(alist, alist[1:])) works nicely. (Especially if an index is not needed elsewhere in the algorithm).

all(a == nexta for a, nexta in zip(strings, strings[1:])) # All equal
all(a < nexta for a, nexta in zip(strings, strings[1:])) # Strictly ascending

len(set(strings)) == 1  # Concise all equal
sorted(strings, reverse=True) == strings  # Concise (but not particularly efficient) ascending


Equivalently, we can also use additional list arguments with map rather than zip,

and, if we wish, pass functional forms of standard operators to either of them:

from operator import (eq, lt)


xs = ["alpha", "beta", "gamma", "delta", "epsilon", "zeta",
      "eta", "theta", "iota", "kappa", "lambda", "mu"]

ys = ["alpha", "beta", "gamma", "delta", "epsilon", "zeta",
      "eta", "theta", "iota", "kappa", "lambda", "mu"]

az = sorted(xs)

print (
    all(map(eq, xs, ys)),

    all(map(lt, xs, xs[1:])),

    all(map(lt, az, az[1:]))
)
Output:
True False True

Quackery

Idiomatically the strings would be stored in a nest which need not be named. The words allthesame and allinorder both take a nest of strings from the stack and return a boolean.

The word $> compares two strings using the QACSFOT lexical ordering. (QACSFOT - Quackery Arbitrary Character Sequence For Ordered Text. It is less arbitrary than the ASCII sequence.)

  [ [ true swap 
      dup size 1 > while
      behead swap
      witheach 
        [ over != if
            [ dip not conclude ] ] ]
      drop ]                         is allthesame ( [ --> b )

  [ [ true swap 
      dup size 1 > while
      behead swap
      witheach 
        [ tuck $> if
            [ dip not conclude ] ] ]
    drop ]                           is allinorder ( [ --> b )

R

We can test, first, whether all elements of vector `strings` are equal to the first element; and, second, whether the sorted order of the vector is equal to the original vector.

all(strings == strings[1])
all(strings == sort(strings))

Testing:

manyStrings=list(
  "a",
  c("a", "b", "c"),
  c("a", "c", "b"),
  c("A", "A"),
  c("a", "A"), 
  c(123, "A", "Aaron", "beryllium", "z"),
  c(123, "A", "z", "Aaron", "beryllium", "z")
)

for (strings in manyStrings) {
  print(strings)
  print(all(strings == strings[1]))
  print(all(strings == sort(strings)))
}

Result:

"a"
TRUE
TRUE
"a" "b" "c"
FALSE
TRUE
"a" "c" "b"
FALSE
FALSE
"A" "A"
TRUE
TRUE
"a" "A"
FALSE
TRUE
"123"       "A"         "Aaron"     "beryllium" "z"        
FALSE
TRUE
"123"       "A"         "z"         "Aaron"     "beryllium" "z"        
FALSE
FALSE

For `NULL` input returns `TRUE` to both tests, for all missing (`NA`) input returns `NA` to first test, `TRUE` to second.

Racket

Racket mostly has this... see documentation of string=? and string<?.

There are two small issues:

  • Racket will not cope with comparing less than 2 strings
  • also string=? and string<? take variable arguments, so the list has to be applyed to the functions

Hence the wrapper in the code below:

#lang racket/base
(define ((list-stringX? stringX?) strs)
  (or (null? strs) (null? (cdr strs)) (apply stringX? strs)))
(define list-string=? (list-stringX? string=?))
(define list-string<? (list-stringX? string<?))

(module+ test
  (require tests/eli-tester)
  (test
   (list-string=? '()) => #t
   (list-string=? '("a")) => #t
   (list-string=? '("a" "a")) => #t
   (list-string=? '("a" "a" "a")) => #t
   (list-string=? '("b" "b" "a")) => #f)
  
  (test
   (list-string<? '()) => #t
   (list-string<? '("a")) => #t
   (list-string<? '("a" "b")) => #t
   (list-string<? '("a" "a")) => #f
   (list-string<? '("a" "b" "a")) => #f
   (list-string<? '("a" "b" "c")) => #t))

Raku

(formerly Perl 6)

In Raku, putting square brackets around an infix operator turns it into a listop that effectively works as if the operator had been put in between all of the elements of the argument list (or in technical terms, it folds/reduces the list using that operator, while taking into account the operator's inherent associativity and identity value):

[eq] @strings  # All equal
[lt] @strings  # Strictly ascending

Red

Red []

list1: ["asdf" "Asdf" "asdf"]
list2: ["asdf" "bsdf" "asdf"]
list3: ["asdf" "asdf" "asdf"]

all-equal?: func [list][   1 = length? unique/case list  ]
sorted?: func [list][   list == sort/case copy list ]  ;; sort without copy would modify list !

print all-equal? list1
print sorted? list1

print all-equal? list2
print sorted? list2

print all-equal? list3
print sorted? list3
Output:
false
false
false
false
true
true

REXX

version 1

/* REXX ---------------------------------------------------------------
* 28.06.2014 Walter Pachl
*--------------------------------------------------------------------*/
Call mklist 'ABC','AA','BB','CC'
Call test 'ABC'
Call mklist 'AAA','AA','AA','AA'
Call mklist 'ACB','AA','CC','BB'
Call test 'AAA'
Call test 'ACB'
Exit

mklist:
  list=arg(1)
  do i=1 by 1 To arg()-1
    call value list'.'i,arg(i+1)
    End
  Call value list'.0',i-1
  Return

test:
Parse Arg list
all_equal=1
increasing=1
Do i=1 To value(list'.0')-1 While all_equal | increasing
  i1=i+1
  Select
    When value(list'.i1')==value(list'.i') Then increasing=0
    When value(list'.i1')<<value(list'.i') Then Do
                                                all_equal=0
                                                increasing=0
                                                End
    When value(list'.i1')>>value(list'.i') Then all_equal=0
    End
  End
Select
  When all_equal Then
    Say 'List' value(list)': all elements are equal'
  When increasing Then
    Say 'List' value(list)': elements are in increasing order'
  Otherwise
    Say 'List' value(list)': neither equal nor in increasing order'
  End
Return
Output:
List ABC: elements are in increasing order
List AAA: all elements are equal
List ACB: neither equal nor in increasing order

version 2

Programming note:   If a caseless compare (case insensitive) is desired, the two

  • parse arg x       (on lines 14   &   20)

REXX statements could be replaced with either of   (they're equivalent):

  • parse upper arg x
  • arg x
/*REXX program compares a list of  (character) strings for:   equality,  all ascending. */
@.1= 'ayu dab dog gar panda tui yak'             /*seven strings: they're all ascending.*/
@.2= 'oy oy oy oy oy oy oy oy oy oy'             /*  ten strings:         all equal.    */
@.3= 'somehow   somewhere  sometime'             /*three strings:   ¬equal,  ¬ascending.*/
@.4= 'Hoosiers'                                  /*only a single string is defined.     */
@.5=                                             /*Null.   That is,  no strings here.   */
         do j=1  for 5;    say;   say            /* [↓]  traipse through all the lists. */
         say center(' '@.j, 50, "═")             /*display a centered title/header.     */
         if ifEqual( @.j)  then  say 'strings are all equal.'
         if ifAscend(@.j)  then  say 'strings are ascending.'
         end   /*j*/
exit                                             /*stick a fork in it,  we're all done. */
/*──────────────────────────────────────────────────────────────────────────────────────*/
ifEqual:  procedure; parse arg strings           /*set  STRINGS  to a string in the list*/
            do k=2  to words(strings)            /*scan the strings in the list.        */
            if word(strings,k)\==word(strings,k-1)  then return 0        /*string=prev? */
            end   /*k*/                          /* [↑]     0=false,   [↓] 1=true.      */
          return 1                               /*indicate that all strings are equal. */
/*──────────────────────────────────────────────────────────────────────────────────────*/
ifAscend: procedure; parse arg strings           /*set  STRINGS  to a string in the list*/
            do k=2  to words(strings)            /*scan the strings in the list.        */
            if word(strings,k)<<=word(strings,k-1)  then return 0        /*string>prev? */
            end   /*k*/                          /*  [↑]    0=false,     [↓]    1=true. */
          return 1                               /*indicate that strings are ascending. */
output   when using the supplied lists:
══════════ ayu dab dog gar panda tui yak══════════
  The strings are ascending.


══════════ oy oy oy oy oy oy oy oy oy oy══════════
  The strings are all equal.


══════════ somehow   somewhere  sometime══════════


════════════════════ Hoosiers═════════════════════
  The strings are all equal.
  The strings are ascending.


════════════════════════ ═════════════════════════
  The strings are all equal.
  The strings are ascending.

version 3

This REXX version is more idiomatic.

/*REXX program compares a list of strings for:  equality, all ascending.                */
@.1= 'ayu dab dog gar panda tui yak'             /*seven strings: they're all ascending.*/
@.2= 'oy oy oy oy oy oy oy oy oy oy'             /*  ten strings:         all equal.    */
@.3= 'somehow   somewhere  sometime'             /*three strings:   ¬equal,  ¬ascending.*/
@.4= 'Hoosiers'                                  /*only a single string is defined.     */
@.5=                                             /*Null.   That is,  no strings here.   */
#= 5;         do j=1  for #;    say;   say       /* [↓]  traipse through all the lists. */
              say center(' '@.j, 50, "═")        /*display a centered title/header.     */
              if cStr(@.j, 'Equal'    )  then  say  "  The strings are all equal."
              if cStr(@.j, 'Ascending')  then  say  "  The strings are ascending."
              end   /*j*/
exit                                             /*stick a fork in it,  we're all done. */
/*──────────────────────────────────────────────────────────────────────────────────────*/
cStr: procedure; parse arg x;  arg , how 2       /*set X to list; get 1st char of arg #2*/
              do k=2  to words(x)                /*scan the strings in the list.        */
              if how=='E'  then if word(x,k) \== word(x,k-1)  then return 0   /*¬=prev.?*/
              if how=='A'  then if word(x,k) <<= word(x,k-1)  then return 0   /*≤ prev.?*/
              end   /*k*/                        /* [↓]   1=true.        [↑]   0=false. */
      return 1                                   /*indicate strings have true comparison*/
output   is identical to the above REXX version.



RPL

Works with: HP version 48G
IF DUP SIZE 2 < THEN 1 ELSE ≪ == ≫ DOSUBS ΠLIST END
≫ ‘ALLSAME?' STO

≪ DUP SORT ==
≫ ‘ALLORDERED?' STO

Ruby

strings.uniq.one?                 # all equal?
strings == strings.uniq.sort      # ascending?

Short circuiting:

strings.all?{|str| str == strings.first} # all equal?
strings.each_cons(2).all?{|str1, str2| str1 < str2} # ascending?

Rust

fn strings_are_equal(seq: &[&str]) -> bool {
    match seq {
        &[] | &[_] => true,
        &[x, y, ref tail @ ..] if x == y => strings_are_equal(&[&[y], tail].concat()),
        _ => false
    }
}

fn asc_strings(seq: &[&str]) -> bool {
    match seq {
        &[] | &[_] => true,
        &[x, y, ref tail @ ..] if x < y => asc_strings(&[&[y], tail].concat()),
        _ => false
    }
}

S-lang

"Simple Loop" and "Array Idiomatic" versions:

define equal_sl(sarr)
{
  variable n = length(sarr), a0, i;
  if (n < 2) return 1;  

  a0 = sarr[0];
  _for i (1, length(sarr)-1, 1)
    if (sarr[i] != a0) return 0;

  return 1;
}
define ascending_sl(sarr) {
  variable n = length(sarr), a0, i;
  if (n < 2) return 1;  

  _for i (0, length(sarr)-2, 1)
    if (sarr[i] >= sarr[i+1]) return 0;

  return 1;
}


define equal_ai(sarr) {
  if (length(sarr) < 2) return 1;
  variable s0 = sarr[0];
  return all(sarr[[1:]] == s0);
}

define ascending_ai(sarr) {
  variable la = length(sarr);
  if (la < 2) return 1;  
  return all(sarr[[0:la-2]] < sarr[[1:la-1]]);
}

define atest(a) {
  () = printf("\n");
  print(a);

  () = printf("equal_sl=%d, ascending_sl=%d\n",
              equal_sl(a), ascending_sl(a));
  () = printf("equal_ai=%d, ascending_ai=%d\n",
              equal_ai(a), ascending_ai(a));
}
              
atest(["AA","BB","CC"]);
atest(["AA","AA","AA"]);
atest(["AA","CC","BB"]);
atest(["AA","ACB","BB","CC"]);
atest(["single_element"]);
atest(NULL);
Output:
"AA"
"BB"
"CC"
equal_sl=0, ascending_sl=1
equal_ai=0, ascending_ai=1

"AA"
"AA"
"AA"
equal_sl=1, ascending_sl=0
equal_ai=1, ascending_ai=0

"AA"
"CC"
"BB"
equal_sl=0, ascending_sl=0
equal_ai=0, ascending_ai=0

"AA"
"ACB"
"BB"
"CC"
equal_sl=0, ascending_sl=1
equal_ai=0, ascending_ai=1

"single_element"
equal_sl=1, ascending_sl=1
equal_ai=1, ascending_ai=1

NULL
equal_sl=1, ascending_sl=1
equal_ai=1, ascending_ai=1

Scala

Functions implemented in Scala following a functional paradigm

def strings_are_equal(seq:List[String]):Boolean = seq match {
    case Nil => true
    case s::Nil => true
    case el1 :: el2 :: tail => el1==el2 && strings_are_equal(el2::tail)
}

def asc_strings(seq:List[String]):Boolean = seq match {
    case Nil => true
    case s::Nil => true
    case el1 :: el2 :: tail => el1.compareTo(el2) < 0
}
Output:

'''Sample tests:'''

scala> strings_are_equal(List("asdf"))
res3: Boolean = true

scala> strings_are_equal(List("asdf","asdf","sf"))
res5: Boolean = false

scala> asc_strings(List())
res10: Boolean = true

scala> asc_strings(List("asdfas","fds"))
res11: Boolean = true

scala> asc_strings(List("sdfa","asfsdf","afas","asf"))
res8: Boolean = false

Scheme

For known lists that are 'short-enough', the simplest solution uses 'apply', but that relies on the list being shorter than the maximum number of arguments a function can accept. Better is to write a simple loop:

(define (compare-strings fn strs)
  (or (null? strs)                             ; returns #t on empty list
      (null? (cdr strs))                       ; returns #t on list of size 1
      (do ((fst strs (cdr fst))
           (snd (cdr strs) (cdr snd)))
        ((or (null? snd)
             (not (fn (car fst) (car snd))))
         (null? snd)))))                       ; returns #t if the snd list is empty, meaning all comparisons are exhausted

(compare-strings string=? strings) ; test for all equal
(compare-strings string<? strings) ; test for in ascending order

Seed7

$ include "seed7_05.s7i";

const func boolean: allTheSame (in array string: strings) is func
  result
    var boolean: allTheSame is TRUE;
  local
    var integer: index is 0;
  begin
    for index range 2 to length(strings) until not allTheSame do
      if strings[pred(index)] <> strings[index] then
        allTheSame := FALSE;
      end if;
    end for;
  end func;

const func boolean: strictlyAscending (in array string: strings) is func
  result
    var boolean: strictlyAscending is TRUE;
  local
    var integer: index is 0;
  begin
    for index range 2 to length(strings) until not strictlyAscending do
      if strings[pred(index)] >= strings[index] then
        strictlyAscending := FALSE;
      end if;
    end for;
  end func;

SenseTalk

analyze ["AA","BB","CC"]
analyze ["AA","AA","AA"]
analyze ["AA","CC","BB"]
analyze ["AA","ACB","BB","CC"]
analyze ["single_element"]

to analyze aList
	put "List: " & aList
	put "   " & (if allEqual(aList) then "IS" else "Is NOT") && "all equal"
	put "   " & (if isAscending(aList) then "IS" else "Is NOT") && "strictly ascending"
end analyze

to handle allEqual strings
	return the number of items in the unique items of strings is less than 2
end allEqual

to handle isAscending strings
	repeat with n = 2 to the number of items in strings
		if item n of strings isn't more than item n-1 of strings then
			return False
		end if
	end repeat
	return True
end isAscending
Output:
List: ["AA","BB","CC"]
   Is NOT all equal
   IS strictly ascending
List: ["AA","AA","AA"]
   IS all equal
   Is NOT strictly ascending
List: ["AA","CC","BB"]
   Is NOT all equal
   Is NOT strictly ascending
List: ["AA","ACB","BB","CC"]
   Is NOT all equal
   IS strictly ascending
List: ["single_element"]
   IS all equal
   IS strictly ascending

Sidef

Short-circuiting:

1..arr.end -> all{ arr[0] == arr[_] }   # all equal
1..arr.end -> all{ arr[_-1] < arr[_] }  # strictly ascending

Non short-circuiting:

arr.uniq.len == 1      # all equal
arr == arr.uniq.sort   # strictly ascending

Tailspin

Note that we choose here to use 1 as true and 0 as false since Tailspin doesn't (yet?) have booleans

// matcher testing if the array contains anything not equal to the first element
templates allEqual
  when <[](..1)> do 1 !
  when <[<~=$(1)>]> do 0 !
  otherwise 1 !
end allEqual

templates strictAscending
  def a: $;
  1 -> #
  when <$a::length..> do 1 !
  when <?($a($) <..~$a($+1)>)> do $ + 1 -> #
  otherwise 0 !
end strictAscending

// Of course we could just use the same kind of loop for equality
templates strictEqual
  def a: $;
  1 -> #
  when <$a::length..> do 1 !
  when <?($a($) <=$a($+1)>)> do $ + 1 -> #
  otherwise 0 !
end strictEqual

Tcl

The command form of the eq and < operators (introduced in Tcl 8.5) handle arbitrarily many arguments and will check if they're all equal/ordered. Making the operators work with a list of values is just a matter of using the expansion syntax with them.

tcl::mathop::eq {*}$strings;		# All values string-equal
tcl::mathop::< {*}$strings;		# All values in strict order

Transd

#lang transd

MainModule: {
    _start: (λ 
        (for v in [["aa","ab","ad","ae"],["ab","ab","ab","ab"]] do
            (lout :boolalpha v)
            (lout (not (any v (λ (ret (neq @it (get v 0)))))))
            (lout (not (any Range(in: v 1 -0) 
                (λ (ret (leq @it (get v (- @idx 1))))))) "\n")
        )
    )
}
Output:
["aa", "ab", "ad", "ae"]
false
true

["ab", "ab", "ab", "ab"]
true
false

VBA

Private Function IsEqualOrAscending(myList) As String
Dim i&, boolEqual As Boolean, boolAsc As Boolean

    On Error Resume Next
    If UBound(myList) > 0 Then
        If Err.Number > 0 Then
            IsEqualOrAscending = "Error " & Err.Number & " : Empty array"
            On Error GoTo 0
            Exit Function
        Else
            For i = 1 To UBound(myList)
                If myList(i) <> myList(i - 1) Then boolEqual = True
                If myList(i) <= myList(i - 1) Then boolAsc = True
            Next
        End If
    End If
    IsEqualOrAscending = "List : " & Join(myList, ",") & ", IsEqual : " & (Not boolEqual) & ", IsAscending : " & Not boolAsc
End Function

Call :

Sub Main()
Dim List
    Debug.Print IsEqualOrAscending(Array("AA", "BB", "CC"))
    Debug.Print IsEqualOrAscending(Array("AA", "AA", "AA"))
    Debug.Print IsEqualOrAscending(Array("AA", "CC", "BB"))
    Debug.Print IsEqualOrAscending(Array("AA", "ACB", "BB", "CC"))
    Debug.Print IsEqualOrAscending(Array("single_element"))
    Debug.Print IsEqualOrAscending(Array("AA", "BB", "BB"))
    'test with Empty Array :
    Debug.Print IsEqualOrAscending(List)
End Sub
Output:
List : AA,BB,CC, IsEqual : False, IsAscending : True
List : AA,AA,AA, IsEqual : True, IsAscending : False
List : AA,CC,BB, IsEqual : False, IsAscending : False
List : AA,ACB,BB,CC, IsEqual : False, IsAscending : True
List : single_element, IsEqual : True, IsAscending : True
List : AA,BB,BB, IsEqual : False, IsAscending : False
Error 13 : Empty array

VBScript

Function string_compare(arr)
	lexical = "Pass"
	ascending = "Pass"
	For i = 0 To UBound(arr)
		If i+1 <= UBound(arr) Then
			If arr(i) <> arr(i+1) Then
				lexical = "Fail"
			End If
			If arr(i) >= arr(i+1) Then
				ascending = "Fail"
			End If 
		End If	
	Next
	string_compare = "List: " & Join(arr,",") & vbCrLf &_
					 "Lexical Test: " & lexical & vbCrLf &_
					 "Ascending Test: " & ascending & vbCrLf
End Function

WScript.StdOut.WriteLine string_compare(Array("AA","BB","CC"))
WScript.StdOut.WriteLine string_compare(Array("AA","AA","AA"))
WScript.StdOut.WriteLine string_compare(Array("AA","CC","BB"))
WScript.StdOut.WriteLine string_compare(Array("AA","ACB","BB","CC"))
WScript.StdOut.WriteLine string_compare(Array("FF"))
Output:
List: AA,BB,CC
Lexical Test: Fail
Ascending Test: Pass

List: AA,AA,AA
Lexical Test: Pass
Ascending Test: Fail

List: AA,CC,BB
Lexical Test: Fail
Ascending Test: Fail

List: AA,ACB,BB,CC
Lexical Test: Fail
Ascending Test: Pass

List: FF
Lexical Test: Pass
Ascending Test: Pass

V (Vlang)

Translation of: go
fn all_equal(strings []string) bool {
	for s in strings {
		if s != strings[0] {
			return false
		}
	}
	return true
}
 
fn all_less_than(strings []string) bool {
	for i := 1; i < strings.len(); i++ {
		if !(strings[i - 1] < s) {
			return false
		}
	}
	return true
}

Wren

Library: Wren-sort
import "./sort" for Sort

var areEqual = Fn.new { |strings|
    if (strings.count < 2) return true
    return (1...strings.count).all { |i| strings[i] == strings[i-1] }
}

var areAscending = Fn.new { |strings| Sort.isSorted(strings) }
    
var a = ["a", "a", "a"]
var b = ["a", "b", "c"]
var c = ["a", "a", "b"]
var d = ["a", "d", "c"]
System.print("%(a) are all equal : %(areEqual.call(a))")
System.print("%(b) are ascending : %(areAscending.call(b))")
System.print("%(c) are all equal : %(areEqual.call(c))")
System.print("%(d) are ascending : %(areAscending.call(d))")
Output:
[a, a, a] are all equal : true
[a, b, c] are ascending : true
[a, a, b] are all equal : false
[a, d, c] are ascending : false

XPL0

include xpllib; \For StrCmp

func AreAllEqual(Strings, Size);
int  Strings, Size, I;
[for I:= 1 to Size-1 do
    if StrCmp(Strings(I), Strings(0)) # 0 then return false;
return true;
];

func AreAscending(Strings, Size;
int  Strings, Size, I;
[for I:= 1 to Size-1 do
    if StrCmp(Strings(I-1), Strings(I)) >= 0 then return false;
return true;
];

int A, B, C, D;
[A:= ["a", "a", "a"];
 B:= ["a", "b", "c"];
 C:= ["a", "a", "b"];
 D:= ["a", "d", "c"];
Text(0, if AreAllEqual (A, 3) then "true" else "false");  CrLf(0);
Text(0, if AreAscending(B, 3) then "true" else "false");  CrLf(0);
Text(0, if AreAllEqual (C, 3) then "true" else "false");  CrLf(0);
Text(0, if AreAscending(D, 3) then "true" else "false");  CrLf(0);
]
Output:
true
true
false
false

XProfan

Proc allsame
   Parameters long liste
   var int result = 1
   var int cnt = GetCount(liste)
   Case cnt == 0 : Return 0
   Case cnt == 1 : Return 1
   WhileLoop 1, cnt-1
      If GetString$(liste,&loop - 1) <> GetString$(liste,&loop)
         result = 0
         BREAK
      EndIf
   EndWhile
   Return result
EndProc

Proc strict_order
   Parameters long liste
   var int result = 1
   var int cnt = GetCount(liste)
   Case cnt == 0 : Return 0
   Case cnt == 1 : Return 1
   WhileLoop 1, cnt-1
      If GetString$(liste,&loop) <= GetString$(liste,&loop - 1)
         result = 0
         BREAK
      EndIf
   EndWhile
   Return result
EndProc

cls
declare string s[4]
s[0] = "AA,BB,CC"
s[1] = "AA,AA,AA"
s[2] = "AA,CC,BB"
s[3] = "AA,ACB,BB,CC"
s[4] = "single_element"

WhileLoop 0,4
   ClearList 0
   Move("StrToList",s[&loop],",")
   Print "list:",s[&loop]
   Print "...is " + if(allsame(0), "", "not ") + "lexically equal"
   Print "...is " + if(strict_order(0), "", "not ") + "in strict ascending order"
EndWhile

ClearList 0
WaitKey
end
Output:
list: AA,BB,CC
...is not lexically equal
...is in strict ascending order
list: AA,AA,AA
...is lexically equal
...is not in strict ascending order
list: AA,CC,BB
...is not lexically equal
...is not in strict ascending order
list: AA,ACB,BB,CC
...is not lexically equal
...is in strict ascending order
list: single_element
...is lexically equal
...is in strict ascending order

zkl

These short circuit.

fcn allEQ(strings){ (not strings.filter1('!=(strings[0]))) }
fcn monoUp(strings){
   strings.len()<2 or
   strings.reduce(fcn(a,b){ if(a>=b) return(Void.Stop,False); b }).toBool() 
}
allEQ(T("AA")).println();                //True
allEQ(T("AA","AA","AA","AA")).println(); //True
allEQ(T("A", "AA","AA","AA")).println(); //False

monoUp(T("a")).println();                   //True
monoUp(T("a","aa","aaa","aaaa")).println(); //True
monoUp(T("a","aa","aaa","aaa")).println();  //False
monoUp(T("a","b","c","cc")).println();      //True

zonnon

module CompareStrings;
type
	Vector = array * of string;
var
	v,w: Vector;
	i: integer;
	all,ascending: boolean;
begin
	v := new Vector(3);
	v[0] := "uno";
	v[1] := "uno";
	v[2] := "uno";

	all := true;
	for i := 1 to len(v) - 1 do
		all := all & (v[i - 1] = v[i]);
	end;

	w := new Vector(3);
	w[0] := "abc";
	w[1] := "bcd";
	w[2] := "cde";
	v := w;
	ascending := true;
	for i := 1 to len(v) - 1 do
		ascending := ascending & (v[i - 1] <= v[i])
	end;

	write("all equals?: ");writeln(all);
	write("ascending?: ");writeln(ascending)
end CompareStrings.

ZX Spectrum Basic

Translation of: AWK
10 FOR j=160 TO 200 STEP 10
20 RESTORE j
30 READ n
40 LET test1=1: LET test2=1
50 FOR i=1 TO n
60 READ a$
70 PRINT a$;" ";
80 IF i=1 THEN GO TO 110
90 IF p$<>a$ THEN LET test1=0
100 IF p$>=a$ THEN LET test2=0
110 LET p$=a$
120 NEXT i
130 PRINT 'test1'test2
140 NEXT j
150 STOP 
160 DATA 3,"AA","BB","CC"
170 DATA 3,"AA","AA","AA"
180 DATA 3,"AA","CC","BB"
190 DATA 4,"AA","ACB","BB","CC"
200 DATA 1,"single_element"
Cookies help us deliver our services. By using our services, you agree to our use of cookies.