Keep duplicates from array

[blumi]

How to keep the duplicates from an array?

[1,1,2,3,3,3,3,4,5,5]

the Result should be

[1,3,5]

[rootx]

https://www.autoitscript.com/autoit3/docs/libfunctions/_ArrayUnique.htm

[jchd]

_ArrayUnique isn't the right answer here, since the OP wants to keep only a single instance of all entries which have dups in the input array. _ArrayUnique would include 4, which is single, hence shouldn't be selected.

@blumi, I'd convert the array into a Map datatype (supported in AutoIt beta) keeping the count of multiple instances and converting back entries with count > 1 into an output array. This will work faster than scanning the array in O² and runtime will remain unaffected by array order.

AutoIt beta has been out for long enough to be considered stable eough for practical purposes. If you're still reluctant about beta, use a scripting dictionary object instead.

[rootx]

I'm sorry I have read to fast!!! Your solution..

#include<array.au3>

Local $aArray[10]    = [1,1,2,3,3,3,3,4,5,5]

for $i = ubound($aArray) - 1 to 0 step -1
        $aFound = _ArrayFindAll($aArray , $aArray[$i])
        $i -= ubound($aFound) - 1
        If ubound($aFound) > 1 Then
            ConsoleWrite("This Value "&$aArray[$aFound[0]] & " Is duplicate N Times " & ubound($aFound)&@CRLF)
        EndIf
next

 

[jchd]

Correct, but _ArrayFindAll runs in O². That won't make a difference in small samples but will explode on serious input.

[Skysnake]

For large volume data, use a SQLite memory db?  

CREATE TABLE `mytable` (
    `value` integer,
    `id`    integer PRIMARY KEY AUTOINCREMENT
);
------------------------
-- [1,1,2,3,3,3,3,4,5,5]
insert into mytable (value)
values (1)
    , (1)
    , (2)
    , (3)
    , (3)
    , (3)
    , (3)
    , (4)
    , (5)
    , (5)
    ;
-----------------------------   
select value, count(value) from mytable group by value having count(value) > 1;

Returns

value count(value)
"1"    "2"
"3"    "4"
"5"    "2"
3 rows returned in 18ms from: select value, count(value) from mytable group by value having count(value) > 1;

 

[mikell]

?

#include <Array.au3>

Local $a[11]    = [1,5,2,3,3,3,3,4,1,5,6]

_ArraySort($a)
Local $b[UBound($a)], $n
For $i = 0 to UBound($a)-2
    If $a[$i] = $a[$i+1] Then
         $b[$n] = $a[$i]
         $n += 1
    EndIf
Next
ReDim $b[$n]
$b = _ArrayUnique($b, 0, 0, 0, 0)
_ArrayDisplay($b)

 

[iamtheky]

dont know how it compares to the redim penalty, but here is an array dance that does it in one loop

#include<array.au3>

Local $aArray[10]    = [1,1,2,3,3,3,3,4,5,5]

_ArraySort($aArray)

For $i = ubound($aArray) - 1 to 1 step -1

        If $aArray[$i] = $aArray[$i - 1] Then
            _ArrayDelete($aArray , $i)
            $i -= 1
            If $i = 0 then exitloop
            $flag = 1
        EndIf

        If $aArray[$i] <> $aArray[$i - 1] And $flag = 0 Then
            _ArrayDelete($aArray , $i)
        ElseIf $aArray[$i] <> $aArray[$i - 1] And $flag = 1 Then
            $flag = 0
        EndIf

        If $flag = 1 then $i += 1
Next

_ArrayDisplay($aArray , $i)

 

[iamtheky]

Also, this  ($2 contains the non-duplicates)

Local $aArray[10]    = [1,1,2,3,3,3,3,4,5,5]
$flag = 0

$str = StringRegExpReplace(_ArrayToString($aArray , "") , "(?:(.)\1+)+|(.)" , "$1")

msgbox(0 , '' , $str)

 

Edited December 10, 2017 by iamtheky

[kylomas]

iamthekey,

Try it like this...

#include <array.au3>

Local $aArray[20]    = [1,1,2,3,3,3,3,4,5,5,6,6]

$str = StringRegExpReplace(_ArrayToString($aArray , "") , "(?:(.)\1+)+|(.)" , "$1")

msgbox(0 , '' , $str)

Trying to understand the pattern to see where it is failing.

kylomas

[iamtheky]

its a straight copy/paste from https://stackoverflow.com/questions/29480579/regex-finding-non-duplicate-characters

I have no idea why it works/doesnt work either.

**And its all sorts of broke, if you add another 4 to your array it only returns 1 and 6

Edited December 10, 2017 by iamtheky

[iamtheky]

awww shit, think i fixed it...

;~ Local $aArray[20]    = [1,1,2,2,2,3,3,3,3,4,4,4,4,4,4,5,6,6]

;~ Local $aArray[20]    = [1,1,2,3,3,3,3,4,5,5,6,6]

Local $aArray[20]    = ["a",1,2,3,3,3,3,"a",4,5,5,6,6]

_ArraySort($aArray)

$str = StringRegExpReplace(_ArrayToString($aArray , "") , "(.).*\1|(.)" , "$1")

edit: bear in mind this only works for sorted arrays, with single character values, or rather the first character of entries. (take off the sort and the last example returns undesired values).  someone smart can probably make it more extensible.

Edited December 10, 2017 by iamtheky

[LarsJ]

For the general solution (not limited to integers, not limited to single character values, not limited to sorted arrays) you'll not find anything faster than this. Try yourself.

#include <Array.au3>

Global $aArray[10] = [1,1,2,3,3,3,3,4,5,5]
Global $aRes = Example( $aArray )
_ArrayDisplay( $aRes )


Global $iItems = 100000
ReDim $aArray[$iItems]
For $i = 0 To $iItems - 1 Step 2
  $aArray[$i] = $i/2
  $aArray[$i+1] = $i/2
Next
_ArrayDisplay( $aArray )
$aRes = Example( $aArray )
_ArrayDisplay( $aRes )


Func Example( $aArray )
  Local $oTst = ObjCreate( "Scripting.Dictionary" )
  Local $oRes = ObjCreate( "Scripting.Dictionary" )

  For $i = 0 To UBound( $aArray ) - 1
    If Not $oTst.Exists( $aArray[$i] ) Then
      $oTst( $aArray[$i] ) = 1
    ElseIf Not $oRes.Exists( $aArray[$i] ) Then
      $oRes( $aArray[$i] ) = 1
    EndIf
  Next

  Return $oRes.Keys()
EndFunc

 

[jchd]

Maps are faster:

expandcollapse popup

#include <Array.au3>

Global $iItems = 100000
Global $aArray[$iItems]
For $i = 0 To $iItems - 1
  $aArray[$i] = Random(0, $iItems * 0.2, 1)
Next
Local $t, $aRes
_ArrayDisplay( $aArray )

$t = TimerInit()
$aRes = Example( $aArray )
ConsoleWrite(TimerDiff($t) & @LF)
_ArrayDisplay( $aRes )


$t = TimerInit()
$aRes = MapExample( $aArray )
ConsoleWrite(TimerDiff($t) & @LF)
_ArrayDisplay( $aRes )

Func Example( ByRef $aArray )
  Local $oTst = ObjCreate( "Scripting.Dictionary" )
  Local $oRes = ObjCreate( "Scripting.Dictionary" )

  For $i = 0 To UBound( $aArray ) - 1
    If Not $oTst.Exists( $aArray[$i] ) Then
      $oTst( $aArray[$i] ) = 1
    ElseIf Not $oRes.Exists( $aArray[$i] ) Then
      $oRes( $aArray[$i] ) = 1
    EndIf
  Next

  Return $oRes.Keys()
EndFunc

Func MapExample(ByRef $aArray)
  Local $mRes[]

  For $i = 0 To UBound($aArray) - 1
    If MapExists($mRes, $aArray[$i]) Then
      $mRes[$aArray[$i]] += 1
    Else
      $mRes[$aArray[$i]] = 1
    EndIf
  Next
  For $i = 0 To UBound($aArray) - 1
    If $mRes[$aArray[$i]] = 1 Then
      MapRemove($mRes, $aArray[$i])
    EndIf
  Next

  Return $mRes.Keys()
EndFunc

 

[Champak]

I'm looking in the help file (v3.3.16.0) and it still says maps are experimental and may be removed and this is a 5 year old thread. I would think after 5 years it's no longer experimental. Before I use this, was that overlooked and it's good to go, or is the map feature still not fleshed out?

[jchd]

AFAICT all quirks have been fixed long ago.

[Champak]

thanks

[Champak]

I've been playing with this and can't figure a way to ALSO return triplicates, and not increase the time if only looking for duplicates? I have a search function that has an option to filter the search with 2 or 3 conditions, so sometimes I need it to return based on 3 similar indexes being found. Right now I'm just running the function twice, although it is still relatively fast, I'm hoping it can be combined to reduce the time.

Func _MapSearch(ByRef $aArray, $bDupeReturn = False)
    Local $mRes[]

    For $i = 0 To UBound($aArray) - 1
        If MapExists($mRes, $aArray[$i]) Then
            $mRes[$aArray[$i]] += 1
        Else
            $mRes[$aArray[$i]] = 1
        EndIf
    Next

    If $bDupeReturn = False Then

        For $i = 0 To UBound($aArray) - 1
            If $mRes[$aArray[$i]] = 1 Then
                MapRemove($mRes, $aArray[$i])
            EndIf
        Next

    Else

        For $i = 0 To UBound($aArray) - 1
            If $mRes[$aArray[$i]] <> 1 Then
                MapRemove($mRes, $aArray[$i])
            EndIf
        Next

    EndIf

    Return $mRes.Keys()

EndFunc