pseudo logic suggestions (Solved)

kylomas · December 16, 2011

J1,

This is providing a distraction from the boring shit that I was working on...The solutions offered so far do not seem complete to me.

Please post your solution for comparison (learning) as I'm also coding something to solve this...

Incidentally, is it true that each part of the filename (before and after the "_") are the same length within one file but may vary from file to file?

kylomas

JohnOne · December 16, 2011

I have not done anything with it yet, been eating and watching a TV show. But I'm certain

that bit of logic will get me underway.

The two sides of _ are of varying length at their are no rules/pattern governing that.

With the code posted by bothose, it will give me the first true file it finds. I will then remove all instances of files

that are equal to the first part Or equal to the second, and loop it.

kylomas · December 16, 2011

J1,

Exactly what I am working on now...

kylomas

jaberwacky · December 16, 2011

J1, we are all awash in a quagmire of uncertainties. When you're buried under an avalanche of snow the best to determine your orientation is to spit. If the spit dribbles into your eye then you're upside down.

iamtheky · December 16, 2011

also bear in mind, my script returns all duplicates under the assumption that only the proper filename will be listed multiple times. So if there is a chance that these "wrong" numbers could end up having duplicates within the group you are evaluating....the quagmire expands.

JohnOne · December 16, 2011

I don't doubt it, a smart person would use _ArrayUnique on the array first, to be sure they're

not tampering with files that do not belong there.

JohnOne · December 16, 2011

J1, we are all awash in a quagmire of uncertainties. When you're buried under an avalanche of snow the best to determine your orientation is to spit. If the spit dribbles into your eye then you're upside down.

haha

Some of us only have piss and dribble to look forward to, but I prefer mine down my chin.

EDIT:

Dribble that is.

Edited December 16, 2011 by JohnOne

kylomas · December 16, 2011

J1,

The logic that I am persueing is:

"Ladies and gentlemen, take my advice, pull down your pants and slide on the ice."

- Sidney Friedman (MASH)

kylomas

jaberwacky · December 16, 2011

oh lawd I lolled so hard at that =D

edit: that includes kylomas's post now

Edited December 16, 2011 by LaCastiglione

kylomas · December 17, 2011

J1,

Given the following test data (from kaotibliss's post)

$a10[0]= "332333_14631"
$a10[1]= "88888_44444"
$a10[2]= "46588_44444"
$a10[3]= "33333_65788"
$a10[4]= "33333_11112"

Which would you say are good file names?

kylomas

JohnOne · December 17, 2011

One cannot be deduced from those.

1,2,3 and 4 are candidates for half of separate good file names, but neither can be confirmed.

Edited December 17, 2011 by JohnOne

kylomas · December 17, 2011

J1,

Yes, then M23's assertion is true. Can you add data to the above example that would deliniate a good file?

kylomas

JohnOne · December 17, 2011

$a10[0]= "332333_14631"
$a10[1]= "88888_44444"
$a10[2]= "46588_44444"
$a10[3]= "33333_65788"
$a10[4]= "33333_11112"
$a10[5]= "46588_45723"

I have added [5]

which you can use to deduce a good file name.

You see it now?

kylomas · December 17, 2011

J1,

The good file name would be 46588_45723 and not 46588_44444 because 46588_44444 has dup last node?

Also,

Given

"123_321" (good)
"267_876" (good)
"12_98"  (good)
"2223_7881" (good)
"123_946"   (good)
"887_321" (bad)
"456_321" (bad)
"123_998" (good)

Is this correct?

kylomas

Edit: I see your dilema, there is nothing here to distinguish between 123_321 and 887_321....

Edit2: back to M23's assertion...to find bad files we have to uniquely id good files...

Edited December 17, 2011 by kylomas

JohnOne · December 17, 2011

No, its good because both nodes are dupe.

Any other file containing either first or last node

are bad.

I'm working with this at the moment.

#include <File.au3>
#include <Array.au3>
$dir = @ScriptDir & ""
$good = $dir & "good"
$dump = $dir & "dump"
While Not @error
$FArray = _FileListToArray($dir, "*.lmi", 1)
$tmp = Work($FArray) ; the name of the first file found which should be good name
If FileExists($tmp) Then
  FileMove($dir & $tmp, $good & $tmp) ; if it is exist the move it
EndIf
$aTmp = StringSplit($tmp, "_", 3) ;these loops below get rid of other files that have either the first part or secong part in them
For $i = $FArray[0] To 1 Step -1
  If StringInStr($FArray[$i], $aTmp[0]) Then
   FileMove($dir & $FArray[$i], $dump & $FArray[$i], 8)
  EndIf
Next
For $i = $FArray[0] To 1 Step -1
  If StringInStr($FArray[$i], $aTmp[1]) Then
   FileMove($dir & $FArray[$i], $dump & $FArray[$i], 8)
  EndIf
Next
WEnd

Func Work($Array)
Local $column1[$Array[0] + 1]
Local $column2[$Array[0] + 1]
For $i = 0 To UBound($Array) - 1
  $Temp = StringSplit($Array[$i], "_")
  If @error Then ContinueLoop
  $column1[$i] = $Temp[1]
  $column2[$i] = $Temp[2]
Next
Local $FOUND = 0
For $k = UBound($column1) - 1 To 0 Step -1
  For $i = UBound($column1) - 1 To 0 Step -1
   If $column1[$k] = $column1[$i] And $i <> $k Then $FOUND = 1
  Next
  If $FOUND = 1 Then
   $FOUND = 0
  Else
   _ArrayDelete($column1, $k)
  EndIf
Next
For $k = UBound($column2) - 1 To 0 Step -1
  For $i = UBound($column2) - 1 To 0 Step -1
   If $column2[$k] = $column2[$i] And $i <> $k Then $FOUND = 1
  Next
  If $FOUND = 1 Then
   $FOUND = 0
  Else
   _ArrayDelete($column2, $k)
  EndIf
Next
If Not IsArray($column1) Or Not IsArray($column2) Then
  Return SetError(1)
EndIf
$Answer = $column1[0] & "_" & $column2[0]
Return $Answer
EndFunc   ;==>Work

EDIT: tags are a pain.

Edited December 17, 2011 by JohnOne

kaotkbliss · December 17, 2011

No, its good because both nodes are dupe.

That's where I was (quickly) trying to go with my 2 minute code. Get a list of all 1st nodes that were duped, get a list of all 2nd nodes that were duped

Then start piecing them together looking for filenames.

kylomas · December 17, 2011

J1,

This code will create an array of all files that are good. Without further qualifying criteria I cannot figure out how to determine which are "good" files from the list of "might be good" files.

@koatkbliss - Seems like such a trivial thing, at first glance...right...

;
; hammer and tongs method
;
#include <array.au3>
local $a10[10]
local $a20[ubound($a10)][2]
$a10[0]= "332333_14631"
$a10[1]= "88888_44444"
$a10[2]= "46588_44444"
$a10[3]= "33333_65788"
$a10[4]= "33333_11112"
$a10[5]= "46588_45723"
;---------------------------------------
; winnow out files that have to be good
;---------------------------------------
; split file name to 2x array
for $i = 0 to ubound($a10) - 1
if stringlen($a10[$i]) = 0 then continueloop
$a30 = stringsplit($a10[$i],'_',2)
$a20[$i][0] = $a30[0]
$a20[$i][1] = $a30[1]
Next
; find unique first nodes - iterate each element of array through itself - save dups to string
local $save_ele, $save_dups
for $i = 0 to ubound($a20) - 1
$save_ele = $a20[$i][0]
for $j = 0 to ubound($a20) - 1
  if $j <> $i and $a20[$j][0] == $save_ele then $save_dups &= $a20[$j][0] & 0x0
Next
next
; find unique second nodes - iterate each element of array through itself - save dups to string
for $i = 0 to ubound($a20) - 1
$save_ele = $a20[$i][1]
for $j = 0 to ubound($a20) - 1
  if $i <> $j and $a20[$j][1] == $save_ele then $save_dups &= $a20[$j][1] & 0x0
Next
next
; run through array and blank all dups
$a30 = stringsplit($save_dups,0x0,2)
for $i = 0 to ubound($a30) - 1
for $j = 0 to ubound($a20) - 1
  if $a20[$j][0] = $a30[$i] then
   $a20[$j][0] = ''
   $a20[$j][1] = ''
  endif
  if $a20[$j][1] = $a30[$i] then
   $a20[$j][0] = ''
   $a20[$j][1] = ''
  endif
next
next
_arraydisplay($a20)

I keep thinking that some clever SQL might do this nicely, however, I don't have the skills for this.

Good Luck!!!

kylomas

Edited December 17, 2011 by kylomas

kylomas · December 17, 2011

J1,

One last thing (I just washed my keyboard with a large coffee).

Does this make sense: if each node has dups than the entry that contains either node, where one of the nodes is unique, is the good file?

Gotta go some of my keys are on coffee auto-repeat...

Kylomas

JohnOne · December 17, 2011

I'm not sure I follow, but I think you might be overcomplicating it (if that's possible).

Given that this is a good file...

123_456

Then if another file begins with 123, then its opposite end will be totally unique.

If a file ends in 456 then its opposite end will be totally unique.

The only exception to this rule, is the good file, it is the only file where

both ends have dupes.

Edited December 17, 2011 by JohnOne

jaberwacky · December 17, 2011

Why don't you ask Marilyn Vos Savant?

So if a good file has 123_456 then any file that exists which has either the 123 or the 456 will considered a bad file?

pseudo logic suggestions (Solved)

Recommended Posts

Link to comment

Share on other sites

Top Posters In This Topic

Top Posters In This Topic

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in

Recently Browsing 0 members