Remove Array() by index

[CrAz3D]

I am aware that this has recently been posted, yet does not function to my needs.


I have my Array

http://
feany
gay
bob

I can remove the first Array index(http://) correctly.  However, if I try to remove the second index(feany) it removes the http:// aswell.  But wait, there's more.  When I remove the last index(bob) it just duplicates the first index(http://) aswell.

Remove Code

Code Select Expand

For i = 0 To UBound(PhraseList) - 1
               If LCase(pA(1)) = LCase(PhraseList(i)) Then
                   SendW "Removed phrase: " & PhraseList(i), strAccount
                   PhraseList(i) = vbNullString
                   ReDim Preserve PhraseList(UBound(PhraseList) - 1)
               End If
           Next i

Help is appreciatd!  

[Stealth]

Here's the most basic way to remove an item from an ARRAY. If you'll be doing it often you may want to look into a collection.

1. Create a new temporary array, dimensioned to (size - 1) of your start array.
2. Loop throguh your start array, copying everything but the index you want to remove. The temp array now contains your entire start array except the item you want to remove.
3. Redimension your start array to the size of the temp array.
4. Copy the temp array into the start array.

[CrAz3D]

Here's the most basic way to remove an item from an ARRAY. If you'll be doing it often you may want to look into a collection.

1. Create a new temporary array, dimensioned to (size - 1) of your start array.
2. Loop throguh your start array, copying everything but the index you want to remove. The temp array now contains your entire start array except the item you want to remove.
3. Redimension your start array to the size of the temp array.
4. Copy the temp array into the start array.

I'll try a collection...thnx

[Grok]

I'll try a collection...thnx

Bah.  You learn not!

[Lenny]

Just a note, collections are much slower and less efficient than arrays, they should be avoided in general...

In addition, you could merely overwrite the array(index) you want to erase with array(index+1) and loop this, after this is finished you resize the array to (Ubound - 1) because the last array value will be a copy of the second to last....

[Stealth]

Just a note, collections are much slower and less efficient than arrays, they should be avoided in general...

In addition, you could merely overwrite the array(index) you want to erase with array(index+1) and loop this, after this is finished you resize the array to (Ubound - 1) because the last array value will be a copy of the second to last....

That is indeed more efficient than I described. Does it work in all situations?

[Flame]

I did something similar to what Lenny said, but only for one purpose (coincidence, Phraseban!) and it seems to work fine.  I can't explain it too well... but I'll just paste my code up here to see if you can get the idea.  I haven't used it a lot, but with the little I have used it, it functions well.

Code Select Expand


           Dim i As Integer
           Dim RFP As Integer
           If Phrasebans(0) = "" Then Exit Sub
           For i = 0 To UBound(Phrasebans)
               If LCase(Mid$(Message, 11)) = LCase(Phrasebans(i)) Then
                   Phrasebans(i) = ""
                   For RFP = i To UBound(Phrasebans)
                       If Phrasebans(RFP) = "" And RFP < UBound(Phrasebans) Then
                           Phrasebans(RFP) = Phrasebans(RFP + 1)
                           Phrasebans(RFP + 1) = ""
                       End If
                   Next RFP
                   If UBound(Phrasebans) = 0 Then
                       Phrasebans(0) = ""
                   Else
                       ReDim Preserve Phrasebans(0 To UBound(Phrasebans) - 1)
                   End If
                   Exit For
               End If
           Next i


[Stealth]

That appears to be exactly what Lenny described.

[Fr0z3N]

uhh, I think Stealth has a good way, thats what I use.

[o.OV]

I read collections are slower if working
with about 100 items or less.
I don't use collections
so I can't give an opinion on that.

If the array does not need to be
in an order of any kind.
If a string.. switch the descriptor with last item
If a value.. item to remove equal to last item
last item not neccessarily the ubound,
and then zero out the last item
to kill off the extra item,
depending on how you want to handle it.

[Adron]

Just a note, collections are much slower and less efficient than arrays, they should be avoided in general...

Where did you find this information? I was assuming that collections were some kind of binary tree which would make them very efficient for things like removing an item from the middle, looking up an item from a key, growing and shrinking.

[Grok]

http://www.mvps.org/vb/hardcore/html/collections201indexing.htm

and

http://www.mvps.org/vb/hardcore/html/walkingclistcollection.htm

[CrAz3D]

From what I see an Array is better in most opinions & according to facts.  I will redo it to use an Array instead, thank you all.


There still seems to be a few problems that I cannot fiugre out.

[8:05:42 AM] «CrAz3D[xL]» \flist
[8:05:42 AM] «0x69» Current Filters are: http, stealth, bob, gay, feany, telrin
[8:05:46 AM] «CrAz3D[xL]» \frem http
[8:05:46 AM] «0x69» Removed filter: http
[8:05:48 AM] «CrAz3D[xL]» \flist
[8:05:48 AM] «0x69» Current Filters are: stealth, bob, gay, feany, telrin
[8:05:53 AM] «CrAz3D[xL]» \frem telrin
[8:05:53 AM] «0x69» Removed filter: telrin
[8:05:55 AM] «CrAz3D[xL]» \flist
[8:05:55 AM] «0x69» Current Filters are: bob, gay, feany, feany

Code Select Expand

For i = 0 To UBound(FilterList) - 1
               If LCase(pA(1)) = LCase(FilterList(i)) Then
                   SendW "Removed filter: " & FilterList(i), strAccount
                   FilterList(i) = FilterList(i + 1)
                   ReDim Preserve FilterList(UBound(FilterList) - 1)
               End If
           Next i
I have a feeling the it is the "ReDim Preserve" portion that I am screwing up.

[drivehappy]

Your ReDim statement appears correct.

Code Select Expand


FilterList(i) = FilterList(i + 1)

This however only moves the next element into the current's position - not all subsequent elements in the array. If 'i' is at the end of the array an out of bounds error may occur if you try accessing the next element that is not there. I cannot see how your first element is removed though.

[CrAz3D]

hmm, ok...I'll mess w/the Filterlist - w/e