Text manipulation

"Is there some VBA code which could delete all first, second or third
characters of a text? Could it be done to the three last characters
from this same text and these be displayed on reverse order?"

Example:
AAAASAHDASK
AAASAHDASK
AASAHDASK
ASAHDASK
SADHASAAAA
ADHASAAAA
DHASAAAA

I'm guessing you could use the LEFT, MID and RIGHT functions, but I've not
used them in VBA.

"paulinoluciano" <paulinoluciano@zipmail.com.br> wrote in message
news:1135761865.111822.80740@g47g2000cwa.googlegroups.com...
> "Is there some VBA code which could delete all first, second or third
> characters of a text? Could it be done to the three last characters
> from this same text and these be displayed on reverse order?"
>
> Example:
> AAAASAHDASK
> AAASAHDASK
> AASAHDASK
> ASAHDASK
> SADHASAAAA
> ADHASAAAA
> DHASAAAA
>

You are right Barb Reinhardt... But in this case I would need to use
some kind of VBA code because I would not like to have to put each
sequence o characters and functions for each cell independently. I
would like to perform an automatic approach because I have a lot of
files to do that.
Thank you anyway!
Luciano

Sub Test()
Dim iLastRow As Long
Dim i As Long, j As Long
Dim temp
Const nIndex As Long = 3 'every third, change to suit

iLastRow = Cells(Rows.Count, "A").End(xlUp).Row
For i = 1 To iLastRow
If nIndex = 1 Then
Cells(i, "A").Value = Right(Cells(i, "A").Value, _
Len(Cells(i, "A").Value) - 1)
Else
Cells(i, "A").Value = Left(Cells(i, "A").Value, nIndex - 1) & _
Right(Cells(i, "A").Value, Len(Cells(i,
"A").Value) - nIndex)
End If
Next i

End Sub


--
HTH

Bob Phillips

(remove nothere from email address if mailing direct)

"paulinoluciano" <paulinoluciano@zipmail.com.br> wrote in message
news:1135761865.111822.80740@g47g2000cwa.googlegroups.com...
> "Is there some VBA code which could delete all first, second or third
> characters of a text? Could it be done to the three last characters
> from this same text and these be displayed on reverse order?"
>
> Example:
> AAAASAHDASK
> AAASAHDASK
> AASAHDASK
> ASAHDASK
> SADHASAAAA
> ADHASAAAA
> DHASAAAA
>

Thank you Bob Phillips! It solve my problem in part. However the major
question is just a few more complex. I explained it better in topic
Text subsequences. In the present topic instead subtract one letter any
time I need a macro capable to let the first cell (e.g. A1) as it is
and remove the characters only in the next row (A2) that it would be
identical to the previous without the first letter.
Luciano

I didn't understand the other one!

--
HTH

Bob Phillips

(remove nothere from email address if mailing direct)

"paulinoluciano" <paulinoluciano@zipmail.com.br> wrote in message
news:1135780063.756751.135410@z14g2000cwz.googlegroups.com...
> Thank you Bob Phillips! It solve my problem in part. However the major
> question is just a few more complex. I explained it better in topic
> Text subsequences. In the present topic instead subtract one letter any
> time I need a macro capable to let the first cell (e.g. A1) as it is
> and remove the characters only in the next row (A2) that it would be
> identical to the previous without the first letter.
> Luciano
>

In fact, the other topic is just a few more complex until to explain.
Let me try explain better. I have a sequence of characters like:

AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ

This sequence must be put in cell A2.
Thus, I have to perform some specific operations in this text:

Example 1:
Rules:
a) Fragment the sequence before K but not always (you could have lost
cut).
b) Sequence is not cut if K is found before FP

Results:

AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ

0 lost cut = Cutting the sequence all the time in which K is present
(The subsequences of this process should be put in B column:
AASSASDK
ASASDASFAFSASASADK
ASASAFPKQREWEAQEOK
SPADAOEK
OQPPDAOPSK
AEPQ

1 lost cut = Cutting the sequence after the first K present in the
sequence (The subsequences of this process should be put in C column::
AASSASDKASASDASFAFSASASADK
ASASAFPKQREWEAQEOKSPADAOEK
OQPPDAOPSKAEPQ
AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
SPADAOEKOQPPDAOPSKAEPQ

2 lost cut = = Cutting the sequence after the second K (just for the
third and following) present in the sequence (The subsequences of this
process should be put in D column:
AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
SPADAOEKOQPPDAOPSKAEPQ

Repair that in some cases I need lost cuts in which you cut after 1, 2,
3, 4,... specific characters.
I have to specify such rules in some place of the sheet containing the
precursor text.
The rules are:

Cut after "XXX" (In this example I have put K but the some cell in the
sheet must contain what is the character in which the sequence will be
fragmented). In some cases it could be more than only one character
(e.g. K and R; nor necessarily together)
Cut before "XXX" (The cut may be after like previous example or before
the character)

Never before "XXX" (In some cases I have prohibitive situations; e.g.
It must not cut a sequence in K if K is preceeded by P or by RP)
Never after "XXX" (Same for after)

Number of times that the character could be missed prior cut "XXX" (In
some place of the sheet I must explicit how many characters could be
"lost" prior cut (see example).

Just out of curiosity, what not-Excel, real world problem is this?

--
Kind regards,

Niek Otten

"paulinoluciano" <paulinoluciano@zipmail.com.br> wrote in message
news:1135798777.456374.90110@f14g2000cwb.googlegroups.com...
> In fact, the other topic is just a few more complex until to explain.
> Let me try explain better. I have a sequence of characters like:
>
> AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ
>
> This sequence must be put in cell A2.
> Thus, I have to perform some specific operations in this text:
>
> Example 1:
> Rules:
> a) Fragment the sequence before K but not always (you could have lost
> cut).
> b) Sequence is not cut if K is found before FP
>
> Results:
>
> AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ
>
> 0 lost cut = Cutting the sequence all the time in which K is present
> (The subsequences of this process should be put in B column:
> AASSASDK
> ASASDASFAFSASASADK
> ASASAFPKQREWEAQEOK
> SPADAOEK
> OQPPDAOPSK
> AEPQ
>
> 1 lost cut = Cutting the sequence after the first K present in the
> sequence (The subsequences of this process should be put in C column::
> AASSASDKASASDASFAFSASASADK
> ASASAFPKQREWEAQEOKSPADAOEK
> OQPPDAOPSKAEPQ
> AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
> SPADAOEKOQPPDAOPSKAEPQ
>
> 2 lost cut = = Cutting the sequence after the second K (just for the
> third and following) present in the sequence (The subsequences of this
> process should be put in D column:
> AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
> SPADAOEKOQPPDAOPSKAEPQ
>
> Repair that in some cases I need lost cuts in which you cut after 1, 2,
> 3, 4,... specific characters.
> I have to specify such rules in some place of the sheet containing the
> precursor text.
> The rules are:
>
> Cut after "XXX" (In this example I have put K but the some cell in the
> sheet must contain what is the character in which the sequence will be
> fragmented). In some cases it could be more than only one character
> (e.g. K and R; nor necessarily together)
> Cut before "XXX" (The cut may be after like previous example or before
> the character)
>
> Never before "XXX" (In some cases I have prohibitive situations; e.g.
> It must not cut a sequence in K if K is preceeded by P or by RP)
> Never after "XXX" (Same for after)
>
> Number of times that the character could be missed prior cut "XXX" (In
> some place of the sheet I must explicit how many characters could be
> "lost" prior cut (see example).
>

On 28 Dec 2005 11:39:37 -0800, "paulinoluciano" <paulinoluciano@zipmail.com.br>
wrote:

>In fact, the other topic is just a few more complex until to explain.
>Let me try explain better. I have a sequence of characters like:
>
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ
>
>This sequence must be put in cell A2.
>Thus, I have to perform some specific operations in this text:
>
>Example 1:
>Rules:
>a) Fragment the sequence before K but not always (you could have lost
>cut).
>b) Sequence is not cut if K is found before FP
>
>Results:
>
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ
>
>0 lost cut = Cutting the sequence all the time in which K is present
>(The subsequences of this process should be put in B column:
>AASSASDK
>ASASDASFAFSASASADK
>ASASAFPKQREWEAQEOK
>SPADAOEK
>OQPPDAOPSK
>AEPQ
>
>1 lost cut = Cutting the sequence after the first K present in the
>sequence (The subsequences of this process should be put in C column::
>AASSASDKASASDASFAFSASASADK
>ASASAFPKQREWEAQEOKSPADAOEK
>OQPPDAOPSKAEPQ
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
>SPADAOEKOQPPDAOPSKAEPQ
>
>2 lost cut = = Cutting the sequence after the second K (just for the
>third and following) present in the sequence (The subsequences of this
>process should be put in D column:
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
>SPADAOEKOQPPDAOPSKAEPQ
>
>Repair that in some cases I need lost cuts in which you cut after 1, 2,
>3, 4,... specific characters.
>I have to specify such rules in some place of the sheet containing the
>precursor text.
>The rules are:
>
>Cut after "XXX" (In this example I have put K but the some cell in the
>sheet must contain what is the character in which the sequence will be
>fragmented). In some cases it could be more than only one character
>(e.g. K and R; nor necessarily together)
>Cut before "XXX" (The cut may be after like previous example or before
>the character)
>
>Never before "XXX" (In some cases I have prohibitive situations; e.g.
>It must not cut a sequence in K if K is preceeded by P or by RP)
>Never after "XXX" (Same for after)
>
>Number of times that the character could be missed prior cut "XXX" (In
>some place of the sheet I must explicit how many characters could be
>"lost" prior cut (see example).

You may want to look into "regular expressions" to do what you are trying to
describe. If you download and install Longre's free morefunc.xll add-in from
http://xcell05.free.fr/ you will see that you can use them as worksheet
functions and also call them from a VBA module.

What you write is a bit confusing. For example, one rule you give is:
"Sequence is not cut if K is found before FP" but in your example you seem to
be acting as if the rule applies if K is found AFTER FP.

I am assuming the output starts in B1; if it starts in a different row, then
adjust the ROW() function to result in a 1 as the output:

seq is the character sequence (Insert/Name/Define and set seq = "your string")

For the "0 lost cuts"

B1: =REGEX.MID(seq,"(\w+?([^FP]K|$)){"&COLUMN()-1&"}",ROW())

ROW() resolves to a '1' which means take the 'first' sequence that matches the
pattern. As you copy/drag the formula down, ROW() will resolve to '2', '3',
etc. which means match the 2nd, 3rd, etc sequence that matches the pattern.

The basic pattern is defined by "(\w+?([^FP]K|$)){" which means look for a
sequence of letters that ends with a K that is not preceded by an FP, or that
is at the end of the string.

The {"&COLUMN()-1&"}" resolves, in Column B, to {1} which means look for one
occurrence of the preceding pattern.

If you copy/drag the formula down until you get blanks for the results, you
will see what you posted in your previous message.

If you copy/drag across to column D, you will see the results of "1 lost cut"
or "2 lost cuts".

I think once you understand the formula construction and the regular
expressions, it will be simple to use this for your other rules.

Without the COLUMN and ROW functions, the formulas would look like:

B1: =REGEX.MID(seq,"(\w+?([^FP]K|$)){1}",1)
B2: =REGEX.MID(seq,"(\w+?([^FP]K|$)){1}",2)

C1: =REGEX.MID(seq,"(\w+?([^FP]K|$)){2}",1)
C2: =REGEX.MID(seq,"(\w+?([^FP]K|$)){2}",2)

To use this in VBA, you would use the RUN method which is outlined in HELP for
morefunc.xll
--ron

On 28 Dec 2005 11:39:37 -0800, "paulinoluciano" <paulinoluciano@zipmail.com.br>
wrote:

>1 lost cut = Cutting the sequence after the first K present in the
>sequence (The subsequences of this process should be put in C column::
>AASSASDKASASDASFAFSASASADK
>ASASAFPKQREWEAQEOKSPADAOEK
>OQPPDAOPSKAEPQ
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
>SPADAOEKOQPPDAOPSKAEPQ

See my other answer. But I did not understand how you obtained the last two
lines in the "1 lost cut" sequence.

They are identical to the two lines in the "2 lost cut" sequence, so I thought
this might be a typo. But perhaps I am missing something?


--ron

paulinoluciano wrote...
....
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ
>
>This sequence must be put in cell A2.
....
>Rules:
>a) Fragment the sequence before K but not always (you could have lost cut).
>b) Sequence is not cut if K is found before FP
>
>Results:
>
>ASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOKSPADAOEKOQPPDAOPSKAEPQ
>
>0 lost cut = Cutting the sequence all the time in which K is present
>(The subsequences of this process should be put in B column:
>AASSASDK
>ASASDASFAFSASASADK
>ASASAFPKQREWEAQEOK
>SPADAOEK
>OQPPDAOPSK
>AEPQ

You could use formulas.

B2:
=LEFT($A$2,FIND("K",SUBSTITUTE($A$2,"FPK","###")))

B3:
=REPLACE(LEFT($A$2,FIND("K",SUBSTITUTE($A$2,"FPK","###")&"K",
SUMPRODUCT(LEN(B$2:B2))+1)),1,SUMPRODUCT(LEN(B$2:B2)),"")

Fill B3 down as needed. Filling it into B4:B8 (one cell more than
needed) gives B2:B8

AASSASDK
ASASDASFAFSASASADK
ASASAFPKQREWEAQEOK
SPADAOEK
OQPPDAOPSK
AEPQ
<blank>

>1 lost cut = Cutting the sequence after the first K present in the
>sequence (The subsequences of this process should be put in C column::
>AASSASDKASASDASFAFSASASADK
>ASASAFPKQREWEAQEOKSPADAOEK
>OQPPDAOPSKAEPQ
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
>SPADAOEKOQPPDAOPSKAEPQ

You have all the information you need for this in column B.

C2:
=INDEX(B$2:B$99,2*ROWS(C$2:C2)-1)&INDEX(B$2:B$99,2*ROWS(C$2:C2))

Fill C2 down as needed. Filling it into C3:C5 (one more than needed)
gives C2:C5

AASSASDKASASDASFAFSASASADK
ASASAFPKQREWEAQEOKSPADAOEK
OQPPDAOPSKAEPQ
<blank>

>2 lost cut = = Cutting the sequence after the second K (just for the
>third and following) present in the sequence (The subsequences of this
>process should be put in D column:
>AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
>SPADAOEKOQPPDAOPSKAEPQ

D2:
=INDEX(B$2:B$99,3*ROWS(D$2:D2)-2)&INDEX(B$2:B$99,3*ROWS(D$2:D2)-1)
&INDEX(B$2:B$99,3*ROWS(D$2:D2))

Fill D2 down as needed. Filling it into D3:D4 (one more than needed)
gives D2:D4

AASSASDKASASDASFAFSASASADKASASAFPKQREWEAQEOK
SPADAOEKOQPPDAOPSKAEPQ
<blank>

>Repair that in some cases I need lost cuts in which you cut after 1, 2,
>3, 4,... specific characters.
>I have to specify such rules in some place of the sheet containing the
>precursor text.
>The rules are:
>
>Cut after "XXX" (In this example I have put K but the some cell in the
>sheet must contain what is the character in which the sequence will be
>fragmented). In some cases it could be more than only one character
>(e.g. K and R; nor necessarily together)
>Cut before "XXX" (The cut may be after like previous example or before
>the character)
>
>Never before "XXX" (In some cases I have prohibitive situations; e.g.
>It must not cut a sequence in K if K is preceeded by P or by RP)

The RP preceding K is redundant if P alone preceding K indicates a
prohibited situation. You'd only need to check for PK.

>Never after "XXX" (Same for after)
>
>Number of times that the character could be missed prior cut "XXX" (In
>some place of the sheet I must explicit how many characters could be
>"lost" prior cut (see example).

Generalizing the formulas above with the character to match in a cell
named CC and the prohibited sequence (in this case FPK) in a cell named
PS,

B2:
=LEFT($A$2,FIND(CC,SUBSTITUTE($A$2,PS,REPT("#",LEN(PS)))))

B3:
=REPLACE(LEFT($A$2,FIND(CC,SUBSTITUTE($A$2,PS,REPT("#",LEN(PS)))&CC,
SUMPRODUCT(LEN(B$2:B2))+1)),1,SUMPRODUCT(LEN(B$2:B2)),"")

The column C, D, etc. formulas wouldn't need to change.

If you have multiple prohibited sequences, then regular expressions
would be MUCH BETTER tools for doing this. Symbolic processing like
this is reducible to text processing, but Excel provides poor built-in
tools for text processing, but since it was meant to calculate numbers
this shouldn't be surprising.

Bob Phillips wrote...
....
> If nIndex = 1 Then
> Cells(i, "A").Value = Right(Cells(i, "A").Value, _
> Len(Cells(i, "A").Value) - 1)
> Else
> Cells(i, "A").Value = Left(Cells(i, "A").Value, nIndex - 1) & _
> Right(Cells(i, "A").Value, Len(Cells(i, "A").Value) - nIndex)
> End If
....

Could simplify: Right(x, Len(x) - y) == Mid(x, y + 1)

Then again, the whole If block could be replaced by

Cells(i, "A").Value = _
Application.WorksheetFunction.Replace(Cells(i, "A").Value, nIndex,
1, "")

Thank you Harlan Grove!
In fact, this solve my first "problem" related to this kind of text
manipulation. However, now I have to specify in such VBA code that the
sequence could be cut after or before two or more letters
simultaneously.
Example:

The sequence is

IADASFDTYEREPWQNMSDFGHKEASADSASSASADRASERAS

cut after K or R

0 lost:
IADASFDTYER
EPWQNMSDFGHK
EASADSASSASADR
SER
AS

1 lost cut:
IADASFDTYEREPWQNMSDFGHK
EPWQNMSDFGHKEASADSASSASADR
EASADSASSASADRSER
SERAS
......


Harlan Grove wrote:
> Bob Phillips wrote...
> ...
> > If nIndex = 1 Then
> > Cells(i, "A").Value = Right(Cells(i, "A").Value, _
> > Len(Cells(i, "A").Value) - 1)
> > Else
> > Cells(i, "A").Value = Left(Cells(i, "A").Value, nIndex - 1) & _
> > Right(Cells(i, "A").Value, Len(Cells(i, "A").Value) - nIndex)
> > End If
> ...
>
> Could simplify: Right(x, Len(x) - y) == Mid(x, y + 1)
>
> Then again, the whole If block could be replaced by
>
> Cells(i, "A").Value = _
> Application.WorksheetFunction.Replace(Cells(i, "A").Value, nIndex,
> 1, "")

paulinoluciano wrote:
> Thank you Harlan Grove!
> In fact, this solve my first "problem" related to this kind of text
> manipulation. However, now I have to specify in such VBA code that the
> sequence could be cut after or before two or more letters
> simultaneously.
> Example:
>
> The sequence is
>
> IADASFDTYEREPWQNMSDFGHKEASADSASSASADRASERAS
>
> cut after K or R
>
> 0 lost:
> IADASFDTYER
> EPWQNMSDFGHK
> EASADSASSASADR
> SER
> AS
>
> 1 lost cut:
> IADASFDTYEREPWQNMSDFGHK
> EPWQNMSDFGHKEASADSASSASADR
> EASADSASSASADRSER
> SERAS
> .....
>
>
> Harlan Grove wrote:
> > Bob Phillips wrote...
> > ...
> > > If nIndex = 1 Then
> > > Cells(i, "A").Value = Right(Cells(i, "A").Value, _
> > > Len(Cells(i, "A").Value) - 1)
> > > Else
> > > Cells(i, "A").Value = Left(Cells(i, "A").Value, nIndex - 1) & _
> > > Right(Cells(i, "A").Value, Len(Cells(i, "A").Value) - nIndex)
> > > End If
> > ...
> >
> > Could simplify: Right(x, Len(x) - y) == Mid(x, y + 1)
> >
> > Then again, the whole If block could be replaced by
> >
> > Cells(i, "A").Value = _
> > Application.WorksheetFunction.Replace(Cells(i, "A").Value, nIndex,
> > 1, "")