How to extract just the letters from a string

Hi im replying to this because im trying to take only alphabetical characters out of what the users input to the text field (txtinput.text).. So if i put in "The yellow man had 32 teeth!" then im trying to get results like" Theyellowmanhadteeth" anything besides characters should be stripped out of the inputed field.

Hi im replying to this because im trying to take only alphabetical characters out of what the users input to the text field (txtinput.text).. So if i put in "The yellow man had 32 teeth!" then im trying to get results like" Theyellowmanhadteeth" anything besides characters should be stripped out of the inputed field.

The simple way would be to loop (probably using a For loop) through each character in the string (most likely using Mid() function) and if it is a letter, append it to another, output string.

Psedudo-code... Oh, and by the way - you can't really copy "anything besides characters" because they're all characters.

Hi im replying to this because im trying to take only alphabetical characters out of what the users input to the text field (txtinput.text).. So if i put in "The yellow man had 32 teeth!" then im trying to get results like" Theyellowmanhadteeth" anything besides characters should be stripped out of the inputed field.

I'm speechless: Killer didn't use Replace! Just brute-force it:
and, you can put it in a loop so that it is only three lines.

But, also do it Killer's way so that you can learn how to use the Mid function.

BTW, if you're using VB.NET, these functions are part of the Strings module, so you say Strings.Mid(...) and Strings.Replace(...)

I think it would be a bit more than three lines, Sammy. Or are you assuming that letters and numbers are the only possible characters?

Im trying to take only alphabetical characters out of what the users input to the text field (txtinput.text)..
So if i put in "The yellow man had 32 teeth!"
then im trying to get results like" Theyellowmanhadteeth"
anything besides characters should be stripped out of the inputed field.

this is in visual basic

Take two strings say strChar and strNumber

Read the length of your string....Loop through your string.....With the Substring() function Read characters one by one........Check characters with IsNumeric() Function whether its a number or not.....if number then strip off the character and store it to strNumber Otherwise store it to strChar.......

after the loop completes strChar will be your string without number.....
Implement it in code.......

it will be Something like....

...Check characters with IsNumeric() Function whether its a number or not.....if number then strip off the character and store it to strNumber Otherwise store it to strChar.......

What happens to spaces, and punctuation?

I'd do this with ascii characters for each letter, similar to what was said earlier with between "a" and "z".

This is proabably over complicating the problem quite a bit, but it works, it keeps the letters in the case they were found in as well ^^

Hope this helps.

I have a similar problem. can you guys help?
...

If I'm not mistaken, this sounds like a question for the Access forum. Would you like me to move it over there? You're likely to get better information there, if this is SQL-related.

I'd do this with ascii characters for each letter, similar to what was said earlier with between "a" and "z".
...

Looks good.

I do have a couple of issues with it, just on performance grounds. There seems to be a lot of duplicated effort.

For example, why use Left() then Right() function? Is this faster than the Mid() function? (this is a serious question - sometimes the longer coding can be more efficient). Or perhaps it's just a personal preference?

And your second If test doesn't need to be executed if the first was satisfied. Personally, I'd short-cut the process by using ElseIf or similar. Plus I would place the Asc() value into a variable so the function is only called once.

In fact, one of the main reasons I like to use Select Case for these kind of situations is that it allows you much more flexibility in your conditions (if checking one value). For example, where you are doing two IF tests, I would have used something like... However, while this seems (to me) "nicer" to code, I don't know how it compares on performance. Must check it out one of these days...

Oh, and one more point - I would always use a Long in favour of an Integer, unless space was very tight. Believe it or not, Long data type is (very slightly) faster to use on a 32 bit processor.

If I'm not mistaken, this sounds like a question for the Access forum.

right. I got carried away.

right. I got carried away.

Hang on, I'll do a bit of Moderator Magic...

Ok, here is the new thread.

I'd do this with ascii characters for each letter, similar to what was said earlier with between "a" and "z".

This is proabably over complicating the problem quite a bit, but it works, it keeps the letters in the case they were found in as well ^^

Expand|Select|Wrap|Line Numbers

1.  
2.     Dim strText As String, strLeft As String, strRight As String
3.     Dim intA As Integer
4.     strText = Text1.Text
5.     For intA = 1 To Len(strText)
6.         strLeft = Left(strText, intA)
7.         strRight = Right(strLeft, 1)
8.  
9.         If Asc(strRight) > 97 And Asc(strRight) < 122 Then
10.             Label1.Caption = Label1.Caption & strRight
11.         End If
12.         If Asc(strRight) > 65 And Asc(strRight) < 90 Then
13.             Label1.Caption = Label1.Caption & strRight
14.         End If
15.     Next intA
16.  
17.  

Hope this helps.

Izzit this code is for vb6?in vb 2005 is there also can use the Left method?Thanks

Izzit this code is for vb6?in vb 2005 is there also can use the Left method?Thanks

Yes, I write in VB6. VB 2005 certainly does have comparable functions, but I don't know the details. I think if you look in the doco, either on your PC or at MS' website, you should find it easily enough.

I always prefer to keep them from putting the characters I don't like into the field in the first place by intercepting the keypress event:

You can then allow back space, delete etc by adding the appropriate character codes. I use this a lot when I just want numbers in auto-calculating forms which I write a lot of.