VBA: Help with Regular Expression

[Converemon]

I need help generating a regular expression that does not match a certain form of string. The string can be:

-K09+M72+L86+B12 . . .
Or even K09 + M72 +L86 + B12

But cannot be:
-K098+M72+L86+B12
-kshbd+M72+L86+B12
-K98+M72+Ldsjfjhb86+B12
-etc.

Right now I have ^(?!.*^\s*[A-Z][0-9][0-9]\s*(\+\s*[A-Z][0-9][0-9]\s*)*$ as my expression which does seem to work very well with what I'm trying to do. Again I am trying to match strings that do not comply with the standards (i.e. the second group). I hope a fresh pair of eyes can put out what I'm doing wrong.

</SPAN> <TBODY> </TBODY> </SPAN>

<TBODY>
</TBODY>

 

[Rick Rothstein]

I need help generating a regular expression that does not match a certain form of string. The string can be:

-K09+M72+L86+B12 . . .
Or even K09 + M72 +L86 + B12

But cannot be:
-K098+M72+L86+B12
-kshbd+M72+L86+B12
-K98+M72+Ldsjfjhb86+B12
-etc.

Right now I have ^(?!.*^\s*[A-Z][0-9][0-9]\s*(\+\s*[A-Z][0-9][0-9]\s*)*$ as my expression which does seem to work very well with what I'm trying to do. Again I am trying to match strings that do not comply with the standards (i.e. the second group). I hope a fresh pair of eyes can put out what I'm doing wrong.

<tbody> </tbody>

<tbody>
</tbody>

I do not do Regular Expressions, but there must by a "not" meta-characters (perhaps the bang character?) which could be used... maybe wrap the whole expression with parentheses and put the not meta-character in front of it? Anyway, that is not why I responded, rather, I wanted to mention the VB Like operator which maybe of interest to you. You would have to process your data one at a time in a loop (assuming you have more than one text string to process), but you can test for text that does not match your pattern using it. First though, I assumed those dashes in the front are not part of your text (because I do not see a dash in your pattern string), rather, I assumed you used them like "bullet points". Given that assumption, this test code will test a single text string (embed it in a loop if you have a list of text strings to process)...

If Not (Relace(TextString, " ", "") Like "[A-Z]##+"[A-Z]##+"[A-Z]##+"[A-Z]##") Then
    '
    '  TextString did not match the pattern, so do whatever to it
    '
End If

 

[Converemon]

I did in fact consider using the Like operator for this, but my main problem is I don't know how many "codes" will be in the string. For example, it could be M71+L09+J82, M71, or even M71+L09+J82+P55+Q21+T57+U82+M08+O99 . . . Apologies for the dashes, they were meant as bullet points.

 

[Rick Rothstein]

I did in fact consider using the Like operator for this, but my main problem is I don't know how many "codes" will be in the string. For example, it could be M71+L09+J82, M71, or even M71+L09+J82+P55+Q21+T57+U82+M08+O99 . . . Apologies for the dashes, they were meant as bullet points.

Oh, so that is what the series of dots at the end of your first example meant... all your examples had the same number of codes in them, so I figured that four codes per text string was a fixed requirement. Okay, you can still use the Like operator to test your text strings (rather than using Regular Expressions to do it), consider this...

Repeats = UBound(Split(TextString, "+")) + 1
If Not (Replace(TextString & "+", " ", "") Like Replace(Space(Repeats), " ", "[A-Z]##+")) Then
    '
    '  TextString did not match the pattern, so do whatever to it
    '
End If

Note that I introduced a new variable named Repeats which should be Dim'med as Long if you, in fact, Dim your variables.

 

[Converemon]

From all of the tests I ran, your method seems to work perfectly for my situation. I am going to implement it into my project right away and I will let you know how it goes. I really appreciate your time. Also, I had another question concerning the "Split" method you used. I am using it to parse the strings in my project ("M71+L22+K55") using "+" as the delimiter. However, when I try to do this, it always skips the very first "code" ("M71"). Do you have any idea what might be happening?

 

[Rick Rothstein]

From all of the tests I ran, your method seems to work perfectly for my situation. I am going to implement it into my project right away and I will let you know how it goes. I really appreciate your time. Also, I had another question concerning the "Split" method you used. I am using it to parse the strings in my project ("M71+L22+K55") using "+" as the delimiter. However, when I try to do this, it always skips the very first "code" ("M71"). Do you have any idea what might be happening?

The Split function always produces a zero-based array (even if the Option Base is set to 1), so you must start your loops at 0, not 1

 

[Converemon]

Silly me, haha. I definitely knew that I just don't know why I didn't consider it. Thank you very much for your help; I do really appreciate it.

Also, your method worked perfectly in my project.