#REGEX0 discussing

Am I bad... the formula allowed only for lower case.

This one works (provided you have a data input flow that cleans the data and puts it into UPPERCASE before validation):

NOT(REGEX(BIC__c, "^[A-Z]{6}[0-9A-Z]{2}([0-9A-Z]{3})?\\z"))

I've managed to do it ! :)   

String recordDescription = formRecord.SVMX_PS_Description__c
String wordToFind = 'WOT Order :';
            Pattern word = Pattern.compile(wordToFind);
            Matcher match = word.matcher(formDescription);
            while (match.find()) {
                  substring = formDescription.mid(match.start(), 50);
                  System.debug('substring '+ substring );
                  if(substring.contains('If the user is')){
                      String newString = formDescription.mid(match.start()+12, 7);
                      listOfWOTwithFirstValidation.add(newString);
                      formRecord.WOT_Order_Future_date_not_allowed__c = String.Join(new List<String>(listOfWOTwithFirstValidation), '');
                  }
                  if(substring.contains('The Task has be')){
                      String newString = formDescription.mid(match.start()+12, 7);
                      listOfWOTwithSecondValidation.add(newString);
                      formRecord.WOT_Order_Task_Already_Completed__c = String.Join(new List<String>(listOfWOTwithSecondValidation), '');
                  }
            }

Hi Laura,

I guess I'll join the free-for-all to get best answer on this thread :)

First of all, here is a regex that might be what you need:

NOT(REGEX(Phone, "[+]{0,1}[-\\s\\.\\/0-9()]+"))

Now I'll explain what the various parts of it do:

[+]{0,1} means the character + can appear between 0 and 1 times. Because it's in the beginning of this expression, this character is only accepted at the beginning of the text we're matching.

[-\\s\\.\\/0-9()]+ means that any of the characters inside the square brackets (will detail in a bit) are allowed any number of times (but must appear at least once). The + just after the closing bracket means "at least once".

Now as for the specific characters here:

- means the literal hyphen character

\s means a whitespace character

\. means a literal dot/period

\/ means the literal character slash

0-9 means any digit between 0 and 9

() means closing or opening brackets

Oh, and to cover the elements in your original regex, I can explain those as well:

\D means any digit (same as 0-9)

{3} and {4} mean that the preceding block has to appear exactly 3 or 4 times, respectively.

You will notice that in this regex we actually use a double backslash \\ instead of a single one. This is actually not part of the Regex - it's for Salesforce to understand we mean a literal backslash and not a special character (like a tab \t or a line break \n). So we're actually double-encoding the expression, once for Salesforce, and a second time for the Regex engine.

One of my favourite tools for writing and examining regular expressions is Regex101. You can try this one here. You can hover over the various elements of the expression to see what they mean, and see the results in real time.

Hi  Brenden ,

 Try this

AND( 

NOT(MobilePhone ="") ,

LEN(MobilePhone) > 0 , 

NOT(REGEX( MobilePhone ,"(^ [+] 614[0-9]{2}(| )[0-9]{3}(| )[0-9]{3}$)|(^04[0-9]{2}(| )[0-9]{3}(| )[0-9]{3}$)"))

) 

Hi @Charles Porch III   

REGEX are sometimes very clunky but it does work very well if it comes to validation. For example, in your case, every character has its own functionality but if we want to pass through that special character we need to pass with \ backslash.  

Try to use this:

NOT(REGEX(Field Name ,"^[a-zA-Z&'*@|^:,;~_\\-=!\\/`>{}()\\[\\]<#%.+?\"]$"))

Note:  You may use character class “[^\w]” but again in your case you want to consider only specific characters to pass through.

Thanks.