Advertisement
Advertisement


How to use Regular Expressions (Regex) in Microsoft Excel both in-cell and loops


Question

How can I use regular expressions in Excel and take advantage of Excel's powerful grid-like setup for data manipulation?

  • In-cell function to return a matched pattern or replaced value in a string.
  • Sub to loop through a column of data and extract matches to adjacent cells.
  • What setup is necessary?
  • What are Excel's special characters for Regular expressions?

I understand Regex is not ideal for many situations (To use or not to use regular expressions?) since excel can use Left, Mid, Right, Instr type commands for similar manipulations.

2019/05/24
1
605
5/24/2019 3:01:53 PM

Accepted Answer

Regular expressions are used for Pattern Matching.

To use in Excel follow these steps:

Step 1: Add VBA reference to "Microsoft VBScript Regular Expressions 5.5"

  • Select "Developer" tab (I don't have this tab what do I do?)
  • Select "Visual Basic" icon from 'Code' ribbon section
  • In "Microsoft Visual Basic for Applications" window select "Tools" from the top menu.
  • Select "References"
  • Check the box next to "Microsoft VBScript Regular Expressions 5.5" to include in your workbook.
  • Click "OK"

Step 2: Define your pattern

Basic definitions:

- Range.

  • E.g. a-z matches an lower case letters from a to z
  • E.g. 0-5 matches any number from 0 to 5

[] Match exactly one of the objects inside these brackets.

  • E.g. [a] matches the letter a
  • E.g. [abc] matches a single letter which can be a, b or c
  • E.g. [a-z] matches any single lower case letter of the alphabet.

() Groups different matches for return purposes. See examples below.

{} Multiplier for repeated copies of pattern defined before it.

  • E.g. [a]{2} matches two consecutive lower case letter a: aa
  • E.g. [a]{1,3} matches at least one and up to three lower case letter a, aa, aaa

+ Match at least one, or more, of the pattern defined before it.

  • E.g. a+ will match consecutive a's a, aa, aaa, and so on

? Match zero or one of the pattern defined before it.

  • E.g. Pattern may or may not be present but can only be matched one time.
  • E.g. [a-z]? matches empty string or any single lower case letter.

* Match zero or more of the pattern defined before it. - E.g. Wildcard for pattern that may or may not be present. - E.g. [a-z]* matches empty string or string of lower case letters.

. Matches any character except newline \n

  • E.g. a. Matches a two character string starting with a and ending with anything except \n

| OR operator

  • E.g. a|b means either a or b can be matched.
  • E.g. red|white|orange matches exactly one of the colors.

^ NOT operator

  • E.g. [^0-9] character can not contain a number
  • E.g. [^aA] character can not be lower case a or upper case A

\ Escapes special character that follows (overrides above behavior)

  • E.g. \., \\, \(, \?, \$, \^

Anchoring Patterns:

^ Match must occur at start of string

  • E.g. ^a First character must be lower case letter a
  • E.g. ^[0-9] First character must be a number.

$ Match must occur at end of string

  • E.g. a$ Last character must be lower case letter a

Precedence table:

Order  Name                Representation
1      Parentheses         ( )
2      Multipliers         ? + * {m,n} {m, n}?
3      Sequence & Anchors  abc ^ $
4      Alternation         |

Predefined Character Abbreviations:

abr    same as       meaning
\d     [0-9]         Any single digit
\D     [^0-9]        Any single character that's not a digit
\w     [a-zA-Z0-9_]  Any word character
\W     [^a-zA-Z0-9_] Any non-word character
\s     [ \r\t\n\f]   Any space character
\S     [^ \r\t\n\f]  Any non-space character
\n     [\n]          New line

Example 1: Run as macro

The following example macro looks at the value in cell A1 to see if the first 1 or 2 characters are digits. If so, they are removed and the rest of the string is displayed. If not, then a box appears telling you that no match is found. Cell A1 values of 12abc will return abc, value of 1abc will return abc, value of abc123 will return "Not Matched" because the digits were not at the start of the string.

Private Sub simpleRegex()
    Dim strPattern As String: strPattern = "^[0-9]{1,2}"
    Dim strReplace As String: strReplace = ""
    Dim regEx As New RegExp
    Dim strInput As String
    Dim Myrange As Range

    Set Myrange = ActiveSheet.Range("A1")

    If strPattern <> "" Then
        strInput = Myrange.Value

        With regEx
            .Global = True
            .MultiLine = True
            .IgnoreCase = False
            .Pattern = strPattern
        End With

        If regEx.Test(strInput) Then
            MsgBox (regEx.Replace(strInput, strReplace))
        Else
            MsgBox ("Not matched")
        End If
    End If
End Sub

Example 2: Run as an in-cell function

This example is the same as example 1 but is setup to run as an in-cell function. To use, change the code to this:

Function simpleCellRegex(Myrange As Range) As String
    Dim regEx As New RegExp
    Dim strPattern As String
    Dim strInput As String
    Dim strReplace As String
    Dim strOutput As String


    strPattern = "^[0-9]{1,3}"

    If strPattern <> "" Then
        strInput = Myrange.Value
        strReplace = ""

        With regEx
            .Global = True
            .MultiLine = True
            .IgnoreCase = False
            .Pattern = strPattern
        End With

        If regEx.test(strInput) Then
            simpleCellRegex = regEx.Replace(strInput, strReplace)
        Else
            simpleCellRegex = "Not matched"
        End If
    End If
End Function

Place your strings ("12abc") in cell A1. Enter this formula =simpleCellRegex(A1) in cell B1 and the result will be "abc".

results image


Example 3: Loop Through Range

This example is the same as example 1 but loops through a range of cells.

Private Sub simpleRegex()
    Dim strPattern As String: strPattern = "^[0-9]{1,2}"
    Dim strReplace As String: strReplace = ""
    Dim regEx As New RegExp
    Dim strInput As String
    Dim Myrange As Range

    Set Myrange = ActiveSheet.Range("A1:A5")

    For Each cell In Myrange
        If strPattern <> "" Then
            strInput = cell.Value

            With regEx
                .Global = True
                .MultiLine = True
                .IgnoreCase = False
                .Pattern = strPattern
            End With

            If regEx.Test(strInput) Then
                MsgBox (regEx.Replace(strInput, strReplace))
            Else
                MsgBox ("Not matched")
            End If
        End If
    Next
End Sub

Example 4: Splitting apart different patterns

This example loops through a range (A1, A2 & A3) and looks for a string starting with three digits followed by a single alpha character and then 4 numeric digits. The output splits apart the pattern matches into adjacent cells by using the (). $1 represents the first pattern matched within the first set of ().

Private Sub splitUpRegexPattern()
    Dim regEx As New RegExp
    Dim strPattern As String
    Dim strInput As String
    Dim Myrange As Range

    Set Myrange = ActiveSheet.Range("A1:A3")

    For Each C In Myrange
        strPattern = "(^[0-9]{3})([a-zA-Z])([0-9]{4})"

        If strPattern <> "" Then
            strInput = C.Value

            With regEx
                .Global = True
                .MultiLine = True
                .IgnoreCase = False
                .Pattern = strPattern
            End With

            If regEx.test(strInput) Then
                C.Offset(0, 1) = regEx.Replace(strInput, "$1")
                C.Offset(0, 2) = regEx.Replace(strInput, "$2")
                C.Offset(0, 3) = regEx.Replace(strInput, "$3")
            Else
                C.Offset(0, 1) = "(Not matched)"
            End If
        End If
    Next
End Sub

Results:

results image


Additional Pattern Examples

String   Regex Pattern                  Explanation
a1aaa    [a-zA-Z][0-9][a-zA-Z]{3}       Single alpha, single digit, three alpha characters
a1aaa    [a-zA-Z]?[0-9][a-zA-Z]{3}      May or may not have preceding alpha character
a1aaa    [a-zA-Z][0-9][a-zA-Z]{0,3}     Single alpha, single digit, 0 to 3 alpha characters
a1aaa    [a-zA-Z][0-9][a-zA-Z]*         Single alpha, single digit, followed by any number of alpha characters

</i8>    \<\/[a-zA-Z][0-9]\>            Exact non-word character except any single alpha followed by any single digit
2020/02/13
971
2/13/2020 2:10:41 PM

To make use of regular expressions directly in Excel formulas the following UDF (user defined function) can be of help. It more or less directly exposes regular expression functionality as an excel function.

How it works

It takes 2-3 parameters.

  1. A text to use the regular expression on.
  2. A regular expression.
  3. A format string specifying how the result should look. It can contain $0, $1, $2, and so on. $0 is the entire match, $1 and up correspond to the respective match groups in the regular expression. Defaults to $0.

Some examples

Extracting an email address:

=regex("Peter Gordon: [email protected], 47", "\[email protected]\w+\.\w+")
=regex("Peter Gordon: [email protected], 47", "\[email protected]\w+\.\w+", "$0")

Results in: [email protected]

Extracting several substrings:

=regex("Peter Gordon: [email protected], 47", "^(.+): (.+), (\d+)$", "E-Mail: $2, Name: $1")

Results in: E-Mail: [email protected], Name: Peter Gordon

To take apart a combined string in a single cell into its components in multiple cells:

=regex("Peter Gordon: [email protected], 47", "^(.+): (.+), (\d+)$", "$" & 1)
=regex("Peter Gordon: [email protected], 47", "^(.+): (.+), (\d+)$", "$" & 2)

Results in: Peter Gordon [email protected] ...

How to use

To use this UDF do the following (roughly based on this Microsoft page. They have some good additional info there!):

  1. In Excel in a Macro enabled file ('.xlsm') push ALT+F11 to open the Microsoft Visual Basic for Applications Editor.
  2. Add VBA reference to the Regular Expressions library (shamelessly copied from Portland Runners++ answer):
    1. Click on Tools -> References (please excuse the german screenshot) Tools -> References
    2. Find Microsoft VBScript Regular Expressions 5.5 in the list and tick the checkbox next to it.
    3. Click OK.
  3. Click on Insert Module. If you give your module a different name make sure the Module does not have the same name as the UDF below (e.g. naming the Module Regex and the function regex causes #NAME! errors).

    Second icon in the icon row -> Module

  4. In the big text window in the middle insert the following:

    Function regex(strInput As String, matchPattern As String, Optional ByVal outputPattern As String = "$0") As Variant
        Dim inputRegexObj As New VBScript_RegExp_55.RegExp, outputRegexObj As New VBScript_RegExp_55.RegExp, outReplaceRegexObj As New VBScript_RegExp_55.RegExp
        Dim inputMatches As Object, replaceMatches As Object, replaceMatch As Object
        Dim replaceNumber As Integer
    
        With inputRegexObj
            .Global = True
            .MultiLine = True
            .IgnoreCase = False
            .Pattern = matchPattern
        End With
        With outputRegexObj
            .Global = True
            .MultiLine = True
            .IgnoreCase = False
            .Pattern = "\$(\d+)"
        End With
        With outReplaceRegexObj
            .Global = True
            .MultiLine = True
            .IgnoreCase = False
        End With
    
        Set inputMatches = inputRegexObj.Execute(strInput)
        If inputMatches.Count = 0 Then
            regex = False
        Else
            Set replaceMatches = outputRegexObj.Execute(outputPattern)
            For Each replaceMatch In replaceMatches
                replaceNumber = replaceMatch.SubMatches(0)
                outReplaceRegexObj.Pattern = "\$" & replaceNumber
    
                If replaceNumber = 0 Then
                    outputPattern = outReplaceRegexObj.Replace(outputPattern, inputMatches(0).Value)
                Else
                    If replaceNumber > inputMatches(0).SubMatches.Count Then
                        'regex = "A to high $ tag found. Largest allowed is $" & inputMatches(0).SubMatches.Count & "."
                        regex = CVErr(xlErrValue)
                        Exit Function
                    Else
                        outputPattern = outReplaceRegexObj.Replace(outputPattern, inputMatches(0).SubMatches(replaceNumber - 1))
                    End If
                End If
            Next
            regex = outputPattern
        End If
    End Function
    
  5. Save and close the Microsoft Visual Basic for Applications Editor window.

2015/09/28

Expanding on patszim's answer for those in a rush.

  1. Open Excel workbook.
  2. Alt+F11 to open VBA/Macros window.
  3. Add reference to regex under Tools then References
    ![Excel VBA Form add references
  4. and selecting Microsoft VBScript Regular Expression 5.5
    ![Excel VBA add regex reference
  5. Insert a new module (code needs to reside in the module otherwise it doesn't work).
    ![Excel VBA insert code module
  6. In the newly inserted module,
    ![Excel VBA insert code into module
  7. add the following code:

    Function RegxFunc(strInput As String, regexPattern As String) As String
        Dim regEx As New RegExp
        With regEx
            .Global = True
            .MultiLine = True
            .IgnoreCase = False
            .pattern = regexPattern
        End With
    
        If regEx.Test(strInput) Then
            Set matches = regEx.Execute(strInput)
            RegxFunc = matches(0).Value
        Else
            RegxFunc = "not matched"
        End If
    End Function
    
  8. The regex pattern is placed in one of the cells and absolute referencing is used on it. ![Excel regex function in-cell usage Function will be tied to workbook that its created in.
    If there's a need for it to be used in different workbooks, store the function in Personal.XLSB

2017/05/23

Here is my attempt:

Function RegParse(ByVal pattern As String, ByVal html As String)
    Dim regex   As RegExp
    Set regex = New RegExp

    With regex
        .IgnoreCase = True  'ignoring cases while regex engine performs the search.
        .pattern = pattern  'declaring regex pattern.
        .Global = False     'restricting regex to find only first match.

        If .Test(html) Then         'Testing if the pattern matches or not
            mStr = .Execute(html)(0)        '.Execute(html)(0) will provide the String which matches with Regex
            RegParse = .Replace(mStr, "$1") '.Replace function will replace the String with whatever is in the first set of braces - $1.
        Else
            RegParse = "#N/A"
        End If

    End With
End Function
2018/06/09

I needed to use this as a cell function (like SUM or VLOOKUP) and found that it was easy to:

  1. Make sure you are in a Macro Enabled Excel File (save as xlsm).
  2. Open developer tools Alt + F11
  3. Add Microsoft VBScript Regular Expressions 5.5 as in other answers
  4. Create the following function either in workbook or in its own module:

    Function REGPLACE(myRange As Range, matchPattern As String, outputPattern As String) As Variant
        Dim regex As New VBScript_RegExp_55.RegExp
        Dim strInput As String
    
        strInput = myRange.Value
    
        With regex
            .Global = True
            .MultiLine = True
            .IgnoreCase = False
            .Pattern = matchPattern
        End With
    
        REGPLACE = regex.Replace(strInput, outputPattern)
    
    End Function
    
  5. Then you can use in cell with =REGPLACE(B1, "(\w) (\d+)", "$1$2") (ex: "A 243" to "A243")

2018/06/09

This isn't a direct answer but may provide a more efficient alternative for your consideration. Which is that Google Sheets has several built in Regex Functions these can be very convenient and help circumvent some of the technical procedures in Excel. Obviously there are some advantages to using Excel on your PC but for the large majority of users Google Sheets will offer an identical experience and may offer some benefits in portability and sharing of documents.

They offer

REGEXEXTRACT: Extracts matching substrings according to a regular expression.

REGEXREPLACE: Replaces part of a text string with a different text string using regular expressions.

SUBSTITUTE: Replaces existing text with new text in a string.

REPLACE: Replaces part of a text string with a different text string.

You can type these directly into a cell like so and will produce whatever you'd like

=REGEXMATCH(A2, "[0-9]+")

They also work quite well in combinations with other functions such as IF statements like so:

=IF(REGEXMATCH(E8,"MiB"),REGEXEXTRACT(E8,"\d*\.\d*|\d*")/1000,IF(REGEXMATCH(E8,"GiB"),REGEXEXTRACT(E8,"\d*\.\d*|\d*"),"")

enter image description here

Hopefully this provides a simple workaround for users who feel taunted by the VBS component of Excel.

2019/11/26