Macro: find multiple terms and replace for a single one. Regex

[eu2020]

Hi,

I’m trying do create a macro to “find and replace”, where in the “find” side there are several regex expressions; and in the “replace” side there is only one word. The following macro works for a single regex, but I don’t know how to put multiple regex in the line “oReplace.SearchString =”

Sub ReplaceFormatting
  'original code : Alex Savitsky
  'modified by : Laurent Godard
  

  Dim oDoc As Object
  Dim oReplace As Object
  Dim SrchAttributes(0) As New com.sun.star.beans.PropertyValue
  Dim ReplAttributes(0) As New com.sun.star.beans.PropertyValue

  oDoc = ThisComponent
  oReplace = oDoc.createReplaceDescriptor

  'Regular expression. Match any text
  oReplace.SearchString = "^\s*A"
  'Note the & places the found text back
  oReplace.ReplaceString = "<&>"
  oReplace.SearchRegularExpression=True  'Use regular expressions
  oReplace.searchStyles=False            
  oReplace.searchAll=True                'Do the entire document

 
  REM Now do the work!
  oDoc.replaceAll(oReplace)
End Sub

I tried things like this, with no success (gives me a syntax error message):

• oReplace.SearchString = “^\sA", "^\s\d”
• oReplace.SearchString = “^\sA" OR "^\s\d”
• oReplace.SearchString = ("^\sA", "^\s\d")
• oReplace.SearchString = Array("^\sA", "^\s\d")

Is it possible? I’m a layman, so please explain in simple terms.

Thanks,

[Lupp]

In RegEx the comma in most places and the 2-letter-word OR always simply are literals without special functionality.
For a list of alternatives the list separator is the ‘Pipe’ symbol |.
For the “layman”: You cannot use Regex successfully without studying them to some depth. There are lots of tutorials in the web. Specifically good information, though probably a bit too complete for a beginner you find here: Regex Tutorial - Literal Characters and Special Characters .
Concerning RegEx usage in LibO specifically, you only need to know that the “flavor” is ICU.
See Regular Expressions - Old location of the ICU User Guide

[eu2020]

It worked, thanks

[eu2020]

Answer: make one giant regex

Thanks Lupp in the comments.

[Lupp]

You should try to avoid the “giant RegEx” by finding common parts and few patterns describing the same set of accepted strings. In many realistic cases a list can be shortened with the help of patterns.
Concerning the original example ^\s(A|\d) would do.

[eu2020]

I understand. Thanks for the explanation.