Say I have the following string:
"Hello how are you."
Since MS Word allows for regular expressions, I can use "*" to find the complete string. But what if I want to exclude the delimiters (the quotes)? I'm afraid that MS Word doesn't support either of the two methods explained here. My question is: would there be any way to do this in one search query?
Thanks in advance.
There are different ways to achieve what you want. Here is one way to find text in VBA Word without the dilimiters using Regex. Let's say you have the following text in Word Document (do not copy and paste it from here as the the website distorts the Double quotes. See the screenshot)
This is a sample
"This is another Sample"
"Wake me up before you go go"
"War of the worlds"
The code to return text using Regex between two quotes is as follows
Sub FindText()
Dim regEx, Match, Matches
Set regEx = New RegExp
regEx.Pattern = "([^“]*)(?=\”)"
regEx.IgnoreCase = False
regEx.Global = True
Set Matches = regEx.Execute(ActiveDocument.Range.Text)
For Each Match In Matches
Debug.Print Match.Value
Next
End Sub
and if you want to say find "Wake me up before you go go" without quotes then you can use this as well
Sub FindText()
Dim regEx, Match, Matches
Dim searchText As String
searchText = "Wake me up before you go go"
Set regEx = New RegExp
regEx.Pattern = "([^“]*)(?=\”)"
regEx.IgnoreCase = False
regEx.Global = True
Set Matches = regEx.Execute(ActiveDocument.Range.Text)
For Each Match In Matches
If Trim(Match.Value) = (searchText) Then
Debug.Print "Found"
End If
Next
End Sub
NOTE: The website distorts the actual double quote so I am posting screenshots.
FOLLOWUP
For the sample file that you posted, use this code
Sub FindText()
Dim regEx, Match, Matches
Set regEx = New RegExp
regEx.Pattern = """([^""]*)"""
regEx.IgnoreCase = False
regEx.Global = True
Set Matches = regEx.Execute(ActiveDocument.Range.Text)
For Each Match In Matches
Debug.Print Match.SubMatches(0)
Next
End Sub
Sample File can be downloaded from here. Please note that this link will be active for 7 days.
Sample File
HTH
Sid
You are wrong. Word does support some wildcards, ? for a single character and * for a series of characters.
This is not a regular expression
means no lookbehind and no lookahead
While there will never be everything in Ms-Word that you want, e.g. like this one where you want to find something else, but want to select only a part of it, there are always macros which you can program to accomplish your task.
Add the following VBA code to your document. You can add a custom button on the toolbar to call it.
Sub FindSpecial()
FindSpecialA
End Sub
Private Sub FindSpecialA(Optional text As String)
Dim ToFind As String
ToFind = InputBox("Enter the text you want to find in double-quotes (without double-quotes):" & vbCrLf & vbCrLf & "(Enter * to match anything within double-quotes)", "Find", text)
If ToFind = "" Then Exit Sub
Selection.Find.ClearFormatting
With Selection.Find
.text = """" & ToFind & """"
.Replacement.text = ""
.Forward = True
.Wrap = wdFindContinue
.Format = False
.MatchCase = False
.MatchWholeWord = False
.MatchWildcards = False
.MatchSoundsLike = False
.MatchAllWordForms = False
If ToFind = "*" Then
.text = "[“""]*[”""]"
.MatchWildcards = True
End If
End With
Selection.Find.Execute
If Selection.Find.Found Then
Selection.MoveStart unit:=wdCharacter, Count:=1
Selection.MoveEnd unit:=wdCharacter, Count:=-1
FindSpecialA ToFind
Else
MsgBox "Not found!"
End If
End Sub
EDIT:
Updated the code to handle wildcard * matches.
Some versions of MS Word support regex-style groups with their "search with wildcards" option, meaning that if you can create a search expression between two quotes -- the one that works for me is "?#" -- you can change it to "(?#)" and enter \1 for the replace text. This will replace the text that was found with just the text that matches the expression between the parentheses, getting rid of your quote marks. (MS Word's ?# is equivalent to .* (non-greedy) in common regex.)
This works for me in Word 2008 for Mac, but I don't have a guide to which versions of Office support this syntax.
Beware! In this search form, Word does not equate the straight quotes on your keyboard with the curly quotes it inserts in order to look pretty. You will need to either turn off "smart quotes" for this document, or construct your search phrase by cutting and pasting the opening and closing quote characters from your document.
I have had luck using color (Format < Font in the Search dialog box while you are clicked into Find What) to solve problems like this. Execute search all content with delimiters (“*” with wildcards checked for this example) and replace using a non-black color like blue. Search and replace delimiter (in this case quote marks) color from blue to black. Perform changes to content in blue. Select all and change to black. If this comes up often, I would suggest macros on a toolbar for the step one (blue, take blue off delimiter) and step two (change all to black).
Related
I have trouble finding working solution for couple of hours now. I hope you will help me.
My problem:
I need to find and select in Word a whole sentence after providing the starting and ending strings of particular sentence.
For example, when my starting string is "People" and ending string is "apples." I expect Word to select the whole "People like red apples." sentence in my document. (If such a sentence exists)
For this purpose I prepared a macro which works almost like I want. The only problem is that it doesn't select the smallest possible set of characters (which I want it to do). To make it clear let's assume I have this text in my document: People like smoking. People like red apples.
Now, when I provide the starting and ending strings to the macro respectively as "People" and "apples.", it selects all the text, which contains 2 sentences mentioned above. That is my problem: I wanted it to select only the second sentence (People like red apples.), not both of them, even though they start with the same word. So, basically, I always want to select the shortest possible set of characters (which in this case is only the last sentence).
Here is a part of my macro in VBA:
`text_str = startStr & "*" & endStr
With Application.Selection.Find
.ClearFormatting
.Forward = True
.Wrap = wdFindContinue
.Text = text_str
.MatchWildcards = True
.MatchCase = True
.Execute
End With
I know the problem is with the Wildcards (or very limited set of regular expressions), so I also tried something like this as the search string:
text_str = "(" & startStr & "*){1}" & endStr
It also didn't help. I'm stuck here. :/
Thanks for any suggestions!
Selection.Find has something similar to regular expressions,
but in this case you must use real regular expressions.
The pattern (in this particular case) should be:
People[^.]+apples\.
I wrote an example macro, which:
Selects the whole text in the document and assigns it to src
variable (searched by the regex).
Sets the cursor at the beginning of the document.
Checks whether the pattern can be matched (regEx.Test).
Executes the regex.
Assigns the matched string to ret variable.
Displays it in a message box.
Below you have a complete macro. Probably you should change it to
select (find) the text matched (instead of the message box).
Sub Re()
Dim startStr As String: startStr = "People"
Dim endStr As String: endStr = "apples"
Dim pattern As String: pattern = startStr & "[^.]+" & endStr & "\."
Dim regEx As New RegExp
Dim src As String
Dim ret As String
Dim colMatches As MatchCollection
ActiveDocument.Range.Select
src = ActiveDocument.Range.Text
Selection.StartOf
regEx.pattern = pattern
If (regEx.Test(src)) Then
Set colMatches = regEx.Execute(src)
ret = "Match: " & colMatches(0).Value
Else
ret = "Matching Failed"
End If
MsgBox ret, vbOKOnly, "Result"
End Sub
I would like to merge my three regexes which clean text (empty lines, leading and trailing spaces etc.) into, if possible, one regex, or if it is not possible - into two.
My first regex is [ \t]+. It does this sort of cleaning.
My second regex is ^(?:[\t ]*(?:\r?\n|\r))+ Not image included since it won't catch anything if the previous regex has not run.
The third regex is ^[\s\xA0]+|[\s\xA0]+$. It does this sort of cleaning.
EDIT: I have forgotten to mention that in each case I replace match with nothing "".
EDIT 2: I use the following code in Word:
With selection
Dim RegEx As Object
Set RegEx = CreateObject("VBScript.RegExp")
RegEx.Global = True
RegEx.MultiLine = True
' clean selection
RegEx.Pattern = "[ \t]+"
.Text = RegEx.Replace(.Text, " ")
RegEx.Pattern = "^(?:[\t ]*(?:\r?\n|\r))+"
.Text = RegEx.Replace(.Text, "")
' the following is from http://stackoverflow.com/a/24049145/2657875
RegEx.Pattern = "^[\s\xA0]+|[\s\xA0]+$"
.Text = RegEx.Replace(.Text, "")
End With
The last regexps can be merged as
RegEx.Pattern = "^(?:[\t ]*(?:\r?\n|\r)?)*|[ \t]+$"
I do not think there can be a chance to merge all 3 in VBA since you are using two different replacement patterns.
If i am not wrong, you want all your lines/spaces/tabs/white lines to be matched and removed, so you could merge the input strings. Well, that's easy and can be done if you do use the following regex in your replace program/script/command:
/([\s\t]{0,50}\r?\n)+|\s+/s
The regex should work well on windows as well as linux based files.
Not pro but I use multiple regex one after another. If you are not familiar with below code than you should try.
Set regEx_ = new regExp
With regEx_
.Global = True
.MultiLine = True
.IgnoreCase = True
.Pattern = "Pattern 1"
TextLine = regEx_.replace(TextLine, "")
.Pattern = "Pattern 2"
TextLine = regEx_.replace(TextLine, "")
'and so on
End With
I am using Outlook 2010, and I am trying to write a macro to replace the font of text with a different one, if it matches a pattern.
The logic I am trying to apply is simple - in the user selected text, check for a pattern, and on match, change the font for the matched text.
So far I have been able to split the text and apply/check regex, but the replacement is something that I am not clear on how to do.
Dim objOL As Application
Dim objDoc As Object
Dim objSel As Object
Dim regEx As RegExp
Dim matches As MatchCollection
Dim m As Match
Dim lines As Variant
Dim ms As String
Set objOL = Application
Set objDoc = objOL.ActiveInspector.WordEditor
Set objSel = objDoc.Windows(1).Selection
lines = Split(objSel, Chr(13))
For i = 0 To UBound(lines) Step 1
Set regEx = New RegExp
With regEx
.Pattern = "\[(ok|edit|error)\](\[.*\])?" ' <-- this is just one regex, I want to be able to check more regexes
.Global = True
End With
If regEx.Test(lines(i)) Then
Set matches = regEx.Execute(lines(i))
For Each m In matches
ms = m.SubMatches(1)
' ms.Font.Italic = True
' <-- here is where I am not sure how to replace! :( -->
Next
End If
Next i
P.S there seems to be text-search (objSel.Find.Text)and replace (objSel.Find.Replacement.Text) methods in Selection object, but not pattern-search ! (or I am missing it)
--EDIT--
Adding a sample text
user#host> show some data
..<few lines of data>.. <-- these lines as-is (but monospaced)
[ok][2014-11-26 11:05:02]
user#host> edit some other data
[edit data]
user#host(data)% some other command
I want to convert the whole block to a monospaced font (like Courier New, or Consolas)
And change the part that begins with something#somewhere.. and till > or % to dimmer color,
(i.e in this example user#host> and user#host(data)% dimmer/grey)
The rest in that line to bold (show some data et al)
And, all the bracketed text followed by time-stamps (or without timestamps) similar to 2. (i.e, dim/grey)
This is getting closer to being done. The framework is here to make all sorts of changes now. Just need to get some of the regex patterns down to make the changes.
Sub FormatSelection()
Dim objMailItem As Outlook.MailItem
Dim objInspector As Outlook.Inspector: Set objInspector = Application.ActiveInspector
Dim objHtmlEditor As Object
Dim objWord As Object
Dim Range As Word.Selection
Dim objSavedSelection As Word.Selection
Dim objFoundText As Object
' Verify a mail object is in focus.
If objInspector.CurrentItem.Class = olMail Then
' Get the mail object.
Set objMailItem = objInspector.CurrentItem
If objInspector.EditorType = olEditorWord Then
' We are using a Word editor. Get the selected text.
Set objHtmlEditor = objMailItem.GetInspector.WordEditor
Set objWord = objHtmlEditor.Application
Set Range = objWord.Selection
Debug.Print Range.Range
' Set defaults for the selection
With Range.Font
.Name = "Courier"
.ColorIndex = wdAuto
End With
' Stylize the bracketed text
Call FormatTextWithRegex(Range, 2, "\[(.+?)\]")
' Prompt style text.
Call FormatTextWithRegex(Range, 2, "(\w+?#.+?)(?=[\>\%])")
' Text following the prompt.
Call FormatTextWithRegex(Range, 3, "(\w+?#.+?[\>\%])(.+)")
End If
End If
Set objInspector = Nothing
Set Range = Nothing
Set objHtmlEditor = Nothing
Set objMailItem = Nothing
End Sub
Private Sub FormatTextWithRegex(ByRef pRange As Word.Selection, pActionIndex As Integer, pPattern As String)
' This routine will perform a regex replacement on the text in pRange using pPattern
' on text based on the pactionindex passed.
Const intLightColourIndex = 15
Dim objRegex As RegExp: Set objRegex = New RegExp
Dim objSingleMatch As Object
Dim objMatches As Object
' Configure Regex object.
With objRegex
.IgnoreCase = True
.MultiLine = False
.Pattern = pPattern ' Example "\[(ok|edit|error)\](\[.+?\])?"
.Global = True
End With
' Locate all matches if any.
Set objMatches = objRegex.Execute(pRange.Text)
' Find
If (objMatches.Count > 0) Then
Debug.Print objMatches.Count & " Match(es) Found"
For Each objSingleMatch In objMatches
' Locate the text associated to this match in the selection so we can replace it.
Debug.Print "Match Found: '" & objSingleMatch & "'"
With pRange.Find
'.ClearFormatting
.Text = objSingleMatch.Value
.ClearFormatting
Select Case pActionIndex
Case 1 ' Italisize text
.Replacement.Text = objSingleMatch.Value
.Replacement.Font.Bold = False
.Replacement.Font.Italic = True
.Replacement.Font.ColorIndex = wdAuto
.Execute Replace:=wdReplaceAll
Case 2 ' Dim the colour
.Replacement.Text = objSingleMatch.Value
.Replacement.Font.Bold = False
.Replacement.Font.Italic = False
.Replacement.Font.ColorIndex = intLightColourIndex
.Execute Replace:=wdReplaceAll
Case 3 ' Bold that text!
.Replacement.Text = objSingleMatch.Value
.Replacement.Font.Bold = True
.Replacement.Font.Italic = False
.Replacement.Font.ColorIndex = wdAuto
.Execute Replace:=wdReplaceAll
End Select
End With
Next
Else
Debug.Print "No matches found for pattern: " & pPattern
End If
Set objRegex = Nothing
Set objSingleMatch = Nothing
Set objMatches = Nothing
End Sub
So we take what the user has selected and execute the macro. I have my Outlook configured with Word for the editor so that is tested for. Take the selected text and run the regex query against the text saving the matches.
The issue you had is what to do with the match once you found it. In my case since we have the actual text that matched we can run that through a find and replace using the selection once again. Replacing the text with itself instead styled as directed.
Caveats
My testing text was the following:
asdfadsfadsf [ok][Test]dsfadsfasdf asdfadsfadsfasdfasdfadsfadsf [ok][Test]dsfadsfasdf asdfadsfadsfasdf
I had to change your regex in your sample to be less greedy since it was matching both [ok][Test] sections. I don't know what kind of text you are working with so my logic might not apply to your situation. Test with caution.
You also had a comment that you needed to test multiple regexes... regexies.... I don't know what the plural is. Wouldn't be hard to create another function that calls this one for several patterns. Assuming this logic works repeating it should not be a big deal. I would like to make this work for you so if something is wrong let me know.
Code Update
I have changed the code so that the regex replacement is in a sub. So what the code does right now is change the selected text to courier and italisize text based on a regex. Now with how it is set up you can use the sub routine FormatTextWithRegex to make changes. Just need to update the pattern and action index which will perform the different styles. Will be updating this again soon with more information. Right now all that exists is the structure that I think you need.
Having issues with the bolding still but you can see the grey part is working correctly. Also the since this relies on highlighting the multiple calls to the function are causing an issue. Just not sure what it is.
I have just taken up the VBA route to automate a few day today tasks so pls excuse if I sound very naive
I'm trying to open a word document & then searching for a expression to highlight(Bold) it,however Im getting error "User defined type not defined"
I'm able to open the word document but unable to perform the pattern search.I have gathered bits & peices of code from internet, however its not working
I'm using Office 2013 & have added the Microsoft VBscript Reg Ex 5.5 in references.
The pattern Im searching is starting from "Dear" till ) is encountered.
Cheers #GoingMad#
Sub Pattern_Replace()
Dim regEx, Match, Matches
Dim rngRange As Range
Dim pathh As String, i As Integer
pathh = "D:\Docs\Macro.docx"
Dim pathhi As String
Dim from_text As String, to_text As String
Dim WA As Object, WD As Object
Set WA = CreateObject("Word.Application")
WA.Documents.Open (pathh)
WA.Visible = True
Set regEx = New RegExp
regEx.Pattern = "Dear[^0-9<>]+)"
regEx.IgnoreCase = False
regEx.Global = True
Set Matches = regEx.Execute(ActiveDocument.Range.Text)
For Each Match In Matches
ActiveDocument.Range(Match.FirstIndex, Match.FirstIndex + Len(Match.Value)).Bold = True
Next
End Sub
You need to escape the bracket ")" within the regex, using a back-slash:
regex.Pattern = "Dear[^0-9<>]+\)"
This is because it has a particular meaning within a regex expression.
I would personally also split the reference to the Word-Range across a few lines:
Set rngRange = ActiveDocument.Range
rngRange.Expand Unit:=wdStory
Set Matches = regex.Execute(rngRange.Text)
although this isn't necessary.
Consider the following text
Dear aunt sally ) I have gone to school.
Your regex pattern would be "Dear[^)]+"
Find the word Dear
Match Any character that is not ")"
Repeat
Refiddle here
This one will include the parenthesis. Dear[\w\s]+\)
Find the word Dear
Match Any Character or whitespace
Repeat as needed
Until a right parenthesis is found
You don't need regex for this - a wildcard Find/Replace in Word will do the job far more efficiently:
With WA.ActiveDocument.Range.Find
.ClearFormatting
.Text = "Dear[!\)]#\)"
.Replacement.ClearFormatting
.Replacement.Font.Bold = True
.Replacement.Text = "^&"
.Format = True
.Forward = True
.Wrap = wdFindContinue
.MatchWildcards = True
.Execute Replace:=wdReplaceAll
End With
has VBA any good mechanism for checking, if the content of a given Excel Cell matches a specific regex?
In my case i want to know, if some cell has the format
m
m2
m1234
In fact, there's just one defined letter at the beginning, followed by a not specified amount of numbers.
How do I put this into an If-Else construct?
If Doc.Cells(1,1).Value ..... ???
greets, poeschlorn
You can get at the VBScript RegExp objects via Tools->References & adding "Microsoft VBScript Regular Expressions 5.5"
Alternatively a quick way to do it, if you don't need to check for a subsequent letter as in `m1234X1 is:
if Doc.Cells(1,1).Value like "[a-zA-Z]#*" then ...
(This doesn't require a reference to anything)
I don't know VBA, but the regex [a-zA-Z][0-9]* might be able to match what you want.
Here is my RegexContains function. Pass it the cell and the pattern and it will return TRUE or FALSE if it contains it or not.
Function RegexContains(ByVal find_in As String, _
ByVal find_what As String, _
Optional IgnoreCase As Boolean = False) As Boolean
Application.ScreenUpdating = False
Dim RE As Object
Set RE = CreateObject("vbscript.regexp")
RE.Pattern = find_what
RE.IgnoreCase = IgnoreCase
RE.Global = True
RegexContains = RE.Test(find_in)
Application.ScreenUpdating = True
End Function
Now, I'm not sure exactly what you want to find in your example, but if you want to know if the cell contains a single letter followed by one or more letters, then you would use (assuming the cell is A1): =RegexContains(A1, "^\w\d+")
The ^ marks the start of the sentence
The \w marks a single alphabetic
character (a-zA-Z)
The \d+ marks one or more numeric
characters[0-9]
I hope this helps.