The PowerShell -replace operator is an excellent function for substituting characters or sequences of characters in a string. The essential advantage of the operator is that it accepts the substitution of a substring with literal strings, wildcards, and regular expressions.
The catch of using the -replace operator to remove characters is to replace a substring or regex pattern with an empty string (“”).
The -replace Operator
The operator has the following syntax:
|
1 |
"<input_string>" -replace "<regex_pattern>", "<substitute>" |
Where <regex_pattern> is the regular expression (or literal string), we want to find and replace it with <substitute> substring. In our examples, we will use an empty string for <substitute> to effectively remove characters captured by the <regex_pattern>.
Note: The -replace operator is case-sensitive by default. If you want your pattern-matching process to be case-insensitive, use -creplace operator.
Let’s now see some examples.
Examples
This Section covers examples of using the -replace operator to remove characters from a string. In each example, the code comes first, followed by an explanation of the pattern used.
Example 1: Remove all instances of a character or sequence of characters
|
1 2 3 4 5 6 7 |
$string = "This event is awesome. The place is awesome as well" # Replace all "i" characters $result1 = $string -replace "i", "" Write-Output $result1 # Replace the substring "is" $result2 = $string -replace "is", "" Write-Output $result2 |
Output:
Ths event s awesome. The place s awesome as well Th event awesome. The place awesome as well
In the example above, we have seen how to replace a character (replace “i” in the first case) and a sequence of characters (remove all “is” substrings in the second case). Note that the second example replaces all occurrences of “is” in the string, not only the word “is”. If you want to replace a word, see Example 2.
Example 2: Remove a word in a string
|
1 2 3 4 5 6 7 |
$string = "If you need my help, please let me know and I will be helping" # Remove the word "help" $result = $string -replace "\bhelp\b", "" Write-Output $result # Replace the word "is" $result = $string -replace "\bis\b", "" Write-Output $result |
Output:
If you need my , please let me know and I will be helping If you need my help, please let me know and I will be helping
The patterns: “\bhelp\b” and “\bis\b”
Explanation: The “\b” sequence matches the boundary between words in a string. Therefore, the patterns “\bhelp\b” and “\bis\b” matches the words “help” and “is”, respectively. Note that this matches the whole words, not the substrings. That is why the substring “help” in the word “helping” is not removed in the second case.
Example 3: Remove the first and the last word in a string
|
1 2 3 4 |
$string = "If you need my help, please let me know." # Remove the first word $result = $string -replace "^\w+\b", "" Write-Output $result |
Output:
you need my help, please let me know.
The pattern: “^\w+\b”
Explanation: The “^” anchor matches the beginning of the string, \w+ matches one or more alphanumeric characters, and \b matches the word boundary. Therefore, “^\w+\b” will match the first word in our string.
|
1 2 3 4 |
$string = "If you need my help, please let me know.&" # Remove the last word - even with a punctuation mark(s) at the end of the input string. $result = $string -replace "\b(\w+)\W*$", "" Write-Output $result |
Output:
If you need my help, please let me
The pattern: “\b(\w+)\W*$”
Explanation: $ matches the end of our string. Therefore, “\b(w+)$” captures the last word. The parentheses capture groups (we will cover that in the last example). We also added the sequence “\W*” to capture zero or more non-alphanumeric characters like punctuation marks.
Example 4: Replace multiple substrings or words
|
1 2 3 |
$string = "If you need my help, please let me know, and I will help. I will always help." $result = $string -replace "help|know|will", "" Write-Output $result |
Output:
If you need my , please let me and I . I always .
The code above replaces all substrings given in the pattern. If you want to capture words, use the word boundary anchor (\b) discussed in Example 2 with something like this: “\bhelp\b|\bknow\b|\bwill\b”.
You can remove all words starting with a given character with a code like this.
|
1 2 3 |
$string = "If you need my help, please let me know, and I will help. I will always help." $result = $string -replace "m\w+|h\w+", "" Write-Output $result |
Output:
If you need , please let know, and I will . I will always .
Example 5: Remove all punctuations
|
1 2 3 |
$string = "If yo%u n^ee%d& my( hel*p, p'l:e?a{};s<e, l[et- me !k@n>ow." $result = $string -replace "[^\w\d\s]+", "" Write-Output $result |
Output:
If you need my help please let me know
The pattern: “[^\w\d\s]+”
Explanation: When the anchor “^” is used at the beginning of the set operator [ ], it means negation. Therefore, “[^\w\d\s]+” matches any character that is not word character (\w), digit (\d) or whitespace (\s). The pattern, therefore, effectively removes punctuation marks that fall under \W.
Example 6: Remove numbers
There are three cases involving how to remove numbers using the -replace operator. They include
Case 1: Remove unsigned integers like 0, 44, 1096
|
1 2 3 4 |
$string = "If y60ou ne99ed my he78lp, pl78ea6se le2t me kn18ow." # You can also use "[0-9]" instead of "\d". $result = $string -replace "\d+", "" Write-Output $result |
Output:
If you need my help, please let me know.
The pattern: “\d+” or “[0-9]+”
Explanation: The pattern matches one or more number characters.
Case 2: Remove signed and unsigned integers, e.g., +67, -78, 45, 70
Case 2 fixes the limitation of Case 1 – Case 1 cannot capture signed integers.
|
1 2 3 |
$string = "If y-5ou ne-67ed my help, pl45ea+98se le-67t me kn18ow." $result = $string -replace "\+?-?\d+", "" Write-Output $result |
Output:
If you need my help, please let me know.
The pattern: “\+?-?\d+”
Explanation: The pattern captures numbers (\d+) with or without the sign (\+?-?).
Case 3: Remove signed and unsigned floats and integers like -3.14, 67, 4.6 and -78
Case 3 solves the bottlenecks of Cases 1 and 2 – Case 3 capture signed and unsigned decimal numbers and integers.
|
1 2 3 |
$string = "If you ne3.142ed my h-4.6elp, plea+6.76se le-6t me kn0.27ow." $result = $string -replace "\+?-?\d+(\.\d+)?", "" Write-Output $result |
Output:
If you need my help, please let me know.
Example 7: Remove all white spaces in a string
|
1 2 3 |
$string = "If you need my help, please let me know" $result = $string -replace "\s+", "" Write-Output $result |
Output:
Ifyouneedmyhelp,pleaseletmeknow
Example 8: Capture groups and remove some of them
Grouping in regular expressions is used to construct substrings from an input string. Groups in PowerShell regex are captured with parentheses “()”. The captured groups can be identified using integers (the default) or names.
Consider a string containing a salutation, first name, last name, and state of an individual separated by whitespaces. For example,
“Mr Mitch McConnell Kentucky”
We wish to capture the four information pieces by grouping and removing the salutation and state.
Let’s start by using integers to identify the captured groups.
|
1 2 3 |
$name = "Prof John Smith Orlando" $new_name = $name -replace "^(\w+)\s+(\w+)\s+(\w+)\s+(\w+)$", "$2 $3" Write-Output $new_name |
Output:
John Smith
The pattern: “^(\w+)\s+(\w+)\s+(\w+)\s+(\w+)$”
Explanation: The pattern captures four groups (shown by the four parenthesis pairs). We then replace the four groups with only two groups, $2 and $3, effectively removing the $1 (salutation) and $4 (state) groups.
Instead of using integers, we can use named groups, as follows.
|
1 2 3 |
<br>$name = "Mrs Alice Kamalie Texas" $new_name = $name -replace "(?<salutation>\w+)\s+(?<first_name>\w+)\s+(?<last_name>\w+)\s+(?<state>\w+)", '${first_name} ${last_name}' Write-Output $new_name |
Output:
Alice Kamalie
Conclusion
This article discussed using the -replace operator to remove characters from a string in PowerShell. After covering the examples given, you should be able to remove characters from strings based on your needs. However, if you need more information, you may find the following reference links useful to learn more about -replace operator and regex: