Replace multiline text in a file using Powershell without using Regex

Question

I have the following Powershell script:

$oldCode =  @"
            <div id="time_estimate">
                <!-- some large table -->
            </div>
"@

$newCode = @"
            <div id="time_estimate">
                                <!-- nested divs and spans -->
                                <div id="contact-form">

                                        <?php include "contact-form.php"; ?>
                                </div>
                        </div>
"@

ls *.html | foreach { 
        $fileContent = [System.Io.File]::ReadAllText($_.FullName)
        $newFileContent = $fileContent.Replace($oldCode, $newCode)
        [System.Io.File]::WriteAllText($_.FullName, $newFileContent)
        Write-Host  "`r`n"
        Write-Host  "Processed - $($_.Name)...`r`n" }

This doesn't seem to be replacing the text. Is it an issue with the multiline strings, or the limits of the Replace() method? I would prefer to do the replace without bringing in regex.

KevinD · Accepted Answer · 2013-12-17 19:01:49Z

5

What version of PowerShell are you using? If you're using v3 or higher, try this:

ls *.html | foreach { 
    $fileContent = Get-Content $_.FullName -Raw
    $newFileContent = $fileContent -replace $oldCode, $newCode
    Set-Content -Path $_.FullName -Value $newFileContent
    Write-Host  "`r`n"
    Write-Host  "Processed - $($_.Name)...`r`n" 
}

answered Dec 17, 2013 at 19:01

KevinD

3,1732 gold badges24 silver badges26 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

ajk Over a year ago

This is pretty much how I would approach the issue too. The -Raw switch is a nice addition to Powershell v3. In v2, I often relied on Get-Content <file> | Out-String to achieve similar results. There's a little more detail in this answer to a related question.

KalenGi Over a year ago

I upgraded to version 3 and applied this method

Henrik Over a year ago

If anyone wonders about a one-liner for ONE file it could go like this: (Get-Content -raw file.txt) -replace $oldCode, $newCode | Set-Content file.txt

Community · Accepted Answer · 2017-05-23 11:46:11Z

1

For Pete's sake, don't even think about using regex for HTML.

The problem you met is that reading a file will provide you an array of strings. Replace() doesn't know about arrays, so you got to work it by hand. You could create a big string with -join like so,

$fileContent = [System.Io.File]::ReadAllText($_.FullName)
$theOneString = $fileContent -join ' '
$theOneString.Replace($foo, $bar)

... But this will mess up your line breaks. Then again, you could reformat the string with HTML Tidy.

The manual way is to iterate the source array row by row. Until you find the <div>, copy the contents into new destination array. After finding the replacable part, insert rest of the new stuff into the destination array. Keep reading and discarding the source array untill you find the </div> and copy all the rest into the destination array. Finally save the destination array's contents and you are done.

edited May 23, 2017 at 11:46

CommunityBot

11 silver badge

answered Dec 17, 2013 at 17:51

vonPryz

24.3k10 gold badges58 silver badges69 bronze badges

1 Comment

KalenGi Over a year ago

Thanks very much for the link to that hilarious answer! The laughter has helped quite a bit...

Ansgar Wiechers · Accepted Answer · 2013-12-18 19:07:43Z

1

I wouldn't use string replacements for modifying HTML code. To many things that could develop in unexpected directions. Try something like this:

$newCode = @"
<!-- nested divs and spans -->
<div id="contact-form">
  <?php include "contact-form.php"; ?>
</div>
"@

Get-ChildItem '*.html' | % {
  $html = New-Object -COM HTMLFile
  $html.write([IO.File]::ReadAllText($_.FullName))
  $html.getElementById('time_estimate').innerHTML = $newCode
  [IO.File]::WriteAllText($_.FullName, $html.documentElement.outerHTML)
}

If needed you can can prettify the HTML by using Tidy:

$newCode = @"
<!-- nested divs and spans -->
<div id="contact-form">
  <?php include "contact-form.php"; ?>
</div>
"@

[Reflection.Assembly]::LoadFile('C:\path\to\Tidy.dll') | Out-Null
$tidy = New-Object Tidy.DocumentClass

Get-ChildItem '*.html' | % {
  $html = New-Object -COM HTMLFile
  $html.write([IO.File]::ReadAllText($_.FullName))
  $html.getElementById('time_estimate').innerHTML = $newCode
  $tidy.ParseString($html.documentElement.outerHTML)
  $tidy.SaveFile($_.FullName) | Out-Null
}

edited Dec 18, 2013 at 19:07

answered Dec 17, 2013 at 21:22

Ansgar Wiechers

202k27 gold badges286 silver badges363 bronze badges

1 Comment

KalenGi Over a year ago

This actually works, but messes with the formatting of the document by stripping out leading whitespace.

Collectives™ on Stack Overflow

Replace multiline text in a file using Powershell without using Regex

3 Answers 3

3 Comments

1 Comment

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

3 Comments

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related