Powershell: Copy a string/word from a bottom line to a top line

Question

hello. I have several html files in C:\Folder1 And I want to copy a word/string below in a line above. For example I want to copy TWO-2 from the second line into the first line, replacing the word ONE-1

html code ... several lines

text <p> BEBE ONE-1 NERO <div> text...

 other  html code ... several lines

 text <table> OANA TWO-2 BOGDAN <tr> text

 again html code ... several lines

MUST BECOME

html code ... several lines

text <p> BEBE TWO-2 NERO <div> text...

 other  html code ... several lines

 text <table> OANA TWO-2 BOGDAN <tr> text

 again html code ... several lines

can this be done with PowerShell?

Answer 1

Hi @Suzana Eree ,

please try this (last try for today):

$sourcedir = "./Junk/1"  
$resultsdir = "./Junk/2"  
Get-ChildItem -Path $sourcedir -Filter *.html | ForEach-Object {  
    $content = Get-Content -Path $_.FullName -Raw  
    $replaceValue = ((Select-String -Pattern "(BEBE).+(NERO)" -InputObject $content).Matches.Value).Split(" ")[1]    
    $findReplacement = Select-String -Pattern "(OANA).+(BOGDAN)" -InputObject $content  
    $split = ($findReplacement.Matches.Value).Split(" ")  
    $i = 0  
    do {  
        $findBogdan = $split[$i]  
        $i++      
    } until ($findBogdan -eq "BOGDAN")  
    $replacementValue = ($split[1..($i - 2)]) -join ' '  
    $content.Replace("$replaceValue", "$replacementValue") | Out-File -FilePath $resultsdir\$($_.name)    
}  

----------

(If the reply was helpful please don't forget to upvote and/or accept as answer, thank you)

Regards
Andreas Baumgarten

Answer 2

sir, @Andreas Baumgarten Great answer. I modify your last code, both of the updates. One of the problem was the regex. Instead of "^*(BEBE).+(NERO)*" I put (BEBE).*(NERO) Same for the second regex.

$sourcedir = "C:\Folder1\"  
 $resultsdir = "C:\Folder2\"  
 Get-ChildItem -Path $sourcedir -Filter *.html | ForEach-Object {  
     $content = Get-Content -Path $_.FullName -Raw  
     $findReplace = Select-String -Pattern "(BEBE).*(NERO)" -InputObject $content  
     $replaceValue = ($findReplace.Matches.Value).Split(" ")[1]      
     $findReplacement = Select-String -Pattern "(OANA).*(BOGDAN)" -InputObject $content  
     $replacementValue = ($findReplacement.Matches.Value).Split(" ")[1]  
     $content.Replace("$replaceValue", "$replacementValue") | Out-File -FilePath $resultsdir\$($_.name)    
 }  

OR

$sourcedir = "C:\Folder1\"  
 $resultsdir = "C:\Folder2\"  
 Get-ChildItem -Path $sourcedir -Filter *.html | ForEach-Object {  
     $content = Get-Content -Path $_.FullName -Raw    
     $replaceValue = ((Select-String -Pattern "(BEBE).*(NERO)" -InputObject $content).Matches.Value).Split(" ")[1]    
     $replacementValue = ((Select-String -Pattern "(OANA).*(BOGDAN)" -InputObject $content).Matches.Value).Split(" ")[1]  
     $content.Replace("$replaceValue", "$replacementValue") | Out-File -FilePath $resultsdir\$($_.name)    
 }  

THE ONLY PROBLEM IN THIS CASE OF POWESHELL CODE IS:

If I have

BEBE ONE-1 NERO

and 2 words between OANA and BOGDAN on the second line, such as:

OANA XX TWO-2 BOGDAN

in this case the result is BEBE XX NERO, but should be BEBE XX TWO-2 NERO

So, the regex is parsing only one word between OANA and BOGDAN, but he should have taken everything between them