remove the first character of each .txt file with Powershell

I have 50 files in a folder that has to be loaded to SQL. In most of the file(but not in all) the very first character is a hashtag that i would like to delete because it creates difficulty in the column mapping.

So I would not want to loop through all the lines of the file just delete the very first character in case it is a hashtag and save the file.

So i need to analyse also if the first character is a hashtag. If it is, then it should be deleted, if it is not, then i should simply exit from the file.

What is the fastest method for this?

I tried these:

$Path = "D:\folder\myfolder"
    $OldText = "#"
    $NewText = ""

--> this loops through all the file looking for hashtags

(Get-Content 'D:\folder\myfoldermyfile.txt' -raw) -replace '^.' | Set-Content 'D:\folder\myfoldermyfile.txt'

--> this removed the first character of each line, so its not good for me.

Solution

To complement Theo's helpful answer with an eye toward performance:

The only immediate problem with your code was the use of regex '^.', because the . metacharacter matches any character (except a newline, by default); therefore, using '^#' instead of '^.' fixes your problem by only matching files whose first character is # (and by not specifying a substitution operand to the -replace operator, the matched # is replaced with the empty string and therefore effectively removed).

Therefore, applied to all *.txt files in D:\folder\myfolder:

# Note:
# * '^#' instead of '^.'
# * -NoNewLine to prevent Set-Content from writing an extra newline.
# * CAVEATS: 
#   * UPDATES THE FILES IN-PLACE - to be safe, keep backup copies.
#   * The *character encoding may change*, because Set-Content applies
#     its default encoding, irrespective of the input encoding.
#     USE -Encoding AS NEEDED.
Get-ChildItem 'D:\folder\myfolder\*.txt' |
  ForEach-Object {
    ($_ | Get-Content -Raw) -replace '^#' | 
      Set-Content -NoNewLine -LiteralPath $_.FullName
  }

Performance-wise, there are two problems with the above:

Even though using -Raw with Get-Content is the most efficient way to read a file in full, as a single, multiline string (and analogously, so is writing the file content in full as a single string), you don't need to do that if all you want to do is to examine the first character.
By using the results of the -replace operation unconditionally, you end up rewriting even the files that don't need it, i.e. those that do not start with # (if -replace cannot find a match, it passes the input string through).

You can address these issues as follows, building on mclayton's suggestion to (initially) only read the first line of each file:

# Same CAVEATS as above apply.
Get-ChildItem 'D:\folder\myfolder\*.txt' |
  ForEach-Object {
    if (($_ | Get-Content -First 1) -match '^#') {
      ($_ | Get-Content -Raw).Substring(1) | 
        Set-Content -NoNewLine -LiteralPath $_.FullName
    }
  }

Note:

This optimization only makes sense if you expect a good portion of the input files not to need rewriting. If most of them do, the unoptimized solution may perform better.