BAMCIS.ByteArrays.psm1
Function Set-ByteArrayPadding { <# .SYNOPSIS Pads an array with a leading byte so that the resulting array is the length specified. .DESCRIPTION This cmdlet takes an input array, creates a new array initialized with the padding byte specified, and then copies the supplied input array to the end of the new array. So if the array @(0x01, 0x02) was supplied with a padding byte of 0x03 and a length of 4, the resulting array would be @(0x03, 0x03, 0x01, 0x02). The original input array is unchanged and the resulting data is a new object returned to the pipeline if used with InputObject or supplied via the pipeline. The original array can be modified by use of the ReferenceObject parameter. .PARAMETER InputObject The array that needs to be padded. .PARAMETER ReferenceObject A reference to the array that needs to be padded. If this parameter is supplied, no output is returned to the pipeline .PARAMETER Length The desired length of the resulting array. .PARAMETER Padding The byte value used to pad the left hand side of the array. This defaults to 0x00. .PARAMETER PadEnd Instead of padding the left side of the array, the right side of the array is padded, so for an input of @(0x01) and a lenght of 4, the resulting array would be @(0x01, 0x00, 0x00, 0x00). .EXAMPLE $Arr = @(0x01) $Arr = $Arr | Set-ByteArrayPadding -Length 4 The resulting array in $Arr is length 4 and its contents are: @(0x00, 0x00, 0x00, 0x01) .EXAMPLE $Arr = @(0x01) Set-ByteArrayPadding ([ref]$Arr) -Length 2 -Padding 0xFF -PadEnd The variable $Arr is modified by reference so that the original variable's contents after the cmdlet are @(0x01, 0xFF). Here the array is padded on the end with the specified 0xFF byte. .INPUTS System.Byte[] .OUTPUTS None or System.Byte[] .NOTES AUTHOR: Michael Haken LAST UPDATE: 1/19/2018 #> [CmdletBinding()] [OutputType([System.Byte[]])] Param( [Parameter(Position = 0, ValueFromPipeline = $true, ParameterSetName = "Input")] [ValidateNotNull()] [System.Byte[]]$InputObject = @(), [Parameter(Position = 0, ParameterSetName = "Ref", Mandatory = $true)] [Ref]$ReferenceObject, [Parameter(Mandatory = $true, Position = 1)] [System.Int32]$Length, [Parameter(Position = 3)] [ValidateNotNull()] [System.Byte]$Padding = 0x00, [Parameter()] [Switch]$PadEnd ) Begin { } Process { switch ($PSCmdlet.ParameterSetName) { "Input" { if ($InputObject.Length -lt $Length) { [System.Byte[]]$NewData = [System.Byte[]](,$Padding * $Length) [System.Int32]$CopyStartIndex = $Length - $InputObject.Length if ($PadEnd) { $CopyStartIndex = 0 } # Copy the original data to the new data array starting at the padding count index # which will pad the new array with leading padding bytes [System.Array]::Copy($InputObject, 0, $NewData, $CopyStartIndex, $InputObject.Length) $InputObject = $NewData } Write-Output -InputObject $InputObject break } "Ref" { if ($ReferenceObject.Value.Length -lt $Length) { [System.Byte[]]$NewData = [System.Byte[]](,$Padding * $Length) [System.Int32]$CopyStartIndex = $Length - $ReferenceObject.Value.Length if ($PadEnd) { $CopyStartIndex = 0 } [System.Array]::Copy($ReferenceObject.Value, 0, $NewData, $CopyStartIndex, $ReferenceObject.Value.Length) $ReferenceObject.Value = $NewData } break } default { Write-Error -Exception (New-Object -TypeName System.ArgumentException("Unknown parameter set, $($PSCmdlet.ParameterSetName), for $($MyInvocation.MyCommand).")) -ErrorAction Stop } } } End { } } Function Remove-ByteArrayPadding { <# .SYNOPSIS Removes padding from the beginning or end of a byte array. .DESCRIPTION This cmdlet removes the padding from the beginning or end of an array and provides a new or modified array with those bytes removed. The caller can specify the byte that is used as padding and provide the input array either through the pipeline, parameter, or by reference. When passed by reference, nothing is returned to the pipeline. .PARAMETER InputObject The byte array to remove padding from. .PARAMETER ReferenceObject The reference to a byte array that will have padding removed from it. The reference will point to a different location in memory after the cmdlet is complete if any modifications have been done. .PARAMETER FromEnd Specifies that padding is removed from the tail end of the array instead of the beginning. .PARAMETER Padding The byte character that is used as padding to be removed. This defaults to 0x00. .EXAMPLE $Arr = @(0x00, 0x00, 0x00, 0x01) Remove-ByteArrayPadding -InputObject $Arr The results of this cmdlet will produce a new array with contents @(0x01). .EXAMPLE $Arr = @(0x00, 0x00, 0x00, 0x01, 0xFF, 0xFF) ([ref]$Arr) | Remove-ByteArrayPadding -Padding 0xFF -FromEnd This example demonstrates several things. First the input array can be passed by reference through the pipeline. After the cmdlet complete, the variable $Arr will contain @(0x00, 0x00, 0x00, 0x01). The cmdlet specified the padding character to 0xFF and it removed the padding from the end of the array instead of the beginning. .INPUTS System.Byte[] .OUTPUTS System.Byte[] or None .NOTES AUTHOR: Michael Haken LAST UPDATE: 1/22/2018 #> [CmdletBinding()] [OutputType([System.Byte[]])] Param( [Parameter(ValueFromPipeline = $true, Position = 0, ParameterSetName = "Input")] [ValidateNotNull()] [System.Byte[]]$InputObject, [Parameter(ValueFromPipeline = $true, Position = 0, ParameterSetName = "Ref")] [ValidateNotNull()] [Ref]$ReferenceObject, [Parameter()] [Switch]$FromEnd, [Parameter()] [ValidateNotNull()] [System.Byte]$Padding = 0x00 ) Begin { [System.Byte[]]$Bytes = @() } Process { switch ($PSCmdlet.ParameterSetName) { "Input" { $Bytes += $InputObject break } "Ref" { $Bytes = $ReferenceObject.Value break } default { Write-Error -Exception (New-Object -TypeName System.ArgumentException("Unknown parameter set, $($PSCmdlet.ParameterSetName), for $($MyInvocation.MyCommand).")) -ErrorAction Stop } } } End { if ($FromEnd) { $StartIndex = $Bytes.Length - 1 while ($StartIndex -ge 0 -and $Bytes[$StartIndex] -eq $Padding) { $StartIndex-- } if ($StartIndex -ge 0) { # Since StartIndex is the index where the first non zero byte is, add 1 to make it a length [System.Byte[]]$FinalBytes = New-Object -TypeName System.Byte[] -ArgumentList ($StartIndex + 1) [System.Array]::Copy($Bytes, 0, $FinalBytes, 0, $StartIndex + 1) $Bytes = $FinalBytes } } else { $StartIndex = 0 while ($StartIndex -lt $Bytes.Length -and $Bytes[$StartIndex] -eq $Padding) { $StartIndex++ } if ($StartIndex -lt $Bytes.Length) { [System.Byte[]]$FinalBytes = New-Object -TypeName System.Byte[] -ArgumentList ($Bytes.Length - $StartIndex) [System.Array]::Copy($Bytes, $StartIndex, $FinalBytes, 0, $Bytes.Length - $StartIndex) $Bytes = $FinalBytes } } switch ($PSCmdlet.ParameterSetName) { "Input" { Write-Output -InputObject $Bytes break } "Ref" { $ReferenceObject.Value = $Bytes break } } } } Function Out-Hex { <# .SYNOPSIS Outputs a byte array to stdout in a hex representation. .DESCRIPTION This cmdlet takes a byte array and writes it to the console in line lengths that are easier to read. If the input array is sent through the pipeline, it does not need to be "protected" with a ',' to stop it from being unrolled. .PARAMETER InputObject The byte array to write to the console in hex. .PARAMETER Length The maximum number of bytes written to each line. This defaults to 8. .PARAMETER Delimiter The character used to separate each byte representation. This defaults to ' '. .EXAMPLE $Arr = @(0x01, 0x02, 0x03, 0x04) $Arr | Out-Hex -Length 2 This writes out: 01 02 03 04 In this case, the array $Arr is unrolled as it is sent to Out-Hex, so it ends up processing 4 separate Length 1 byte arrays. .EXAMPLE $Arr = @(0x01, 0x02, 0x03, 0x04, 0x05) $Arr | Out-Hex -Length 3 This writes out: 01 02 03 04 05 In this case, the array $Arr is not unrolled as it is sent to Out-Hex, so it ends up processing 1 array of length 5. .INPUTS System.Byte[] .OUTPUTS None .NOTES AUTHOR: Michael Haken LAST UPDATE: 1/19/2018 #> [CmdletBinding()] [OutputType()] Param( [Parameter(ValueFromPipeline = $true, Position = 0)] [ValidateNotNull()] [System.Byte[]]$InputObject, [Parameter()] [ValidateRange(1, [System.Int32]::MaxValue)] [System.Int32]$Length = 8, [Parameter()] [ValidateNotNull()] [System.Char]$Delimiter = ' ' ) Begin { # This method will handle an array passed via pipeline whether it is unrolled or not # Keep track on how many items we've added to a single line so we know when to add a # new line [System.Int32]$LineCounter = 0 [System.Text.StringBuilder]$SB = New-Object -TypeName System.Text.StringBuilder } Process { for ($i = 0; $i -lt $InputObject.Length; $i++) { $SB.Append($InputObject[$i + $j].ToString("X2") + $Delimiter) | Out-Null $LineCounter++ # This pipeline input has finished the line, reset the line counter # and add the new line if ($LineCounter -eq $Length) { $LineCounter = 0 # Trim the extra delimiter $SB.Length = $SB.Length - 1 # Add the new line $SB.Append("`r`n") | Out-Null } } } End { # Trim off the extra delimiters or new lines if they were added Write-Host ($SB.ToString().Trim()) } } Function ConvertTo-OIDString { <# .SYNOPSIS Converts a byte array into an OID string. .DESCRIPTION This cmdlet accepts a byte array which is converted into an OID string. .PARAMETER InputObject The byte array to convert. .EXAMPLE $Arr = @(0x2A, 0x86, 0x48, 0x86, 0xF7, 0x0D, 0x01, 0x01, 0x01) $OID = ConvertTo-OIDString $Arr This produces 1.2.840.113549.1.1.1 as the OID string, which is the OID for RSA Encryption .INPUTS System.Byte[] .OUTPUTS System.String .NOTES AUTHOR: Michael Haken LAST UPDATE: 1/20/2018 #> [CmdletBinding()] [OutputType([System.String])] Param( [Parameter(Position = 0, Mandatory = $true, ValueFromPipeline = $true)] [System.Byte[]]$InputObject ) Begin { # Make a new array to hold all of the input in case the array is unrolled by the pipeline [System.Byte[]]$Bytes = @() } Process { # Add each input to the collector in case the byte array is unrolled on the pipeline $Bytes += $InputObject } End { if ($Bytes.Length -lt 3) { Write-Error -Exception (New-Object -TypeName System.ArgumentException("The input byte array must have a length of at least 3 bytes.")) -ErrorAction Stop } # Now that we have all the bytes to make the OID string, start processing them $Data = "" for ($i = 0; $i -lt $Bytes.Length; $i++) { if ($i -eq 0) { # Oid A.B.6.1.... # Where A = 1 and B = 3 # The first byte is computed by (A x 40) + B, first node times 40 plus the second node # To work backwords, do a modulo 40 to find the remainder, then subtract that from the value # to get a number divisible by 40 evenly to get the first node $Val = [System.Convert]::ToUInt32($Bytes[$i]) [System.Int32]$SecondNode = $Val % 40 [System.Int32]$FirstNode = ($Val - $SecondNode) / 40 $Data += "$FirstNode.$SecondNode" } else { # All of the rest of the nodes are either less than or equal to 127, in which case they # contain the value directly, or they are 128 and greater and the value is stored in # multiple bytes if (($Bytes[$i] -band (1 -shl 7)) -ne 0) { # Build new array to hold the bytes [System.Byte[]]$Arr = @($Bytes[$i]) $OIDCounter = 1 while (($Bytes[$i + $OIDCounter] -band (1 -shl 7)) -ne 0) { # We don't care about the left most bit, it's just a marker # that it is part of this node $Arr += ($Bytes[$i + $OIDCounter] -band 0x7F) $OIDCounter++ } # This byte did not have a leading 1 and is the last byte # to make up the variable length $Arr += ($Bytes[$i + $OIDCounter] -band 0x7F) # Skip however many bytes beyond the first we added to the array $i += $Arr.Length - 1 # Take 7 bits from each byte and concatenate them # This is the easiest way to make sure we grab 7 bits from each byte and concatenate # them and is just as fast as bit shifting the individual bytes [System.Text.StringBuilder]$SB = New-Object -TypeName System.Text.StringBuilder for ($j = 0; $j -lt $Arr.Length; $j++) { # Convert the byte to base 2, then pad the left with 0's to make sure # each string is 8 digits long $Str = [System.Convert]::ToString($Arr[$j], 2).PadLeft(8, '0') # Drop off the most significant bit and append to the string builder $SB.Append($Str.Substring(1)) | Out-Null } # Convert our concatenated binary string to a UInt64 and tell the converter # we're coming from base 2 [System.UInt64]$Value = [System.Convert]::ToUInt64($SB.ToString(), 2) # This approach also works, but is more prone to problems in case the Multiplier # exceeds 8 bytes <# $Multiplier = 1 for ($j = $Arr.Length - 1; $j -gt 0; $j--) { # Move the bits from the left byte to right byte, # as we move farther down, we need to move more bytes $Arr[$j] = ($Arr[$j - 1] -shl (8 - $Multiplier)) -bor $Arr[$j] # Shift down the left byte $Arr[$j - 1] = $Arr[$j - 1] -shr (1 * $Multiplier++) } $Arr = Set-ByteArrayPadding -InputObject $Arr -Length 8 if ([System.BitConverter]::IsLittleEndian) { [System.Array]::Reverse($Arr) } [System.UInt64]$Value = [System.BitConverter]::ToUInt64($Arr, 0) #> $Data += ".$Value" } else { $Data += ".$([System.Convert]::ToUInt32($Bytes[$i]))" } } } Write-Output -InputObject $Data } } Function ConvertFrom-OIDString { <# .SYNOPSIS Creates a byte array from an OID string. .DESCRIPTION This cmdlet take an OID string in an X.Y.Z.W format and produces a byte array that is used to represent the OID. .PARAMETER OID The OID string to convert to bytes. .EXAMPLE $Bytes = ConvertFrom-OIDString -OID "1.2.840.113549.1.1.1" The contents of $Bytes is (in decimal): 42 134 72 134 247 13 1 1 1 In hex the contents are: 2A 86 48 86 F7 0D 01 01 01 This is the RSA Encryption OID. .INPUTS System.String .OUTPUTS System.Byte[] .NOTES AUTHOR: Michael Haken LAST UPDATE: 1/22/2018 #> [CmdletBinding()] [OutputType([System.String])] Param( [Parameter(Mandatory = $true, Position = 0, ValueFromPipeline = $true)] [ValidateNotNullOrEmpty()] [System.String]$OID ) Begin { } Process { # In case the OID components are seperated by spaces, replace those with # periods, then split the parts on the periods [System.String[]]$Parts = $OID.Trim().Replace(' ', '.').Split('.') if ($Parts.Length -gt 2) { # This will hold all of the resulting bytes [System.Byte[]]$Results = @() # Iterate each component of the OID string for ($i = 0; $i -lt $Parts.Length; $i++) { # The first two components require special handling, they # are composed from data in a single byte if ($i -eq 0) { [System.UInt32]$High = $Parts[$i] [System.UInt32]$Low = $Parts[$i + 1] [System.UInt32]$FirstByte = ($High * 40) + $Low [System.Byte[]]$Bytes = [System.BitConverter]::GetBytes($FirstByte) if ([System.BitConverter]::IsLittleEndian) { [System.Array]::Reverse($Bytes) } # The GetBytes function will add 0x00 bytes to the result so that it is # 4 bytes in length, but our result will only be 1 byte as part of the OID byte # array $Bytes = Remove-ByteArrayPadding -InputObject $Bytes $Results += $Bytes # Skip the second the part since we already used it $i += 1 } else { # Get the byte integer value [System.UInt32]$Value = $Parts[$i] # If the value is less than 128, its value is just the byte if ($Value -lt 128) { [System.Byte[]]$Bytes = [System.BitConverter]::GetBytes($Value) if ([System.BitConverter]::IsLittleEndian) { [System.Array]::Reverse($Bytes) } # The GetBytes function will add 0x00 bytes to the result so that it is # 4 bytes in length, but our result will only be 1 byte as part of the OID byte # array $Bytes = Remove-ByteArrayPadding -InputObject $Bytes $Results += $Bytes } else { # The value is greater than 128, which means it uses multiple bytes to # store the value. [System.String]$BitString = [System.Convert]::ToString($Value, 2) # Count how many individual bit characters we've added to a string $Counter = 0 # Store all the bytes making up this value here [System.Byte[]]$Bytes = @() # Used as a buffer to store 7 digit bit strings $Line = "" # Create the bytes from the bit string, starting at the end of the string for ($j = $BitString.Length - 1; $j -ge 0; $j += -1) { $Line = "$($BitString[$j])$Line" $Counter++ # Once index 6 is set (meaning we've filled 7 digits), the counter is incremented to 7 # and then we need to reset it to start a new byte, or if this is the last digit # make sure the line is 8 digits long if ($Counter -ge 7 -or $j -eq 0) { # Since the string is 7 digits or less long, add 0's to pad # to 8 digits $Line = $Line.PadLeft(8, '0') [System.Byte]$Byte = [System.Convert]::ToByte($Line, 2) $Bytes += $Byte $Counter = 0 $Line = "" } } # The += adds the items to the end, but we started at the back of the # bit string, so we need to reverse the items [System.Array]::Reverse($Bytes) # Make all of the bytes except the last have a 1 in the most significant bit for ($j = 0; $j -lt $Bytes.Length - 1; $j++) { $Bytes[$j] = $Bytes[$j] -bor 0x80 } $Results += $Bytes } } } Write-Output -InputObject $Results } else { Write-Error -Exception (New-Object -TypeName System.ArgumentException("The OID string was not correctly formatted, it should include at least 2 parts separated by a period.")) -ErrorAction Stop } } End { } } |