Last active
March 29, 2023 13:00
-
-
Save mavaddat/a82cc0328a6b6d80dfbb927e0ac57bf2 to your computer and use it in GitHub Desktop.
Close Elements SGML to XML
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
function Close-Elements | |
{ | |
[CmdletBinding()] | |
param( | |
[Parameter(Mandatory, ParameterSetName="ReaderGiven", ValueFromPipeline, Position = 0)] | |
[ValidateNotNullOrEmpty()] | |
[System.Xml.XmlReader] $Reader, | |
[Parameter(Mandatory, ParameterSetName="InPathGiven", ValueFromPipeline, Position = 0)] | |
[ValidateNotNullOrEmpty()] | |
[string] $InPath, | |
[Parameter(Mandatory, ParameterSetName="WriterGiven", Position = 1)] | |
[Parameter(ParameterSetName="InPathGiven", Position = 1)] | |
[ValidateNotNullOrEmpty()] | |
[System.Xml.XmlWriter] $Writer, | |
[Parameter(Mandatory, ParameterSetName="OutPathGiven", Position = 1)] | |
[Parameter(ParameterSetName="InPathGiven", Position = 1)] | |
[ValidateNotNullOrEmpty()] | |
[string] $OutPath | |
) | |
<# Thanks to Scott Hanselman http://www.hanselman.com/blog/postprocessing-autoclosed-sgml-tags-with-the-sgmlreader #> | |
# $msgBody = $Reader.NameTable.Add('MSGBODY') | |
if($PSCmdlet.ParameterSetName -eq "OutPathGiven") | |
{ | |
$WriterSettings = [System.Xml.XmlWriterSettings]::new() | |
$WriterSettings.Indent = $true | |
$WriterSettings.IndentChars = ' '*4 | |
$WriterSettings.ConformanceLevel = [System.Xml.ConformanceLevel]::Fragment | |
$Writer = [System.Xml.XmlWriter]::Create($OutPath, $WriterSettings) | |
} | |
if($PSCmdlet.ParameterSetName -eq "InPathGiven"){ | |
$Reader = [Sgml.SgmlReader]::Create($InPath) | |
} | |
$selfClosingTags = 'recbomba', 'recinser', 'trbomba', 'trinser', 'spanspec', 'colspec', 'mtoc', 'ptitle', 'pnote', 'usercomm', 'xmit', 'ctoc', 'sbeff', 'unitid', 'mtoc', 'mlep', 'clep', 'itemcol', 'mfmatr', 'sbtitle', 'sbdata', 'sbinfo', 'sblist' | |
foreach ($tag in $selfClosingTags) { | |
$Reader.NameTable.Add($tag) | |
} | |
$previousElement = $null | |
$elementsWeAlreadyEnded = New-Object System.Collections.Stack | |
while ($Reader.Read()) | |
{ | |
switch ($Reader.NodeType) | |
{ | |
XmlNodeType.Element | |
{ | |
$previousElement = $Reader.LocalName | |
$Writer.WriteStartElement($Reader.LocalName) | |
break | |
} | |
XmlNodeType.Text | |
{ | |
if (-not [string]::IsNullOrEmpty($Reader.Value)) | |
{ | |
$Writer.WriteString($Reader.Value.Trim()) | |
if ($null -ne $previousElement -and $previousElement -notin $selfClosingTags) | |
{ | |
$Writer.WriteEndElement() | |
$elementsWeAlreadyEnded.Push($previousElement) | |
} | |
} | |
else | |
{ | |
Write-Debug -Message "Empty text node" | |
} | |
break | |
} | |
XmlNodeType.EndElement | |
{ | |
if ($elementsWeAlreadyEnded.Count -gt 0 -and $elementsWeAlreadyEnded.Peek() -eq $Reader.LocalName) | |
{ | |
# Next in the stack is the current end element, so pop it off | |
$elementsWeAlreadyEnded.Pop() | Out-Null | |
} | |
else | |
{ | |
$Writer.WriteEndElement() | |
} | |
break | |
} | |
default | |
{ | |
$Writer.WriteNode($Reader, $false) | |
break | |
} | |
} | |
} | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment