mavjav-edu · pull · May 22, 2024 · May 22, 2024 · May 22, 2024
diff --git a/PSModule.build.ps1 b/PSModule.build.ps1
@@ -105,33 +105,33 @@ Enter-Build {
         Get-Variable | select-object name, value, visibility | format-table -autosize | out-string | write-verbose
     }
 
-    #Register Nuget
-    if (!(get-packageprovider "Nuget" -ForceBootstrap -ErrorAction silentlycontinue)) {
-        write-verbose "Nuget Provider Not found. Fetching..."
-        Install-PackageProvider Nuget -forcebootstrap -scope currentuser @PassThruParams | out-string | write-verbose
-        write-verboseheader "Installed Nuget Provider Info"
-        Get-PackageProvider Nuget @PassThruParams | format-list | out-string | write-verbose
-    }
-
-    #Fix a bug with the Appveyor 2017 image having a broken nuget (points to v3 URL but installed packagemanagement doesn't query v3 correctly)
-    #Next command will add this back
-    if ($ENV:APPVEYOR -and ($ENV:APPVEYOR_BUILD_WORKER_IMAGE -eq 'Visual Studio 2017')) {
-        write-verbose "Detected Appveyor VS2017 Image, using v2 Nuget API"
-        UnRegister-PackageSource -Name nuget.org
-    }
-
-    #Add the nuget repository so we can download things like GitVersion
-    if (!(Get-PackageSource "nuget.org" -erroraction silentlycontinue)) {
-        write-verbose "Registering nuget.org as package source"
-        Register-PackageSource -provider NuGet -name nuget.org -location http://www.nuget.org/api/v2 -Trusted @PassThruParams  | out-string | write-verbose
-    }
-    else {
-        $nugetOrgPackageSource = Set-PackageSource -name 'nuget.org' -Trusted @PassThruParams
-        if ($PassThruParams.Verbose) {
-            write-verboseheader "Nuget.Org Package Source Info "
-            $nugetOrgPackageSource | format-table | out-string | write-verbose
-        }
-    }
+    # #Register Nuget
+    # if (!(get-packageprovider "Nuget" -ForceBootstrap -ErrorAction silentlycontinue)) {
+    #     write-verbose "Nuget Provider Not found. Fetching..."
+    #     Install-PackageProvider Nuget -forcebootstrap -scope currentuser @PassThruParams | out-string | write-verbose
+    #     write-verboseheader "Installed Nuget Provider Info"
+    #     Get-PackageProvider Nuget @PassThruParams | format-list | out-string | write-verbose
+    # }
+
+    # #Fix a bug with the Appveyor 2017 image having a broken nuget (points to v3 URL but installed packagemanagement doesn't query v3 correctly)
+    # #Next command will add this back
+    # if ($ENV:APPVEYOR -and ($ENV:APPVEYOR_BUILD_WORKER_IMAGE -eq 'Visual Studio 2017')) {
+    #     write-verbose "Detected Appveyor VS2017 Image, using v2 Nuget API"
+    #     UnRegister-PackageSource -Name nuget.org
+    # }
+
+    # #Add the nuget repository so we can download things like GitVersion
+    # if (!(Get-PackageSource "nuget.org" -erroraction silentlycontinue)) {
+    #     write-verbose "Registering nuget.org as package source"
+    #     Register-PackageSource -provider NuGet -name nuget.org -location http://www.nuget.org/api/v2 -Trusted @PassThruParams  | out-string | write-verbose
+    # }
+    # else {
+    #     $nugetOrgPackageSource = Set-PackageSource -name 'nuget.org' -Trusted @PassThruParams
+    #     if ($PassThruParams.Verbose) {
+    #         write-verboseheader "Nuget.Org Package Source Info "
+    #         $nugetOrgPackageSource | format-table | out-string | write-verbose
+    #     }
+    # }
 
     #Move to the Project Directory if we aren't there already
     Set-Location $buildRoot
@@ -476,7 +476,7 @@ task PublishPSGallery -if (-not $SkipPublish) {
 ### SuperTasks
 # These are the only supported items to run directly from Invoke-Build
 task Deploy PreDeploymentChecks,Package,PublishGitHubRelease,PublishPSGallery
-task Build Clean,CopyFilesToBuildDir,UpdateMetadata
+task Build Clean, CopyFilesToBuildDir
 task Test Pester
 
 #Default Task - Build, Test with Pester, Deploy

diff --git a/PowerHTML.psd1 b/PowerHTML.psd1
@@ -12,7 +12,7 @@
 RootModule = 'PowerHTML.psm1'
 
 # Version number of this module.
-ModuleVersion = '0.0.1'
+ModuleVersion = '0.2.0'
 
 # Supported PSEditions
 # CompatiblePSEditions = @()
@@ -27,7 +27,7 @@ Author = 'Justin Grote'
 CompanyName = 'Unspecified'
 
 # Copyright statement for this module
-Copyright = '(c) 2018 Justin Grote. All rights reserved.'
+Copyright = '(c) 2024 Justin Grote. All rights reserved.'
 
 # Description of the functionality provided by this module
 Description = 'Provides a wrapper for HTML Agility Pack for use where the IE HTML DOM from Invoke-WebRequest is not available such as Powershell Core'
@@ -63,7 +63,7 @@ Description = 'Provides a wrapper for HTML Agility Pack for use where the IE HTM
 # TypesToProcess = @()
 
 # Format files (.ps1xml) to be loaded when importing this module
-FormatsToProcess = @('.\Types\*.ps1xml')
+FormatsToProcess = @('.\Types\HtmlAgilityPack.HtmlTextNode.ps1xml')
 
 # Modules to import as nested modules of the module specified in RootModule/ModuleToProcess
 # NestedModules = @()
@@ -95,13 +95,13 @@ PrivateData = @{
     PSData = @{
 
         # Tags applied to this module. These help with module discovery in online galleries.
-        # Tags = @()
+        Tags = @('HTML','Invoke-WebRequest','Scraping','Agility','Pack','ScreenScraping')
 
         # A URL to the license for this module.
-        # LicenseUri = ''
+        LicenseUri = 'https://github.com/JustinGrote/PowerHTML/blob/master/LICENSE'
 
         # A URL to the main website for this project.
-        # ProjectUri = ''
+        ProjectUri = 'https://github.com/JustinGrote/PowerHTML'
 
         # A URL to an icon representing this module.
         # IconUri = ''
@@ -120,5 +120,3 @@ PrivateData = @{
 # DefaultCommandPrefix = ''
 
 }
-
-
diff --git a/PowerHTML.psm1 b/PowerHTML.psm1
@@ -1,9 +1,9 @@
 #Get public and private function definition files.
-$PublicFunctions  = @( Get-ChildItem -Path $PSScriptRoot\Public\*.ps1 -ErrorAction SilentlyContinue )
-$PrivateFunctions = @( Get-ChildItem -Path $PSScriptRoot\Private\*.ps1 -ErrorAction SilentlyContinue )
+$PublicFunctions = @( Get-ChildItem -Path $PSScriptRoot\Public\*.ps1 -ErrorAction Ignore )
+$PrivateFunctions = @( Get-ChildItem -Path $PSScriptRoot\Private\*.ps1 -ErrorAction Ignore )
 
 #Get JSON settings files
-$ModuleSettings = @( Get-ChildItem -Path $PSScriptRoot\Settings\*.json -ErrorAction SilentlyContinue )
+$ModuleSettings = @( Get-ChildItem -Path $PSScriptRoot\Settings\*.json -ErrorAction Ignore )
 
 #Determine which assembly versions to load
 #See if .Net Standard 2.0 is available on the system and if not, load the legacy Net 4.0 library
@@ -40,23 +40,15 @@ if ($AssembliesToLoad) {
 }
 
 #Dot source the files
-Foreach($FunctionToImport in @($PublicFunctions + $PrivateFunctions))
-{
-    Try
-    {
+foreach ($FunctionToImport in @($PublicFunctions + $PrivateFunctions)) {
+    try {
         . $FunctionToImport.fullname
-    }
-    Catch
-    {
+    } catch {
         Write-Error -Message "Failed to import function $($import.fullname): $_"
     }
 }
 
 #Import Settings files as global objects based on their filename
-foreach ($ModuleSettingsItem in $ModuleSettings)
-{
+foreach ($ModuleSettingsItem in $ModuleSettings) {
     New-Variable -Name "$($ModuleSettingsItem.basename)" -Scope Global -Value (convertfrom-json (Get-Content -raw $ModuleSettingsItem.fullname)) -Force
-}
-
-#Export the public functions. This requires them to match the standard Noun-Verb powershell cmdlet format as a safety mechanism
-Export-ModuleMember -Function ($PublicFunctions.Basename | where {$PSitem -match '^\w+-\w+$'})
+}
diff --git a/Public/ConvertFrom-HTML.ps1 b/Public/ConvertFrom-HTML.ps1
@@ -1,122 +1,109 @@
-<#
-.SYNOPSIS
-    Takes an HTML input and converts it to an HTMLAgilityPack htmlNode object that can be navigated using Linq
-.DESCRIPTION
-    Long description
-.EXAMPLE
-    PS C:\> $HTMLString = @"
-    <!DOCTYPE html>
-    <html>
-    <body>
-    <h1>My First Heading</h1>
-    <p>My first paragraph.</p>d
-    </body>
-    </html>
-    "@
- PS C:\> $HTMLString | ConvertFrom-HTML -OutVariable result
-
-NodeType Name      AttributeCount ChildNodeCount ContentLength InnerText
--------- ----      -------------- -------------- ------------- ---------
-Document #document 0              4              103               …
-
- PS C:\> $result.SelectSingleNode("//body/h1")
-
-NodeType Name AttributeCount ChildNodeCount ContentLength InnerText
--------- ---- -------------- -------------- ------------- ---------
-Element  h1   0              1              16            My First Heading
-
-    Convert HTML string to a HtmlNode via the pipeline.
-
-.EXAMPLE
-    PS C:\> $uri = "https://www.powershellgallery.com/"
-    PS C:\> $result = ConvertFrom-HTML -uri $uri
-    PS C:\> $result
-
-NodeType Name      AttributeCount ChildNodeCount ContentLength InnerText
--------- ----      -------------- -------------- ------------- ---------
-Document #document 0              4              17550         …
-
-    Fetch and parse $uri directly via the URI pipeline.
-.EXAMPLE
-    PS C:\>  Get-Item $testFilePath | ConvertFrom-Html
-
-NodeType Name      AttributeCount ChildNodeCount ContentLength InnerText
--------- ----      -------------- -------------- ------------- ---------
-Document #document 0              5              105               …
-
-    Parse an HTML file piped from Get-Item.
-.INPUTS
-    [String[]]
-    [System.IO.FileInfo[]]
-.OUTPUTS
-    [HtmlAgilityPack.HtmlDocument]
-    [HtmlAgilityPack.HtmlNode]
-.NOTES
-    General notes
-#>
+
 function ConvertFrom-Html {
-    [CmdletBinding(DefaultParameterSetName="String")]
-    param (
+    <#
+    .SYNOPSIS
+        Takes an HTML input and converts it to an HTMLAgilityPack htmlNode object that can be navigated using Linq
+    .DESCRIPTION
+        Long description
+    .EXAMPLE
+        $HTMLString = @'
+        <!DOCTYPE html>
+        <html>
+        <body>
+        <h1>My First Heading</h1>
+        <p>My first paragraph.</p>d
+        </body>
+        </html>
+'@ | ConvertFrom-HTML
+
+        $HTMLString
+
+    NodeType Name      AttributeCount ChildNodeCount ContentLength InnerText
+    -------- ----      -------------- -------------- ------------- ---------
+    Document #document 0              4              103               …
+
+        $HTMLString.SelectSingleNode('//body/h1')
+
+    NodeType Name AttributeCount ChildNodeCount ContentLength InnerText
+    -------- ---- -------------- -------------- ------------- ---------
+    Element  h1   0              1              16            My First Heading
+
+        Convert HTML string to a HtmlNode via the pipeline.
+
+    .EXAMPLE
+        $uri = [Uri]'https://www.powershellgallery.com/' | ConvertFrom-HTML
+        $uri
+
+    NodeType Name      AttributeCount ChildNodeCount ContentLength InnerText
+    -------- ----      -------------- -------------- ------------- ---------
+    Document #document 0              4              17550         …
+
+        Fetch and parse a url.
+    .EXAMPLE
+        Get-Item $testFilePath | ConvertFrom-Html
+
+    NodeType Name      AttributeCount ChildNodeCount ContentLength InnerText
+    -------- ----      -------------- -------------- ------------- ---------
+    Document #document 0              5              105               …
+
+        Parse an HTML file piped from Get-Item.
+    .INPUTS
+        [String[]]
+        [System.IO.FileInfo[]]
+        [System.URI[]]
+    .OUTPUTS
+        [HtmlAgilityPack.HtmlDocument]
+        [HtmlAgilityPack.HtmlNode]
+    .NOTES
+        General notes
+    #>
+    [OutputType([HtmlAgilityPack.HtmlNode])]
+    [OutputType([HtmlAgilityPack.HtmlDocument])]
+    [CmdletBinding(DefaultParameterSetName = 'String')]
+    param(
         #The HTML text to parse. Accepts multiple separate documents as an array. This also accepts pipeline from Invoke-WebRequest
-        [Parameter(ParameterSetName="String",Mandatory,ValueFromPipeline,ValueFromPipelineByPropertyName,Position=0)]
-        [String[]]$Content,
+        [Parameter(ParameterSetName = 'String', Mandatory, ValueFromPipeline, ValueFromPipelineByPropertyName, Position = 0)]
+        [String[]] $Content,
 
         #The URI or URIs from which to retrieve content. This may be faster than using Invoke-WebRequest but is less flexible in the method of retrieval (for instance, no POST)
-        [Parameter(ParameterSetName="URI",Mandatory,ValueFromPipeline,ValueFromPipelineByPropertyName)]
-        [System.URI[]]$URI,
+        [Parameter(ParameterSetName = 'URI', Mandatory, ValueFromPipeline, ValueFromPipelineByPropertyName, Position = 0)]
+        [System.URI[]] $URI,
 
         #Path to file or files containing HTML content to convert. This accepts pipeline from Get-Childitem or Get-Item
-        [Parameter(ParameterSetName="Path",Mandatory,ValueFromPipeline,ValueFromPipelineByPropertyName)]
-        [System.IO.FileInfo[]]$Path,
+        [Parameter(ParameterSetName = 'Path', Mandatory, ValueFromPipeline, ValueFromPipelineByPropertyName, Position = 0)]
+        [System.IO.FileInfo[]] $Path,
 
         #Do not return the Linq documentnode, instead return the HTMLDocument object. This is useful if you want to do XPath queries instead of Linq queries
-        [switch]$Raw
-
+        [switch] $Raw
     )
-
     begin {
+        $html = [HtmlAgilityPack.HtmlDocument]::new()
+        $web = [HtmlAgilityPack.HtmlWeb]::new()
     }
-
     process {
-        #Find the type of input and bind it to inputObject
-        $inputObject = $null
-        foreach ($contentType in "Content","URI","Path") {
-            if ((Get-Variable -erroraction SilentlyContinue $contentType).value) {
-                $inputObject = (Get-Variable $contentType).value
-                break
-            }
-        }
-        if (-not $inputObject) {write-error "Input Object Type Not Identified. If you see this then ConvertFrom-HTML needs better input validation"}
-
-        #Unwrap any arrays. This allows us to accept both pipeline and parameter input
-        $inputObject | ForEach-Object {
-            $inputItem = $PSItem
-            $htmlDoc = new-object HtmlAgilityPack.HtmlDocument
-
-            #Process all object types into a common HTML document format
-            switch ($inputItem.GetType().FullName) {
-                "System.String" {
-                    $htmlDoc.LoadHtml($inputItem)
+        switch ($PSCmdlet.ParameterSetName) {
+            'String' {
+                $Content | ForEach-Object {
+                    Write-Verbose "Loading HTML"
+                    $html.LoadHtml($_)
+                    if ($Raw) { $html } else { $html.DocumentNode }
                 }
-                "System.Uri" {
-                    $htmlDoc = (new-object HtmlAgilityPack.HtmlWeb).Load($inputItem)
-                }
-                "System.IO.FileInfo" {
-                    $htmlDoc.Load($inputItem)
-                }
-                Default {
-                    write-error "Object Type not supported or implemented. If you see this error then ConvertFrom-HTML has improper input validation"
-                    continue
+            }
+            'URI' {
+                $URI | ForEach-Object {
+                    Write-Verbose "Loading URI $_"
+                    $site = $web.Load($_)
+                    if ($Raw) { $site } else { $site.DocumentNode }
                 }
             }
-            if ($inputItem) {
-                if ($Raw) {
-                    $htmlDoc
-                } else {
-                    $htmlDoc.DocumentNode
+            'Path' {
+                $Path | ForEach-Object {
+                    Write-Verbose "Loading File $_"
+                    $html.Load($_.FullName)
+                    if ($Raw) { $html } else { $html.DocumentNode }
                 }
             }
         }
-
     }
-}
+
+}