Drop PDF source conversion path

Author: awakecoding

AGENTS.md

@@ -60,5 +60,5 @@ When you add a new exported function, add its name to the `$expected` array in t
 ## Project-specific rules
 
 - Do not remove or rename exported functions without updating `AwakeCoding.OpenSpecs.psd1` and the exports test.
-- Conversion: DOCX is handled in-module via OpenXML; PDF uses external `docling` or `markitdown` when available (see `AwakeCoding.OpenSpecs/Private/Get-OpenSpecToolchain.ps1`). Output is textual (tables, ASCII), not image-based.
+- Conversion: DOCX is handled in-module via OpenXML. PDF is not used as a conversion source. Output is textual (tables, ASCII), not image-based.
 - For bulk or CI conversions, use `-Parallel -ThrottleLimit N` with `Convert-OpenSpecToMarkdown` or `Invoke-OpenSpecConversionPipeline` (PowerShell 7 only).

AwakeCoding.OpenSpecs/Private/ConvertFrom-OpenSpecPdf.ps1

@@ -1,7 +1,6 @@
 function Get-OpenSpecToolchain {
     [CmdletBinding()]
     param(
-        [switch]$RequirePdfConverter,
         [switch]$RequireDocxConverter
     )
 
@@ -22,10 +21,6 @@ function Get-OpenSpecToolchain {
         HasOpenXml = $null -ne $openXmlModule
     }
 
-    if ($RequirePdfConverter -and -not ($toolchain.HasDocling -or $toolchain.HasMarkItDown)) {
-        throw 'No PDF converter detected. Install docling or markitdown.'
-    }
-
     if ($RequireDocxConverter -and -not $toolchain.HasOpenXml) {
         throw 'No DOCX converter detected. Install the OpenXML PowerShell module.'
     }

AwakeCoding.OpenSpecs/Private/Get-OpenSpecToolchain.ps1

@@ -8,7 +8,7 @@ function Convert-OpenSpecToMarkdown {
 
         [string]$OutputPath = (Join-Path -Path (Get-Location) -ChildPath 'converted-specs'),
 
-        [ValidateSet('Auto', 'DOCX', 'PDF')]
+        [ValidateSet('Auto', 'DOCX')]
         [string]$SourceFormat = 'Auto',
 
         [switch]$Force,
@@ -84,6 +84,11 @@ function Convert-OpenSpecToMarkdown {
                 $SourceFormat
             }
 
+            if ($resolvedFormat -eq 'PDF') {
+                Write-Error "PDF source conversion is not supported. Use DOCX input for '$sourcePath'."
+                continue
+            }
+
             if ($resolvedFormat -eq 'Unknown') {
                 Write-Error "Unable to infer source format for '$sourcePath'."
                 continue
@@ -136,12 +141,6 @@ function Convert-OpenSpecToMarkdown {
                 $mediaDirectory = Join-Path -Path $specDirectory -ChildPath 'media'
                 $conversionStep = ConvertFrom-OpenSpecDocx -InputPath $sourcePath -OutputPath $rawMarkdownPath -Toolchain $toolchain -MediaOutputDirectory $mediaDirectory
             }
-            elseif ($resolvedFormat -eq 'PDF') {
-                $toolchain = Get-OpenSpecToolchain -RequirePdfConverter
-                $rawMarkdownPath = Join-Path -Path $artifactDirectory -ChildPath 'raw-pdf.md'
-                $conversionStep = ConvertFrom-OpenSpecPdf -InputPath $sourcePath -OutputPath $rawMarkdownPath -Toolchain $toolchain
-            }
-
             $rawMarkdown = Get-Content -LiteralPath $conversionStep.OutputPath -Raw
             $normalized = ConvertTo-OpenSpecTextLayout -Markdown $rawMarkdown
             $sourceLinkMetadata = if ($conversionStep.PSObject.Properties['LinkMetadata']) { $conversionStep.LinkMetadata } else { $null }

AwakeCoding.OpenSpecs/Public/Convert-OpenSpecToMarkdown.ps1

@@ -5,7 +5,7 @@ function Invoke-OpenSpecConversionPipeline {
 
         [string]$Query,
 
-        [ValidateSet('PDF', 'DOCX', 'Both')]
+        [ValidateSet('DOCX', 'Both')]
         [string]$Format = 'DOCX',
 
         [string]$DownloadPath = (Join-Path -Path (Get-Location) -ChildPath 'downloads-convert'),
@@ -32,6 +32,9 @@ function Invoke-OpenSpecConversionPipeline {
         Save-OpenSpecDocument -Query $Query -Format $Format -OutputPath $DownloadPath -Force:$Force
     }
 
-    $toConvert = $downloadResults | Where-Object { $_.Status -in 'Downloaded', 'Exists' }
+    $toConvert = $downloadResults | Where-Object { $_.Status -in 'Downloaded', 'Exists' -and $_.Format -eq 'DOCX' }
+    if (-not @($toConvert)) {
+        throw 'No DOCX files are available for conversion. PDF source conversion is not supported.'
+    }
     $toConvert | Convert-OpenSpecToMarkdown -OutputPath $OutputPath -Force:$Force -Parallel:$Parallel -ThrottleLimit $ThrottleLimit -RemoveDocumentIndex:$RemoveDocumentIndex
 }