Enhance PDF generation steps to support multiple reference engines

- Modified `step_generate_reference_pdfs` function to accept an `engine` parameter, allowing selection between LibreOffice and MS Office for PDF conversion.
- Updated command execution logic to use the appropriate script based on the selected engine.
- Adjusted argument parsing in the main function to include an `--engine` option for specifying the reference engine.
- Updated documentation and print statements to reflect the changes in the reference engine selection process.

Author: shps951023

.vscode/settings.json

@@ -0,0 +1,4 @@
+{
+    "github.copilot.chat.backgroundCompaction": true,
+    "github.copilot.chat.summarizeAgentConversationHistoryThreshold": 170000
+}

scripts/Run-Benchmark.ps1

@@ -20,6 +20,8 @@ param(
     [switch]$SkipInstall,
     [switch]$WithOffice,
     [switch]$SkipOffice,
+    [ValidateSet("libre", "office")]
+    [string]$Engine = "office",
     [string]$Filter
 )
 
@@ -50,6 +52,7 @@ if ($SkipMiniPdf) { $pyArgs += "--skip-minipdf" }
 if ($SkipReference) { $pyArgs += "--skip-reference" }
 if ($WithOffice) { $pyArgs += "--with-office" }
 if ($SkipOffice) { $pyArgs += "--skip-office" }
+if ($Engine -ne "office") { $pyArgs += "--engine"; $pyArgs += $Engine }
 if ($Filter) { $pyArgs += "--filter"; $pyArgs += $Filter }
 
 # Run the benchmark pipeline

scripts/Run-Benchmark_docx.ps1

@@ -20,6 +20,8 @@ param(
     [switch]$SkipInstall,
     [switch]$WithOffice,
     [switch]$SkipOffice,
+    [ValidateSet("libre", "office")]
+    [string]$Engine = "office",
     [string]$Filter
 )
 
@@ -50,6 +52,7 @@ if ($SkipMiniPdf) { $pyArgs += "--skip-minipdf" }
 if ($SkipReference) { $pyArgs += "--skip-reference" }
 if ($WithOffice) { $pyArgs += "--with-office" }
 if ($SkipOffice) { $pyArgs += "--skip-office" }
+if ($Engine -ne "office") { $pyArgs += "--engine"; $pyArgs += $Engine }
 if ($Filter) { $pyArgs += "--filter"; $pyArgs += $Filter }
 
 # Run the benchmark pipeline

scripts/Run-Benchmark_issues.ps1

@@ -22,7 +22,9 @@ param(
     [switch]$SkipReference,
     [switch]$SkipInstall,
     [switch]$WithOffice,
-    [switch]$SkipOffice
+    [switch]$SkipOffice,
+    [ValidateSet("libre", "office")]
+    [string]$Engine = "office"
 )
 
 $ErrorActionPreference = "Continue"
@@ -93,14 +95,26 @@ if ($xlsxFiles -and $xlsxFiles.Count -gt 0) {
     }
 
     if (-not $CompareOnly -and -not $SkipReference) {
-        Write-Host '[Step 2] Converting XLSX -> PDF (LibreOffice)...' -ForegroundColor Yellow
-        Push-Location $BenchmarkDir
-        try {
-            $refArgs = @("generate_reference_pdfs.py", "--xlsx-dir", $XlsxIssueDir, "--pdf-dir", $RefXlsx)
-            if ($Filter) { $refArgs += @("--filter", $Filter) }
-            python @refArgs
-        } finally {
-            Pop-Location
+        if ($Engine -eq 'office') {
+            Write-Host '[Step 2] Converting XLSX -> PDF (Office / Excel COM)...' -ForegroundColor Yellow
+            Push-Location $BenchmarkDir
+            try {
+                $refArgs = @("generate_office_pdfs.py", "--xlsx-dir", $XlsxIssueDir, "--pdf-dir", $RefXlsx)
+                if ($Filter) { $refArgs += @("--filter", $Filter) }
+                python @refArgs
+            } finally {
+                Pop-Location
+            }
+        } else {
+            Write-Host '[Step 2] Converting XLSX -> PDF (LibreOffice)...' -ForegroundColor Yellow
+            Push-Location $BenchmarkDir
+            try {
+                $refArgs = @("generate_reference_pdfs.py", "--xlsx-dir", $XlsxIssueDir, "--pdf-dir", $RefXlsx)
+                if ($Filter) { $refArgs += @("--filter", $Filter) }
+                python @refArgs
+            } finally {
+                Pop-Location
+            }
         }
     }
 
@@ -153,14 +167,26 @@ if ($docxFiles -and $docxFiles.Count -gt 0) {
     }
 
     if (-not $CompareOnly -and -not $SkipReference) {
-        Write-Host '[Step 2] Converting DOCX -> PDF (LibreOffice)...' -ForegroundColor Yellow
-        Push-Location $BenchmarkDir
-        try {
-            $refArgs = @("generate_reference_pdfs_docx.py", "--docx-dir", $DocxIssueDir, "--pdf-dir", $RefDocx)
-            if ($Filter) { $refArgs += @("--filter", $Filter) }
-            python @refArgs
-        } finally {
-            Pop-Location
+        if ($Engine -eq 'office') {
+            Write-Host '[Step 2] Converting DOCX -> PDF (Office / Word COM)...' -ForegroundColor Yellow
+            Push-Location $BenchmarkDir
+            try {
+                $refArgs = @("generate_office_pdfs_docx.py", "--docx-dir", $DocxIssueDir, "--pdf-dir", $RefDocx)
+                if ($Filter) { $refArgs += @("--filter", $Filter) }
+                python @refArgs
+            } finally {
+                Pop-Location
+            }
+        } else {
+            Write-Host '[Step 2] Converting DOCX -> PDF (LibreOffice)...' -ForegroundColor Yellow
+            Push-Location $BenchmarkDir
+            try {
+                $refArgs = @("generate_reference_pdfs_docx.py", "--docx-dir", $DocxIssueDir, "--pdf-dir", $RefDocx)
+                if ($Filter) { $refArgs += @("--filter", $Filter) }
+                python @refArgs
+            } finally {
+                Pop-Location
+            }
         }
     }
 

tests/MiniPdf.Benchmark/generate_office_pdfs.py

@@ -52,6 +52,8 @@ def main():
                         help="Directory containing .xlsx files")
     parser.add_argument("--pdf-dir", default="office_pdfs",
                         help="Output directory for Office-generated PDFs")
+    parser.add_argument("--filter", default=None, metavar="PATTERN",
+                        help="Only convert files whose name contains this substring")
     args = parser.parse_args()
 
     xlsx_dir = os.path.abspath(args.xlsx_dir)
@@ -69,6 +71,8 @@ def main():
     print()
 
     xlsx_files = sorted(Path(xlsx_dir).glob("*.xlsx"))
+    if args.filter:
+        xlsx_files = [f for f in xlsx_files if args.filter.lower() in f.stem.lower()]
     if not xlsx_files:
         print("No .xlsx files found.")
         sys.exit(1)

tests/MiniPdf.Benchmark/generate_office_pdfs_docx.py

@@ -49,6 +49,8 @@ def main():
                         help="Directory containing .docx files")
     parser.add_argument("--pdf-dir", default="office_pdfs_docx",
                         help="Output directory for Office-generated PDFs")
+    parser.add_argument("--filter", default=None, metavar="PATTERN",
+                        help="Only convert files whose name contains this substring")
     args = parser.parse_args()
 
     docx_dir = os.path.abspath(args.docx_dir)
@@ -66,6 +68,8 @@ def main():
     print()
 
     docx_files = sorted(Path(docx_dir).glob("*.docx"))
+    if args.filter:
+        docx_files = [f for f in docx_files if args.filter.lower() in f.stem.lower()]
     if not docx_files:
         print("No .docx files found.")
         sys.exit(1)