Add OCRModule::PageNeedsOCR sample usage (Python, Ruby, PHP, Go)

Samples/OCRTest/GO/OCR_test.go

@@ -233,6 +233,65 @@ func TestOCR(t *testing.T) {
 		doc.Save(outputPath+"physics.pdf", uint(0))
 		fmt.Println("Example 6: extracting and applying OCR XML from physics.tif")
 
+		// Example 7) Check whether a page needs OCR before running it, to avoid unnecessary OCR
+		// processing on pages that already contain extractable text
+		// --------------------------------------------------------------------------------
+
+		// A) Positive test: lorem_ipsum.pdf is a scanned document with no extractable text, so it should need OCR.
+
+		scannedDoc := NewPDFDoc(inputPath + "../lorem_ipsum.pdf")
+
+		scannedPage := scannedDoc.GetPage(1)
+
+		// Ask OCRModule whether the page needs OCR. Passing accept_invisible_text as false
+		// means invisible text (e.g. from a prior OCR pass) is ignored and only visible content
+		// is considered, while true would treat an existing invisible text layer as sufficient
+		// to consider the page as not needing OCR.
+
+		scannedNeedsOCR := OCRModulePageNeedsOCR(scannedPage, false)
+		if scannedNeedsOCR {
+			fmt.Println("Example 7 (positive test): page 1 of lorem_ipsum.pdf needs OCR")
+		} else {
+			fmt.Println("Example 7 (positive test): page 1 of lorem_ipsum.pdf does not need OCR")
+		}
+
+		// Only run OCR if it is actually needed
+
+		if scannedNeedsOCR {
+			opts = NewOCROptions()
+			if use_iris {
+				opts.SetOCREngine("iris")
+			}
+			opts.AddLang("eng")
+			OCRModuleProcessPDF(scannedDoc, opts)
+			scannedDoc.Save(outputPath+"lorem_ipsum_conditional.pdf", uint(0))
+		}
+
+		// B) Negative test: newsletter.pdf already contains extractable text, so it should not need OCR.
+
+		newsletterDoc := NewPDFDoc(inputPath + "../newsletter.pdf")
+
+		newsletterPage := newsletterDoc.GetPage(1)
+
+		newsletterNeedsOCR := OCRModulePageNeedsOCR(newsletterPage, false)
+		if newsletterNeedsOCR {
+			fmt.Println("Example 7 (negative test): page 1 of newsletter.pdf needs OCR")
+		} else {
+			fmt.Println("Example 7 (negative test): page 1 of newsletter.pdf does not need OCR")
+		}
+
+		// Only run OCR if it is actually needed
+
+		if newsletterNeedsOCR {
+			opts = NewOCROptions()
+			if use_iris {
+				opts.SetOCREngine("iris")
+			}
+			opts.AddLang("eng")
+			OCRModuleProcessPDF(newsletterDoc, opts)
+			newsletterDoc.Save(outputPath+"newsletter_conditional.pdf", uint(0))
+		}
+
 		PDFNetTerminate()
 	}
 }

Samples/OCRTest/PHP/OCRTest.php

@@ -224,6 +224,54 @@
 
 		echo "Example 6: extracting and applying OCR XML from physics.tif \n";
 
+		//--------------------------------------------------------------------------------
+		// Example 7) Check whether a page needs OCR before running it, to avoid unnecessary OCR processing on pages that already contain extractable text
+
+		// A) Positive test: lorem_ipsum.pdf is a scanned document with no extractable text, so it should need OCR.
+
+		$scanned_doc = new PDFDoc($input_path."../lorem_ipsum.pdf");
+		$scanned_page = $scanned_doc->GetPage(1);
+
+		// Ask OCRModule whether the page needs OCR. Passing accept_invisible_text as false
+		// means invisible text (e.g. from a prior OCR pass) is ignored and only visible content
+		// is considered, while true would treat an existing invisible text layer as sufficient
+		// to consider the page as not needing OCR.
+
+		$scanned_needs_ocr = OCRModule::PageNeedsOCR($scanned_page, false);
+		echo "Example 7 (positive test): page 1 of lorem_ipsum.pdf ".($scanned_needs_ocr ? "needs" : "does not need")." OCR \n";
+
+		// Only run OCR if it is actually needed
+
+		if ($scanned_needs_ocr) {
+			$opts = new OCROptions();
+			if ($use_iris) {
+				$opts->SetOCREngine("iris");
+			}
+			$opts->AddLang("eng");
+			OCRModule::ProcessPDF($scanned_doc, $opts);
+			$scanned_doc->Save($output_path."lorem_ipsum_conditional.pdf", 0);
+		}
+
+		// B) Negative test: newsletter.pdf already contains extractable text, so it should not need OCR.
+
+		$newsletter_doc = new PDFDoc($input_path."../newsletter.pdf");
+		$newsletter_page = $newsletter_doc->GetPage(1);
+
+		$newsletter_needs_ocr = OCRModule::PageNeedsOCR($newsletter_page, false);
+		echo "Example 7 (negative test): page 1 of newsletter.pdf ".($newsletter_needs_ocr ? "needs" : "does not need")." OCR \n";
+
+		// Only run OCR if it is actually needed
+
+		if ($newsletter_needs_ocr) {
+			$opts = new OCROptions();
+			if ($use_iris) {
+				$opts->SetOCREngine("iris");
+			}
+			$opts->AddLang("eng");
+			OCRModule::ProcessPDF($newsletter_doc, $opts);
+			$newsletter_doc->Save($output_path."newsletter_conditional.pdf", 0);
+		}
+
 		echo "Done. \n";
 
 	}

Samples/OCRTest/PYTHON/OCRTest.py

@@ -222,6 +222,51 @@ def main():
         doc.Save(output_path + "physics.pdf", 0)
         print("Example 6: extracting and applying OCR XML from physics.tif")
 
+        # Example 7) Check whether a page needs OCR before running it, to avoid unnecessary OCR
+        # processing on pages that already contain extractable text
+        # --------------------------------------------------------------------------------
+
+        # A) Positive test: lorem_ipsum.pdf is a scanned document with no extractable text, so it should need OCR.
+
+        scanned_doc = PDFDoc(input_path + "../lorem_ipsum.pdf")
+        scanned_page = scanned_doc.GetPage(1)
+
+        # Ask OCRModule whether the page needs OCR. Passing accept_invisible_text as False
+        # means invisible text (e.g. from a prior OCR pass) is ignored and only visible content
+        # is considered, while True would treat an existing invisible text layer as sufficient
+        # to consider the page as not needing OCR.
+
+        scanned_needs_ocr = OCRModule.PageNeedsOCR(scanned_page, False)
+        print("Example 7 (positive test): page 1 of lorem_ipsum.pdf " + ("needs" if scanned_needs_ocr else "does not need") + " OCR")
+
+        # Only run OCR if it is actually needed
+
+        if scanned_needs_ocr:
+            opts = OCROptions()
+            if use_iris:
+                opts.SetOCREngine("iris")
+            opts.AddLang("eng")
+            OCRModule.ProcessPDF(scanned_doc, opts)
+            scanned_doc.Save(output_path + "lorem_ipsum_conditional.pdf", 0)
+
+        # B) Negative test: newsletter.pdf already contains extractable text, so it should not need OCR.
+
+        newsletter_doc = PDFDoc(input_path + "../newsletter.pdf")
+        newsletter_page = newsletter_doc.GetPage(1)
+
+        newsletter_needs_ocr = OCRModule.PageNeedsOCR(newsletter_page, False)
+        print("Example 7 (negative test): page 1 of newsletter.pdf " + ("needs" if newsletter_needs_ocr else "does not need") + " OCR")
+
+        # Only run OCR if it is actually needed
+
+        if newsletter_needs_ocr:
+            opts = OCROptions()
+            if use_iris:
+                opts.SetOCREngine("iris")
+            opts.AddLang("eng")
+            OCRModule.ProcessPDF(newsletter_doc, opts)
+            newsletter_doc.Save(output_path + "newsletter_conditional.pdf", 0)
+
         PDFNet.Terminate()
 
 

Samples/OCRTest/RUBY/OCRTest.rb

@@ -249,6 +249,54 @@
 
       doc.Close
 
+      # Example 7) Check whether a page needs OCR before running it, to avoid unnecessary OCR
+      # processing on pages that already contain extractable text
+      # --------------------------------------------------------------------------------
+
+      # A) Positive test: lorem_ipsum.pdf is a scanned document with no extractable text, so it should need OCR.
+      scanned_doc = PDFDoc.new(input_path + "../lorem_ipsum.pdf")
+      scanned_page = scanned_doc.GetPage(1)
+
+      # Ask OCRModule whether the page needs OCR. Passing accept_invisible_text as false
+      # means invisible text (e.g. from a prior OCR pass) is ignored and only visible content
+      # is considered, while true would treat an existing invisible text layer as sufficient
+      # to consider the page as not needing OCR.
+      scanned_needs_ocr = OCRModule.PageNeedsOCR(scanned_page, false)
+      puts "Example 7 (positive test): page 1 of lorem_ipsum.pdf #{scanned_needs_ocr ? 'needs' : 'does not need'} OCR"
+
+      # Only run OCR if it is actually needed
+      if scanned_needs_ocr
+         opts = OCROptions.new
+         if use_iris
+            opts.SetOCREngine("iris")
+         end
+         opts.AddLang("eng")
+         OCRModule.ProcessPDF(scanned_doc, opts)
+         scanned_doc.Save(output_path + "lorem_ipsum_conditional.pdf", 0)
+      end
+
+      scanned_doc.Close
+
+      # B) Negative test: newsletter.pdf already contains extractable text, so it should not need OCR.
+      newsletter_doc = PDFDoc.new(input_path + "../newsletter.pdf")
+      newsletter_page = newsletter_doc.GetPage(1)
+
+      newsletter_needs_ocr = OCRModule.PageNeedsOCR(newsletter_page, false)
+      puts "Example 7 (negative test): page 1 of newsletter.pdf #{newsletter_needs_ocr ? 'needs' : 'does not need'} OCR"
+
+      # Only run OCR if it is actually needed
+      if newsletter_needs_ocr
+         opts = OCROptions.new
+         if use_iris
+            opts.SetOCREngine("iris")
+         end
+         opts.AddLang("eng")
+         OCRModule.ProcessPDF(newsletter_doc, opts)
+         newsletter_doc.Save(output_path + "newsletter_conditional.pdf", 0)
+      end
+
+      newsletter_doc.Close
+
    end
    rescue Exception=>e
       puts e