@pdftron/pdfnet-node-samples 9.2.0 → 9.4.0

package/samples/LogicalStructureTest/LogicalStructureTest.js

@@ -1,251 +1,251 @@
-//---------------------------------------------------------------------------------------
-// Copyright (c) 2001-2021 by PDFTron Systems Inc. All Rights Reserved.
-// Consult legal.txt regarding legal and license information.
-//---------------------------------------------------------------------------------------
-
-//---------------------------------------------------------------------------------------
-// This sample explores the structure and content of a tagged PDF document and dumps
-// the structure information to the console window.
-//
-// In tagged PDF documents StructTree acts as a central repository for information
-// related to a PDF document's logical structure. The tree consists of StructElement-s
-// and ContentItem-s which are leaf nodes of the structure tree.
-//
-// The sample can be extended to access and extract the marked-content elements such
-// as text and images.
-//---------------------------------------------------------------------------------------
-
-
-const { PDFNet } = require('@pdftron/pdfnet-node');
-const PDFTronLicense = require('../LicenseKey/LicenseKey');
-
-((exports) => {
-
-  exports.runLogicalStructureTest = () => {
-
-    const printAndIndent = (printState, indent) => {
-      console.log(printState.str);
-
-      let indentStr = '';
-      for (let i = 0; i < indent; ++i) {
-        indentStr += '  ';
-      }
-    printState.str = indentStr;
-    };
-
-    // Used in code snippet 1.
-    const processStructElement = async(element, indent, printState) => {
-      if (!(await element.isValid())) {
-        return;
-      }
-
-
-      // Print out the type and title info, if any.
-      printAndIndent(printState, indent++);
-      printState.str += 'Type: ' + (await element.getType());
-      if (await element.hasTitle()) {
-        printState.str += '. Title: ' + (await element.getTitle());
-      }
-
-      const num = await element.getNumKids();
-      for (let i = 0; i < num; ++i) {
-        // Check is the kid is a leaf node (i.e. it is a ContentItem).
-        if (await element.isContentItem(i)) {
-          const cont = await element.getAsContentItem(i);
-          const type = await cont.getType();
-
-          const page = await cont.getPage();
-
-          printAndIndent(printState, indent);
-          printState.str += 'Content Item. Part of page #' + (await page.getIndex());
-
-          printAndIndent(printState, indent);
-          switch (type) {
-            case PDFNet.ContentItem.Type.e_MCID:
-            case PDFNet.ContentItem.Type.e_MCR:
-              printState.str += 'MCID: ' + (await cont.getMCID());
-              break;
-            case PDFNet.ContentItem.Type.e_OBJR:
-              {
-                printState.str += 'OBJR ';
-                const refObj = await cont.getRefObj();
-                if (refObj) {
-                  printState.str += '- Referenced Object#: ' + refObj.getObjNum();
-                }
-              }
-              break;
-            default:
-              break;
-          }
-        } else { // the kid is another StructElement node.
-          await processStructElement(await element.getAsStructElem(i), indent, printState);
-        }
-      }
-    };
-
-    // Used in code snippet 2.
-    const processElements = async(reader, printState) => {
-      let element;
-      while (element = await reader.next()) { // Read page contents
-        // In this sample we process only paths & text, but the code can be
-        // extended to handle any element type.
-        const type = await element.getType();
-        if (type === PDFNet.Element.Type.e_path || type === PDFNet.Element.Type.e_text || type === PDFNet.Element.Type.e_path) {
-          switch (type) {
-            case PDFNet.Element.Type.e_path: // Process path ...
-              printState.str += '\nPATH: ';
-              break;
-            case PDFNet.Element.Type.e_text: // Process text ...
-              printState.str += '\nTEXT: ' + (await element.getTextString()) + '\n';
-              break;
-            case PDFNet.Element.Type.e_form: // Process form XObjects
-              printState.str += '\nFORM XObject: ';
-              // reader.formBegin();
-              // await ProcessElements(reader);
-              // reader.end();
-              break;
-          }
-
-          // Check if the element is associated with any structural element.
-          // Content items are leaf nodes of the structure tree.
-          const structParent = await element.getParentStructElement();
-          if (await structParent.isValid()) {
-            // Print out the parent structural element's type, title, and object number.
-            printState.str += ' Type: ' + (await structParent.getType()) + ', MCID: ' + (await element.getStructMCID());
-            if (await structParent.hasTitle()) {
-              printState.str += '. Title: ' + (await structParent.getTitle());
-            }
-            printState.str += ', Obj#: ' + (await (await structParent.getSDFObj()).getObjNum());
-          }
-        }
-      }
-    };
-
-    // Used in code snippet 3.
-    const processElements2 = async(reader, mcidPageMap) => {
-      let element;
-      while (element = await reader.next()) { // Read page contents
-        // In this sample we process only text, but the code can be extended
-        // to handle paths, images, or any other Element type.
-        const mcid = await element.getStructMCID();
-        if (mcid >= 0 && (await element.getType()) === PDFNet.Element.Type.e_text) {
-          const val = await element.getTextString();
-          if (mcid in mcidPageMap) {
-            mcidPageMap[mcid] += val;
-          } else {
-            mcidPageMap[mcid] = val;
-          }
-        }
-      }
-    };
-
-    // Used in code snippet 3.
-    const processStructElement2 = async(element, mcidDocMap, indent, printState) => {
-      if (!(await element.isValid())) {
-        return;
-      }
-
-      // Print out the type and title info, if any.
-      printAndIndent(printState, indent);
-      printState.str += '<' + (await element.getType());
-      if (await element.hasTitle()) {
-        printState.str += ' title="' + (await element.getTitle()) + '"';
-      }
-      printState.str += '>';
-
-      const num = await element.getNumKids();
-      for (let i = 0; i < num; ++i) {
-        if (await element.isContentItem(i)) {
-          const cont = await element.getAsContentItem(i);
-          if ((await cont.getType()) === PDFNet.ContentItem.Type.e_MCID) {
-            const pageNum = await (await cont.getPage()).getIndex();
-            const mcidPageMap = mcidDocMap[pageNum];
-            if (mcidPageMap) {
-              const mcid = await cont.getMCID();
-              if (mcid in mcidPageMap) {
-                printState.str += mcidPageMap[mcid];
-              }
-            }
-          }
-        } else { // the kid is another StructElement node.
-          await processStructElement2(await element.getAsStructElem(i), mcidDocMap, indent + 1, printState);
-        }
-      }
-
-      printAndIndent(printState, indent);
-      printState.str += '</' + (await element.getType()) + '>';
-    };
-
-    const main = async() => {
-      // Relative path to the folder containing test files.
-      const inputPath = '../TestFiles/';
-      const printState = { str: '' };
-      try { // Extract logical structure from a PDF document
-        const doc = await PDFNet.PDFDoc.createFromFilePath(inputPath + 'tagged.pdf');
-        doc.initSecurityHandler();
-
-        let reader = null;
-        let tree = null;
-
-        console.log('____________________________________________________________');
-        console.log('Sample 1 - Traverse logical structure tree...');
-        tree = await doc.getStructTree();
-        if (await tree.isValid()) {
-          console.log('Document has a StructTree root.');
-          for (let i = 0, numKids = await tree.getNumKids(); i < numKids; ++i) {
-            // Recursively get structure info for all child elements.
-            await processStructElement(await tree.getKid(i), 0, printState);
-          }
-        } else {
-          console.log('This document does not contain any logical structure.');
-        }
-        printAndIndent(printState, 0);
-        console.log('Done 1.');
-
-        console.log('____________________________________________________________');
-        console.log('Sample 2 - Get parent logical structure elements from');
-        console.log('layout elements.');
-        reader = await PDFNet.ElementReader.create();
-        for (let itr = await doc.getPageIterator(); await itr.hasNext(); itr.next()) {
-          reader.beginOnPage(await itr.current());
-          await processElements(reader, printState);
-          reader.end();
-        }
-        printAndIndent(printState, 0);
-        console.log('Done 2.');
-
-        console.log('____________________________________________________________');
-        console.log("Sample 3 - 'XML style' extraction of PDF logical structure and page content.");
-        {
-          const mcidDocMap = {};
-          for (let itr = await doc.getPageIterator(); await itr.hasNext(); itr.next()) {
-            const page = await itr.current();
-            reader.beginOnPage(page);
-            const pageNum = await page.getIndex();
-            const pageMcidMap = {};
-            mcidDocMap[pageNum] = pageMcidMap;
-            await processElements2(reader, pageMcidMap);
-            reader.end();
-          }
-
-          tree = await doc.getStructTree();
-          if (await tree.isValid()) {
-            for (let i = 0, numKids = await tree.getNumKids(); i < numKids; ++i) {
-              await processStructElement2(await tree.getKid(i), mcidDocMap, 0, printState);
-            }
-          }
-        }
-        printAndIndent(printState, 0);
-        console.log('Done 3.');
-        await doc.save(inputPath + 'Output/LogicalStructure.pdf', 0);
-      } catch (err) {
-        console.log(err);
-      }
-    };
-
-    PDFNet.runWithCleanup(main, PDFTronLicense.Key).catch(function(error){console.log('Error: ' + JSON.stringify(error));}).then(function(){return PDFNet.shutdown();});
-  };
-  exports.runLogicalStructureTest();
-})(exports);
-// eslint-disable-next-line spaced-comment
+//---------------------------------------------------------------------------------------
+// Copyright (c) 2001-2022 by PDFTron Systems Inc. All Rights Reserved.
+// Consult legal.txt regarding legal and license information.
+//---------------------------------------------------------------------------------------
+
+//---------------------------------------------------------------------------------------
+// This sample explores the structure and content of a tagged PDF document and dumps
+// the structure information to the console window.
+//
+// In tagged PDF documents StructTree acts as a central repository for information
+// related to a PDF document's logical structure. The tree consists of StructElement-s
+// and ContentItem-s which are leaf nodes of the structure tree.
+//
+// The sample can be extended to access and extract the marked-content elements such
+// as text and images.
+//---------------------------------------------------------------------------------------
+
+
+const { PDFNet } = require('@pdftron/pdfnet-node');
+const PDFTronLicense = require('../LicenseKey/LicenseKey');
+
+((exports) => {
+
+  exports.runLogicalStructureTest = () => {
+
+    const printAndIndent = (printState, indent) => {
+      console.log(printState.str);
+
+      let indentStr = '';
+      for (let i = 0; i < indent; ++i) {
+        indentStr += '  ';
+      }
+    printState.str = indentStr;
+    };
+
+    // Used in code snippet 1.
+    const processStructElement = async(element, indent, printState) => {
+      if (!(await element.isValid())) {
+        return;
+      }
+
+
+      // Print out the type and title info, if any.
+      printAndIndent(printState, indent++);
+      printState.str += 'Type: ' + (await element.getType());
+      if (await element.hasTitle()) {
+        printState.str += '. Title: ' + (await element.getTitle());
+      }
+
+      const num = await element.getNumKids();
+      for (let i = 0; i < num; ++i) {
+        // Check is the kid is a leaf node (i.e. it is a ContentItem).
+        if (await element.isContentItem(i)) {
+          const cont = await element.getAsContentItem(i);
+          const type = await cont.getType();
+
+          const page = await cont.getPage();
+
+          printAndIndent(printState, indent);
+          printState.str += 'Content Item. Part of page #' + (await page.getIndex());
+
+          printAndIndent(printState, indent);
+          switch (type) {
+            case PDFNet.ContentItem.Type.e_MCID:
+            case PDFNet.ContentItem.Type.e_MCR:
+              printState.str += 'MCID: ' + (await cont.getMCID());
+              break;
+            case PDFNet.ContentItem.Type.e_OBJR:
+              {
+                printState.str += 'OBJR ';
+                const refObj = await cont.getRefObj();
+                if (refObj) {
+                  printState.str += '- Referenced Object#: ' + refObj.getObjNum();
+                }
+              }
+              break;
+            default:
+              break;
+          }
+        } else { // the kid is another StructElement node.
+          await processStructElement(await element.getAsStructElem(i), indent, printState);
+        }
+      }
+    };
+
+    // Used in code snippet 2.
+    const processElements = async(reader, printState) => {
+      let element;
+      while (element = await reader.next()) { // Read page contents
+        // In this sample we process only paths & text, but the code can be
+        // extended to handle any element type.
+        const type = await element.getType();
+        if (type === PDFNet.Element.Type.e_path || type === PDFNet.Element.Type.e_text || type === PDFNet.Element.Type.e_path) {
+          switch (type) {
+            case PDFNet.Element.Type.e_path: // Process path ...
+              printState.str += '\nPATH: ';
+              break;
+            case PDFNet.Element.Type.e_text: // Process text ...
+              printState.str += '\nTEXT: ' + (await element.getTextString()) + '\n';
+              break;
+            case PDFNet.Element.Type.e_form: // Process form XObjects
+              printState.str += '\nFORM XObject: ';
+              // reader.formBegin();
+              // await ProcessElements(reader);
+              // reader.end();
+              break;
+          }
+
+          // Check if the element is associated with any structural element.
+          // Content items are leaf nodes of the structure tree.
+          const structParent = await element.getParentStructElement();
+          if (await structParent.isValid()) {
+            // Print out the parent structural element's type, title, and object number.
+            printState.str += ' Type: ' + (await structParent.getType()) + ', MCID: ' + (await element.getStructMCID());
+            if (await structParent.hasTitle()) {
+              printState.str += '. Title: ' + (await structParent.getTitle());
+            }
+            printState.str += ', Obj#: ' + (await (await structParent.getSDFObj()).getObjNum());
+          }
+        }
+      }
+    };
+
+    // Used in code snippet 3.
+    const processElements2 = async(reader, mcidPageMap) => {
+      let element;
+      while (element = await reader.next()) { // Read page contents
+        // In this sample we process only text, but the code can be extended
+        // to handle paths, images, or any other Element type.
+        const mcid = await element.getStructMCID();
+        if (mcid >= 0 && (await element.getType()) === PDFNet.Element.Type.e_text) {
+          const val = await element.getTextString();
+          if (mcid in mcidPageMap) {
+            mcidPageMap[mcid] += val;
+          } else {
+            mcidPageMap[mcid] = val;
+          }
+        }
+      }
+    };
+
+    // Used in code snippet 3.
+    const processStructElement2 = async(element, mcidDocMap, indent, printState) => {
+      if (!(await element.isValid())) {
+        return;
+      }
+
+      // Print out the type and title info, if any.
+      printAndIndent(printState, indent);
+      printState.str += '<' + (await element.getType());
+      if (await element.hasTitle()) {
+        printState.str += ' title="' + (await element.getTitle()) + '"';
+      }
+      printState.str += '>';
+
+      const num = await element.getNumKids();
+      for (let i = 0; i < num; ++i) {
+        if (await element.isContentItem(i)) {
+          const cont = await element.getAsContentItem(i);
+          if ((await cont.getType()) === PDFNet.ContentItem.Type.e_MCID) {
+            const pageNum = await (await cont.getPage()).getIndex();
+            const mcidPageMap = mcidDocMap[pageNum];
+            if (mcidPageMap) {
+              const mcid = await cont.getMCID();
+              if (mcid in mcidPageMap) {
+                printState.str += mcidPageMap[mcid];
+              }
+            }
+          }
+        } else { // the kid is another StructElement node.
+          await processStructElement2(await element.getAsStructElem(i), mcidDocMap, indent + 1, printState);
+        }
+      }
+
+      printAndIndent(printState, indent);
+      printState.str += '</' + (await element.getType()) + '>';
+    };
+
+    const main = async() => {
+      // Relative path to the folder containing test files.
+      const inputPath = '../TestFiles/';
+      const printState = { str: '' };
+      try { // Extract logical structure from a PDF document
+        const doc = await PDFNet.PDFDoc.createFromFilePath(inputPath + 'tagged.pdf');
+        doc.initSecurityHandler();
+
+        let reader = null;
+        let tree = null;
+
+        console.log('____________________________________________________________');
+        console.log('Sample 1 - Traverse logical structure tree...');
+        tree = await doc.getStructTree();
+        if (await tree.isValid()) {
+          console.log('Document has a StructTree root.');
+          for (let i = 0, numKids = await tree.getNumKids(); i < numKids; ++i) {
+            // Recursively get structure info for all child elements.
+            await processStructElement(await tree.getKid(i), 0, printState);
+          }
+        } else {
+          console.log('This document does not contain any logical structure.');
+        }
+        printAndIndent(printState, 0);
+        console.log('Done 1.');
+
+        console.log('____________________________________________________________');
+        console.log('Sample 2 - Get parent logical structure elements from');
+        console.log('layout elements.');
+        reader = await PDFNet.ElementReader.create();
+        for (let itr = await doc.getPageIterator(); await itr.hasNext(); itr.next()) {
+          reader.beginOnPage(await itr.current());
+          await processElements(reader, printState);
+          reader.end();
+        }
+        printAndIndent(printState, 0);
+        console.log('Done 2.');
+
+        console.log('____________________________________________________________');
+        console.log("Sample 3 - 'XML style' extraction of PDF logical structure and page content.");
+        {
+          const mcidDocMap = {};
+          for (let itr = await doc.getPageIterator(); await itr.hasNext(); itr.next()) {
+            const page = await itr.current();
+            reader.beginOnPage(page);
+            const pageNum = await page.getIndex();
+            const pageMcidMap = {};
+            mcidDocMap[pageNum] = pageMcidMap;
+            await processElements2(reader, pageMcidMap);
+            reader.end();
+          }
+
+          tree = await doc.getStructTree();
+          if (await tree.isValid()) {
+            for (let i = 0, numKids = await tree.getNumKids(); i < numKids; ++i) {
+              await processStructElement2(await tree.getKid(i), mcidDocMap, 0, printState);
+            }
+          }
+        }
+        printAndIndent(printState, 0);
+        console.log('Done 3.');
+        await doc.save(inputPath + 'Output/LogicalStructure.pdf', 0);
+      } catch (err) {
+        console.log(err);
+      }
+    };
+
+    PDFNet.runWithCleanup(main, PDFTronLicense.Key).catch(function(error){console.log('Error: ' + JSON.stringify(error));}).then(function(){return PDFNet.shutdown();});
+  };
+  exports.runLogicalStructureTest();
+})(exports);
+// eslint-disable-next-line spaced-comment
 //# sourceURL=LogicalStructureTest.js

package/samples/OCRTest/OCRTest.js

@@ -1,5 +1,5 @@
 //---------------------------------------------------------------------------------------
-// Copyright (c) 2001-2021 by PDFTron Systems Inc. All Rights Reserved.
+// Copyright (c) 2001-2022 by PDFTron Systems Inc. All Rights Reserved.
 // Consult legal.txt regarding legal and license information.
 //---------------------------------------------------------------------------------------
 
@@ -18,6 +18,8 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
       try {
 
         PDFNet.addResourceSearchPath('../../lib/');
+
+        const useIRIS = await PDFNet.OCRModule.isIRISModuleAvailable();
         if (!(await PDFNet.OCRModule.isModuleAvailable())) {
           console.log('\nUnable to run OCRTest: PDFTron SDK OCR module not available.');
           console.log('---------------------------------------------------------------');
@@ -39,10 +41,14 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
 
           // A) Setup empty destination doc
           const doc = await PDFNet.PDFDoc.create();
+          
           await doc.initSecurityHandler();
 
+          const opts = new PDFNet.OCRModule.OCROptions();
+          if(useIRIS) opts.setOCREngine('iris');
+
           // B) Run OCR on the .png with options
-          await PDFNet.OCRModule.imageToPDF(doc, input_path + 'psychomachia_excerpt.png');
+          await PDFNet.OCRModule.imageToPDF(doc, input_path + 'psychomachia_excerpt.png', opts);
 
           // C) check the result
           await doc.save(output_path + 'psychomachia_excerpt.pdf', 0);
@@ -62,6 +68,7 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
 
           // B) Setup options with multiple target languages, English will always be considered as secondary language
           const opts = new PDFNet.OCRModule.OCROptions();
+          if(useIRIS) opts.setOCREngine('iris');
           opts.addLang('rus');
           opts.addLang('deu');
 
@@ -85,6 +92,7 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
 
           // B) Setup options with a single language and an ignore zone
           const opts = new PDFNet.OCRModule.OCROptions();
+          if(useIRIS) opts.setOCREngine('iris');
           opts.addLang('deu');
 
           const ignore_zones = [];
@@ -111,6 +119,7 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
 
           // B) Setup options with a single language plus text/ignore zones
           const opts = new PDFNet.OCRModule.OCROptions();
+          if(useIRIS) opts.setOCREngine('iris');
           opts.addLang('eng');
 
           var ignore_zones = [];
@@ -163,8 +172,11 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
           const doc = await PDFNet.PDFDoc.createFromFilePath(input_path + 'zero_value_test_no_text.pdf');
           await doc.initSecurityHandler();
 
+          const opts = new PDFNet.OCRModule.OCROptions();
+          if(useIRIS) opts.setOCREngine('iris');
+
           // B) Run OCR on the .pdf with default English language
-          const json = await PDFNet.OCRModule.getOCRJsonFromPDF(doc);
+          const json = await PDFNet.OCRModule.getOCRJsonFromPDF(doc, opts);
 
           // C) Post-processing step (whatever it might be)
           console.log('Have OCR result JSON, re-applying to PDF ');
@@ -188,9 +200,12 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
           const doc = await PDFNet.PDFDoc.create();
           await doc.initSecurityHandler();
 
+          const opts = new PDFNet.OCRModule.OCROptions();
+          if(useIRIS) opts.setOCREngine('iris');
+
           // B) Run OCR on the .tif with default English language, extracting OCR results in XML format. Note that
           // in the process we convert the source image into PDF. We reuse this PDF document later to add hidden text layer to it.
-          const xml = await PDFNet.OCRModule.getOCRXmlFromImage(doc, input_path + 'physics.tif');
+          const xml = await PDFNet.OCRModule.getOCRXmlFromImage(doc, input_path + 'physics.tif', opts);
 
           // C) Post-processing step (whatever it might be)
           console.log('Have OCR result XML, re-applying to PDF');
@@ -205,36 +220,6 @@ const PDFTronLicense = require('../LicenseKey/LicenseKey');
         } catch (err) {
           console.log(err);
         }
-
-
-        //--------------------------------------------------------------------------------
-        // Example 7) Resolution can be manually set, when DPI missing from metadata or is wrong
-        try {
-          // A) Setup empty destination doc
-          const doc = await PDFNet.PDFDoc.create();
-          await doc.initSecurityHandler();
-
-          // B) Setup options with a text zone
-          const opts = new PDFNet.OCRModule.OCROptions();
-
-          const text_zones = [];
-          text_zones.push(new PDFNet.Rect(140, 870, 310, 920));
-          opts.addTextZonesForPage(text_zones, 1);
-
-          // C) Manually override DPI
-          opts.addDPI(100)
-
-          // D) Run OCR on the .jpg with options
-          await PDFNet.OCRModule.imageToPDF(doc, input_path + 'corrupted_dpi.jpg', opts);
-
-          // E) check the result
-          await doc.save(output_path + 'corrupted_dpi.pdf', 0);
-
-          console.log('Example 7: converting image with corrupted resolution metadata corrupted_dpi.jpg to pdf with searchable text');
-        } catch (err) {
-          console.log(err);
-        }
-
         console.log('Done.');
       } catch (err) {
         console.log(err);