Section:

Extract Text, Read, Parse PDF - TextExtract - Node.js (JavaScript) Sample Code

Sample code for using Apryse SDK to read a PDF (parse and extract text), provided in Python, C++, C#, Java, Node.js (JavaScript), PHP, Ruby, Go and VB. If you'd like to search text on PDF pages, see our code sample for text search.

Learn more about our full PDF Data Extraction SDK Capabilities.

To start your free trial, get stated with Server SDK.

1//---------------------------------------------------------------------------------------
2// Copyright (c) 2001-2024 by Apryse Software Inc. All Rights Reserved.
3// Consult legal.txt regarding legal and license information.
4//---------------------------------------------------------------------------------------
5
6
7const { PDFNet } = require('@pdftron/pdfnet-node');
8const PDFTronLicense = require('../LicenseKey/LicenseKey');
9
10((exports) => {
11
12  exports.runTextExtractTest = async () => {
13    // A utility method used to dump all text content in the console window.
14    const dumpAllText = async (reader) => {
15      let element;
16      let bbox;
17      let arr;
18      while ((element = await reader.next()) !== null) {
19        switch (await element.getType()) {
20          case PDFNet.Element.Type.e_text_begin:
21            console.log('\n--> Text Block Begin');
22            break;
23          case PDFNet.Element.Type.e_text_end:
24            console.log('\n--> Text Block End');
25            break;
26          case PDFNet.Element.Type.e_text:
27            bbox = await element.getBBox();
28            console.log('\n--> BBox: ' + bbox.x1.toFixed(2) + ', ' + bbox.y1.toFixed(2) + ', ' + bbox.x2.toFixed(2) + ', ' + bbox.y2.toFixed(2) + '\n');
29            arr = await element.getTextString();
30            console.log(arr);
31            break;
32          case PDFNet.Element.Type.e_text_new_line:
33            console.log('\n--> New Line');
34            break;
35          case PDFNet.Element.Type.e_form:
36            reader.formBegin();
37            await dumpAllText(reader);
38            reader.end();
39            break;
40        }
41      }
42    };
43
44    // helper method for ReadTextFromRect
45    const rectTextSearch = async (reader, pos, srchStr) => {
46      let element;
47      let arr;
48      while ((element = await reader.next()) !== null) {
49        let bbox;
50        switch (await element.getType()) {
51          case PDFNet.Element.Type.e_text:
52            bbox = await element.getBBox();
53            if (await bbox.intersectRect(bbox, pos)) {
54              arr = await element.getTextString();
55              srchStr += arr + '\n';
56            }
57            break;
58          case PDFNet.Element.Type.e_text_new_line:
59            break;
60          case PDFNet.Element.Type.e_form:
61            reader.formBegin();
62            srchStr += await rectTextSearch(reader, pos, srchStr); // possibly need srchStr = ...
63            reader.end();
64            break;
65        }
66      }
67      return srchStr;
68    };
69
70    const readTextFromRect = async (page, pos, reader) => {
71      let srchStr = '';
72      reader.beginOnPage(page); // uses default parameters.
73      srchStr += await rectTextSearch(reader, pos, srchStr);
74      reader.end();
75      return srchStr;
76    };
77
78    const twoDigitHex = function (num) {
79      const hexStr = num.toString(16).toUpperCase();
80      return ('0' + hexStr).substr(-2);
81    }
82
83    const printStyle = async (s) => {
84      const rgb = await s.getColor();
85      const rColorVal = await rgb.get(0);
86      const gColorVal = await rgb.get(1);
87      const bColorVal = await rgb.get(2);
88      const rgbHex = twoDigitHex(rColorVal) + twoDigitHex(gColorVal) + twoDigitHex(bColorVal)
89      const fontName = await s.getFontName();
90      const fontSize = await s.getFontSize();
91      const serifOutput = ((await s.isSerif()) ? ' sans-serif; ' : ' ');
92      const returnString = ' style="font-family:' + fontName + '; font-size:' + fontSize + ';' + serifOutput + 'color:#' + rgbHex + ';"';
93      return returnString;
94    };
95
96    const main = async () => {
97      // eslint-disable-next-line no-unused-vars
98      let ret = 0;
99
100      // Relative path to the folder containing test files.
101      const inputPath = '../TestFiles/';
102      const inputFilename = 'newsletter.pdf'; // addimage.pdf, newsletter.pdf
103
104      const example1Basic = false;
105      const example2XML = false;
106      const example3Wordlist = false;
107      const example4Advanced = true;
108      const example5LowLevel = false;
109
110      try {
111        await PDFNet.startDeallocateStack();
112        const doc = await PDFNet.PDFDoc.createFromFilePath(inputPath + inputFilename);
113        doc.initSecurityHandler();
114
115        const page = await doc.getPage(1);
116
117        if (page.id === '0') {
118          console.log('Page not found.');
119          return 1;
120        }
121
122        const txt = await PDFNet.TextExtractor.create();
123        txt.begin(page);
124
125        let text;
126        let line;
127        let word;
128
129        // Example 1. Get all text on the page in a single string.
130        // Words will be separated with space or new line characters.
131        if (example1Basic) {
132          const wordCount = await txt.getWordCount();
133          console.log('Word Count: ' + wordCount);
134          text = await txt.getAsText();
135          console.log('\n\n- GetAsText --------------------------');
136          console.log(text);
137          console.log('-----------------------------------------------------------');
138        }
139
140        // Example 2. Get XML logical structure for the page.
141        if (example2XML) {
142          text = await txt.getAsXML(PDFNet.TextExtractor.XMLOutputFlags.e_words_as_elements | PDFNet.TextExtractor.XMLOutputFlags.e_output_bbox | PDFNet.TextExtractor.XMLOutputFlags.e_output_style_info);
143          console.log('\n\n- GetAsXML  --------------------------\n' + text);
144          console.log('-----------------------------------------------------------');
145        }
146
147        // Example 3. Extract words one by one.
148        if (example3Wordlist) {
149          line = await txt.getFirstLine();
150          for (; (await line.isValid()); line = (await line.getNextLine())) {
151            for (word = await line.getFirstWord(); await word.isValid(); word = await word.getNextWord()) {
152              text = await word.getString();
153              console.log(text);
154            }
155          }
156          console.log('-----------------------------------------------------------');
157        }
158
159        // Example 4. A more advanced text extraction example. 
160        // The output is XML structure containing paragraphs, lines, words, 
161        // as well as style and positioning information.
162        if (example4Advanced) {
163          let b;
164          let q;
165          let curFlowID = -1;
166          let curParaID = -1;
167
168          console.log('<PDFText>');
169
170          // For each line on the page...
171          for (line = await txt.getFirstLine(); await line.isValid(); line = await line.getNextLine()) {
172            if ((await line.getNumWords()) === 0) {
173              continue;
174            }
175            if (curFlowID !== await line.getFlowID()) {
176              if (curFlowID !== -1) {
177                if (curParaID !== -1) {
178                  curParaID = -1;
179                  console.log('</Para>');
180                }
181                console.log('</Flow>');
182              }
183              curFlowID = await line.getFlowID();
184              console.log('<Flow id="' + curFlowID + '">');
185            }
186            if (curParaID !== await line.getParagraphID()) {
187              if (curParaID !== -1) {
188                console.log('</Para>');
189              }
190              curParaID = await line.getParagraphID();
191              console.log('<Para id="' + curParaID + '">');
192            }
193            b = await line.getBBox();
194            const lineStyle = await line.getStyle();
195            let outputStringLineBox = '<Line box="' + b.x1.toFixed(2) + ', ' + b.y1.toFixed(2) + ', ' + b.x2.toFixed(2) + ', ' + b.y2.toFixed(2) + '"';
196            outputStringLineBox += (await printStyle(lineStyle));
197            const currentLineNum = await line.getCurrentNum();
198            outputStringLineBox += ' cur_num="' + currentLineNum + '">';
199            console.log(outputStringLineBox);
200
201            // For each word in the line...
202            for (word = await line.getFirstWord(); await word.isValid(); word = await word.getNextWord()) {
203              // output bounding box for the word
204              q = await word.getBBox();
205              const currentNum = await word.getCurrentNum();
206              let outputStringWord = '<Word box="' + q.x1.toFixed(2) + ', ' + q.y1.toFixed(2) + ', ' + q.x2.toFixed(2) + ', ' + q.y2.toFixed(2) + '" cur_num="' + currentNum + '"';
207              const sz = await word.getStringLen();
208              if (sz === 0) {
209                continue;
210              }
211              // if the word style is different from the parent style, output the new style
212              const sty = await word.getStyle();
213              if (!(await sty.compare(lineStyle))) {
214                outputStringWord += await printStyle(sty);
215              }
216              outputStringWord += '>' + (await word.getString()) + '</Word>';
217              console.log(outputStringWord);
218            }
219            console.log('</Line>');
220          }
221          if (curFlowID !== -1) {
222            if (curParaID !== -1) {
223              curParaID = -1;
224              console.log('</Para>');
225            }
226            console.log('</Flow>');
227          }
228          console.log('</PDFText>');
229        }
230        await PDFNet.endDeallocateStack();
231      } catch (err) {
232        console.log(err);
233        console.log(err.stack);
234        ret = 1;
235      }
236
237
238      if (example5LowLevel) {
239        ret = 0;
240        try {
241          await PDFNet.startDeallocateStack();
242          const doc = await PDFNet.PDFDoc.createFromFilePath(inputPath + inputFilename);
243          doc.initSecurityHandler();
244
245          // Example 1. Extract all text content from the document
246          const reader = await PDFNet.ElementReader.create();
247          const itr = await doc.getPageIterator(1);
248
249          //  Read every page
250          for (itr; await itr.hasNext(); itr.next()) {
251            const page = await itr.current();
252            reader.beginOnPage(page);
253            await dumpAllText(reader);
254            reader.end();
255          }
256          // Example 2. Extract text content based on the
257          // selection rectangle.
258          console.log('\n----------------------------------------------------');
259          console.log('Extract text based on the selection rectangle.');
260          console.log('----------------------------------------------------');
261
262
263          const firstPage = await (await doc.getPageIterator()).current();
264          let s1 = await readTextFromRect(firstPage, (await PDFNet.Rect.init(27, 392, 563, 534)), reader);
265          console.log('\nField 1: ' + s1);
266
267          s1 = await readTextFromRect(firstPage, (await PDFNet.Rect.init(28, 551, 106, 623)), reader);
268          console.log('Field 2: ' + s1);
269
270          s1 = await readTextFromRect(firstPage, (await PDFNet.Rect.init(208, 550, 387, 621)), reader);
271          console.log('Field 3: ' + s1);
272
273          // ...
274          console.log('Done');
275          await PDFNet.endDeallocateStack();
276        } catch (err) {
277          console.log(err.stack);
278          ret = 1;
279        }
280      }
281    };
282    PDFNet.runWithCleanup(main, PDFTronLicense.Key).catch(function (error) { console.log('Error: ' + JSON.stringify(error)); }).then(function () { return PDFNet.shutdown(); });
283  };
284  exports.runTextExtractTest();
285})(exports);
286// eslint-disable-next-line spaced-comment
287//# sourceURL=TextExtractTest.js

Did you find this helpful?

Trial setup questions?

Ask experts on Discord

Need other help?

Contact Support

Pricing or product questions?

Contact Sales