Section:

Extract Text, Read, Parse PDF - TextExtract - Go 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) 2001-2021 by PDFTron Systems Inc. All Rights Reserved.
3// Consult LICENSE.txt regarding license information.
4//---------------------------------------------------------------------------------------
5
6package main
7import (
8	"fmt"
9	"strconv"
10	"os"
11	. "pdftron"
12)
13
14import  "pdftron/Samples/LicenseKey/GO"
15
16func PrintStyle (style Style){
17    sansSerifStr := ""
18    if style.IsSerif(){
19        sansSerifStr = " sans-serif;"
20	}
21    rgb := style.GetColor()
22    rgbHex := fmt.Sprintf("%02X%02X%02X;", rgb.Get(0), rgb.Get(1), rgb.Get(2))
23    fontStr := fmt.Sprintf("%g", style.GetFontSize())
24    os.Stdout.Write([]byte(" style=\"font-family:" + style.GetFontName() + "; font-size:" + fontStr + ";" + sansSerifStr + " color:#" + rgbHex + "\""))
25}
26
27func DumpAllText (reader ElementReader){
28    element := reader.Next()
29
30    for element.GetMp_elem().Swigcptr() != 0{
31        etype := element.GetType()
32        if etype == ElementE_text_begin{
33            fmt.Println("Text Block Begin")
34        }else if etype == ElementE_text_end{
35            fmt.Println("Text Block End")
36        }else if etype == ElementE_text{
37            bbox := element.GetBBox()
38            fmt.Println("BBox: " + fmt.Sprintf("%f", bbox.GetX1()) + ", " + fmt.Sprintf("%f", bbox.GetY1()) + ", " +
39						fmt.Sprintf("%f", bbox.GetX2()) + ", " + fmt.Sprintf("%f", bbox.GetY2()))
40            textString := element.GetTextString()
41            fmt.Println(textString)
42        }else if etype == ElementE_text_new_line{
43            fmt.Println("New Line")
44        }else if etype == ElementE_form{
45            reader.FormBegin()
46            DumpAllText(reader)
47            reader.End()
48		}
49        element = reader.Next()
50	}
51}
52
53// A utility method used to extract all text content from
54// a given selection rectangle. The recnagle coordinates are
55// expressed in PDF user/page coordinate system.
56func ReadTextFromRect (page Page, pos Rect, reader ElementReader) string{
57    reader.Begin(page)
58    srchStr := RectTextSearch(reader, pos)
59    reader.End()
60    return srchStr
61}
62//A helper method for ReadTextFromRect
63func RectTextSearch (reader ElementReader, pos Rect) string{
64    element := reader.Next()
65    srchStr2 := ""
66    for element.GetMp_elem().Swigcptr() != 0{
67        etype := element.GetType()
68        if etype == ElementE_text{
69            bbox := element.GetBBox()
70            if (bbox.IntersectRect(bbox, pos)){
71                arr := element.GetTextString()
72                srchStr2 += arr
73                srchStr2 += "\n"
74			}
75        }else if etype == ElementE_text_new_line{
76            //handle text new line here
77        }else if etype == ElementE_form{
78            reader.FormBegin()
79            srchStr2 += RectTextSearch(reader, pos)
80            fmt.Println(srchStr2)
81            reader.End()
82		}
83        element = reader.Next()
84	}
85    return srchStr2
86}
87
88func main(){
89    PDFNetInitialize(PDFTronLicense.Key)
90    
91    // Relative path to the folder containing test files.
92    inputPath :=  "../../TestFiles/newsletter.pdf"
93    example1Basic := false
94    example2Xml := false
95    example3Wordlist := false
96    example4Advanced := true
97    example5LowLevel := false
98   
99    // Sample code showing how to use high-level text extraction APIs.
100    doc := NewPDFDoc(inputPath)
101    doc.InitSecurityHandler()
102    
103    page := doc.GetPage(1)
104    if page == nil{
105        fmt.Println("page no found")
106    }    
107    txt := NewTextExtractor()
108    txt.Begin(page) // Read the page
109    
110    // Example 1. Get all text on the page in a single string.
111    // Words will be separated witht space or new line characters.
112    if example1Basic{
113        fmt.Println("Word count: " + strconv.Itoa(txt.GetWordCount()))
114        txtAsText := txt.GetAsText()
115        fmt.Println("- GetAsText --------------------------" + txtAsText)
116        fmt.Println("-----------------------------------------------------------")
117	}
118    // Example 2. Get XML logical structure for the page.
119    if example2Xml{
120        text := txt.GetAsXML(TextExtractorE_words_as_elements | 
121                            TextExtractorE_output_bbox | 
122                            TextExtractorE_output_style_info)       
123        fmt.Println("- GetAsXML  --------------------------" + text)
124        fmt.Println("-----------------------------------------------------------")
125    }
126    // Example 3. Extract words one by one.
127    if example3Wordlist{
128        word := NewWord()
129        line := txt.GetFirstLine()
130        for line.IsValid(){
131            word = line.GetFirstWord()
132            for word.IsValid(){
133                wordString := word.GetString()
134                fmt.Println(wordString)
135                word = word.GetNextWord()
136			}
137            line = line.GetNextLine()
138		}
139        fmt.Println("-----------------------------------------------------------")
140	}
141    // Example 4. A more advanced text extraction example. 
142    // The output is XML structure containing paragraphs, lines, words, 
143    // as well as style and positioning information.
144    if example4Advanced{
145        bbox := NewRect()
146        curFlowId := -1
147        curParaId := -1
148        
149        fmt.Println("<PDFText>")
150        // For each line on the page...
151        line := txt.GetFirstLine()
152        for line.IsValid(){
153            if line.GetNumWords() == 0{
154                line = line.GetNextLine()			
155                continue
156			}
157            word := line.GetFirstWord()
158            if curFlowId != line.GetFlowID(){
159                if curFlowId != -1{
160                    if curParaId != -1{
161                        curParaId = -1
162                        fmt.Println("</Para>")
163					}
164                    fmt.Println("</Flow>")
165				}
166                curFlowId = line.GetFlowID()
167                fmt.Println("<Flow id=\"" + strconv.Itoa(curFlowId) +"\">")
168            }        
169            if curParaId != line.GetParagraphID(){
170                if curParaId != -1{
171                    fmt.Println("</Para>")
172				}
173                curParaId= line.GetParagraphID()
174                fmt.Println("<Para id=\"" +strconv.Itoa(curParaId)+ "\">")
175            }    
176            bbox = line.GetBBox()
177            lineStyle := line.GetStyle()
178            os.Stdout.Write([]byte(fmt.Sprintf("<Line box=\"%.2f, %.2f, %.2f, %.2f\"", bbox.GetX1(), bbox.GetY1(), bbox.GetX2(), bbox.GetY2())))
179            PrintStyle (lineStyle)
180            os.Stdout.Write([]byte(" cur_num=\"" + strconv.Itoa(line.GetCurrentNum()) + "\"" + ">\n"))
181            
182            // For each word in the line...
183            word = line.GetFirstWord()
184            for word.IsValid(){
185                // Output the bounding box for the word
186                bbox = word.GetBBox()
187				os.Stdout.Write([]byte(fmt.Sprintf("<Word box=\"%.2f, %.2f, %.2f, %.2f\"", bbox.GetX1(), bbox.GetY1(), bbox.GetX2(), bbox.GetY2())))
188                os.Stdout.Write([]byte(" cur_num=\"" + strconv.Itoa(word.GetCurrentNum()) + "\""));
189                sz := word.GetStringLen()
190                if sz == 0{
191                    word = word.GetNextWord()				
192                    continue
193				}
194                // If the word style is different from the parent style, output the new style.
195                s := word.GetStyle()
196                if !s.IsEqual(lineStyle){
197                    PrintStyle (s)
198				}
199                wordString := word.GetString()
200                os.Stdout.Write([]byte(">" + wordString + "</Word>\n"))
201                word = word.GetNextWord()
202			}
203            os.Stdout.Write([]byte("</Line>\n"))               
204            line = line.GetNextLine()
205        }    
206        if curFlowId != -1{
207            if curParaId != -1{
208                curParaId = -1
209                os.Stdout.Write([]byte("</Para>\n"))
210			}
211            os.Stdout.Write([]byte("</Flow>\n"))
212        }
213        txt.Destroy()
214        doc.Close()            
215        fmt.Println("</PDFText>")
216    }
217    // Sample code showing how to use low-level text extraction APIs.
218    if example5LowLevel{
219        doc = NewPDFDoc(inputPath)
220        doc.InitSecurityHandler()
221
222        // Example 1. Extract all text content from the document
223        
224        reader := NewElementReader()
225        itr := doc.GetPageIterator()
226        for itr.HasNext(){
227            reader.Begin(itr.Current())
228            DumpAllText(reader)
229            reader.End()
230            itr.Next()
231        }
232		
233        // Example 2. Extract text content based on the 
234        // selection rectangle.
235        
236        fmt.Println("----------------------------------------------------")
237        fmt.Println("Extract text based on the selection rectangle.")
238        fmt.Println("----------------------------------------------------")
239        
240        itr = doc.GetPageIterator()
241        firstPage := itr.Current()
242        s1 := ReadTextFromRect(firstPage, NewRect(27.0, 392.0, 563.0, 534.0), reader)
243        fmt.Println("Field 1: " + s1)
244
245        s1 = ReadTextFromRect(firstPage, NewRect(28.0, 551.0, 106.0, 623.0), reader);
246        fmt.Println("Field 2: " + s1)
247
248        s1 = ReadTextFromRect(firstPage, NewRect(208.0, 550.0, 387.0, 621.0), reader);
249        fmt.Println("Field 3: " + s1)
250        
251        doc.Close()
252        PDFNetTerminate()
253        fmt.Println("Done.")
254	}
255}

Did you find this helpful?

Trial setup questions?

Ask experts on Discord

Need other help?

Contact Support

Pricing or product questions?

Contact Sales