How to extract text from pdf using golang?

I am trying to extract text from a pdf file in golang. See the code below. For some reason, it's printing complete garbage(some random numbers). Here is the pdf. I believe it's possible to extract text since I am able to copy and paste the text from this file.

package main
import (
"bufio"
"bytes"
"fmt"
"io/ioutil"
"log"
"net/http"
"os"
"strings"
pdf "github.com/unidoc/unipdf/v3/model"
)
func main() {
fmt.Println("Enter URL of PDF file:")
reader := bufio.NewReader(os.Stdin)
url, err := reader.ReadString('\n')
if err != nil {
log.Fatal(err)
}
url = strings.TrimSpace(url)
// Fetch PDF from URL.
resp, err := http.Get(url)
if err != nil {
log.Fatal(err)
}
defer resp.Body.Close()
buf, _ := ioutil.ReadAll(resp.Body)
pdfReader, err := pdf.NewPdfReader(bytes.NewReader(buf))
if err != nil {
log.Fatal(err)
}
// Parse PDF file.
isEncrypted, err := pdfReader.IsEncrypted()
if err != nil {
log.Fatal(err)
}
// If PDF is encrypted, exit with message.
if isEncrypted {
fmt.Println("Error: PDF is encrypted.")
os.Exit(1)
}
// Get number of pages.
numPages, err := pdfReader.GetNumPages()
if err != nil {
log.Fatal(err)
}
// Iterate through pages and print text.
for i := 1; i <= numPages; i++ {
page, err := pdfReader.GetPage(i)
if err != nil {
log.Fatal(err)
}
text, err := page.GetAllContentStreams()
if err != nil {
log.Fatal(err)
}
fmt.Println(text)
}
}

It is possible for GetAllContentStreams might returns formats, graphics, images, and other objects in that page and that might be the reason for printing complete garbage(some random numbers).

> GetAllContentStreams gets all the content streams for a page as one
> string

Instead of GetAllContentStreams, we can use ExtractText method to extract the text.

> ExtractText processes and extracts all text data in content streams
> and returns as a string.

And this should need a licence api key to use the package.

https://github.com/unidoc/unipdf

> This software package (unipdf) is a commercial product and requires a
> license code to operate.
>
> To Get a Metered License API Key in for free in the Free Tier, sign up
> on https://cloud.unidoc.io

The unipdf example code can be found at here

Here is the updated code

func init() {
// Make sure to load your metered License API key prior to using the library.
// If you need a key, you can sign up and create a free one at https://cloud.unidoc.io
err := license.SetMeteredKey("your-metered-api-key")
if err != nil {
panic(err)
}
}
func main() {
//
// The other blocks in your code
//
// Iterate through pages and print text.
for i := 1; i <= numPages; i++ {
pageNum := i + 1
page, err := pdfReader.GetPage(i)
if err != nil {
log.Fatal(err)
}
ex, err := extractor.New(page)
if err != nil {
log.Fatal(err)
}
text, err := ex.ExtractText()
if err != nil {
log.Fatal(err)
}
fmt.Println("------------------------------")
fmt.Printf("Page %d:\n", pageNum)
fmt.Printf(text)
fmt.Println("------------------------------")
}
}

I can not find a free, capable Go package to extract text from PDF. Luckily, there are some free CLI tools that can do this.

pdftotext from Xpdf is a promising choice. See its output:

$ pdftotext -layout -nopgbrk 2023-04-24_BU-12.pdf - | head
                           ALL INDIA TENNIS ASSOCIATION
                                        As on 24TH April , 2023
       BOY'S UNDER-12                                 2011                BEST    BEST    25% BEST POINTS
       24TH April , 2023                                                  Eight   Eight     Eight  CUT FOR     TTL.
                                                                          SING.   DBLS.     DBLS. NO SHOW      PTS.
RANK   NAME OF PLAYER                     REG NO.      DOB       STATE     PTS.   PTS.       PTS.  LATE WL    Final
  1    VIVAAN MIRDHA                      432735    08-Apr-11      (RJ)    485     565     141.25     0        797
  2    SMIT SACHIN UNDRE                  437763    07-Feb-11    (MH)      435     480       120      0      664.25
  3    RISHIKESH MANE                     436806    15-Jan-11    (MH)      420     380        95      0        619
  4    VIRAJ CHOUDHARY                    436648    03-Feb-11      (DL)    415     420       105      0      598.75

On Ubuntu, this tool can be installed with this command:

$ sudo apt install poppler-utils

And it's easy to execute it from Go application with the exec package:

package main

import (
	"bytes"
	"context"
	"fmt"
	"os/exec"
)

func main() {
	// See "man pdftotext" for more options.
	args := []string{
		"-layout",              // Maintain (as best as possible) the original physical layout of the text.
		"-nopgbrk",             // Don't insert page breaks (form feed characters) between pages.
		"2023-04-24_BU-12.pdf", // The input file.
		"-",                    // Send the output to stdout.
	}
	cmd := exec.CommandContext(context.Background(), "pdftotext", args...)

	var buf bytes.Buffer
	cmd.Stdout = &buf

	if err := cmd.Run(); err != nil {
		fmt.Println(err)
		return
	}

	fmt.Println(buf.String())
}