problems with GetPageText()

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

Hello, 
Using Delphi XE2 and version 912 of Debenu Library VCL component.

Doing a GetPageText I get some odd decoding issues with some PDFs.

A snippet of my code to get the page text which is pretty straigt forward:

          for x := 1 to PDFLibrary.PageCount do begin
            PDFLibrary.SelectPage(x);
            Memo1.Text := Memo1.Text + PDFLibrary.GetPageText(7);//passing 7 preserves formatting
          end;

Here is how the text looks on the PDF. You'll have to trust me that it looks like this since I cannot post a screen shot.

Tax
        Labor Tax          @    7.00%        $4.20  
        Parts Tax           @    7.00%      $27.75  
  Tax Total                                       $31.95 

When I get the page text I get stuff like this:

Tax
                                             $4.20
           Labor Tax   @       7.00%        $27.75
                      @
            Parts Tax
                               7.00%
 Tax Total                                  $31.95

I'm guessing there is something wacky about how this PDF is created. Is there anything I can do to get my page text in a format more closely to how it shows on the actual PDF? I need the text to be properly formatted to parse it.

Thanks for any advice.

Regards,
TJ Asher

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
tj asher Members Profile Find Members Posts Beginner Joined: 09 Jan 13 Status: Offline Points: 2	Post Options Post Reply Quote tj asher Report Post Thanks(0) Quote Reply Topic: problems with GetPageText() Posted: 09 Jan 13 at 7:26PM
	Hello, Using Delphi XE2 and version 912 of Debenu Library VCL component. Doing a GetPageText I get some odd decoding issues with some PDFs. A snippet of my code to get the page text which is pretty straigt forward: for x := 1 to PDFLibrary.PageCount do begin PDFLibrary.SelectPage(x); Memo1.Text := Memo1.Text + PDFLibrary.GetPageText(7);//passing 7 preserves formatting end; Here is how the text looks on the PDF. You'll have to trust me that it looks like this since I cannot post a screen shot. Tax Labor Tax @ 7.00% $4.20 Parts Tax @ 7.00% $27.75 Tax Total $31.95 When I get the page text I get stuff like this: Tax $4.20 Labor Tax @ 7.00% $27.75 @ Parts Tax 7.00% Tax Total $31.95 I'm guessing there is something wacky about how this PDF is created. Is there anything I can do to get my page text in a format more closely to how it shows on the actual PDF? I need the text to be properly formatted to parse it. Thanks for any advice. Regards, TJ Asher

Ingo Members Profile Find Members Posts Moderator Group Joined: 29 Oct 05 Status: Offline Points: 3530	Post Options Post Reply Quote Ingo Report Post Thanks(0) Quote Reply Posted: 09 Jan 13 at 8:21PM
	Hi TJ! Option 7 is best for your needs. For better parsing you can try the word-by-word-extraction option. Another idea: Do the extraction with the additional data regarding textformatting and positions. Then you can do the layout by your own. Cheers and welcome here, Ingo

tj asher Members Profile Find Members Posts Beginner Joined: 09 Jan 13 Status: Offline Points: 2	Post Options Post Reply Quote tj asher Report Post Thanks(0) Quote Reply Posted: 09 Jan 13 at 10:02PM
	Ingo, I am using option 7 but for some reason the actual text returned from the page is not how it looks on the page. I will consider the option of the text with data positions. Trying to use Acrobat or Foxit PDF Reader and selecting just the text in question is difficult as other areas get selected that don't appear related so I suspect flaws in the orginization of the PDF document itself. Regards, TJ Asher

AndrewC Members Profile Find Members Posts Moderator Group Joined: 08 Dec 10 Location: Geelong, Aust Status: Offline Points: 841	Post Options Post Reply Quote AndrewC Report Post Thanks(0) Quote Reply Posted: 13 Jan 13 at 1:48PM
	We would need to see the actual PDF to explain exactly why the results looks the way they do. I suspect the PDF uses different fonts and sizes for the text. Text extraction is not an exact science and it is a bit like putting together a jigsaw puzzle. Andrew.