You are viewing a plain text version of this content. The canonical link for it is here.

Posted to users@pdfbox.apache.org by "Hesham G." <he...@gmail.com> on 2011/05/26 18:41:26 UTC

PDFBox extracts fi char wrong

Hello ,

I am using PDFBox version 1.4 to extract text from a PDF, but all the words having "fi" inside them are extracted wrong. You can test the following 1 page PDF sample: http://www.4shared.com/document/GAMnpE9A/the_fi_char.html

I am aware of the post: https://issues.apache.org/jira/browse/PDFBOX-860 which mentions that this is now fixed if I use the ICU4J jar, which already exists inside the PDFBox 1.4 jar as I can see, but still such words are parsed wrong. Am I missing something here ?


Best regards ,
Hesham

Re: PDFBox extracts fi char wrong

Posted by "Hesham G." <he...@gmail.com>.

Thomas ,

Thank you a lot for the information.
I have just tested it now and it extracts those characters fine.

I will upgrade to this version now.


Best regards ,
Hesham

---------------------------------------------
Included message :


> Hi Hesham,
>
>> I am using PDFBox version 1.4 to extract text from a PDF, but all the 
>> words having "fi" inside them are extracted wrong. You can test the 
>> following 1 page PDF sample: 
>> http://www.4shared.com/document/GAMnpE9A/the_fi_char.html
>>
>> I am aware of the post: https://issues.apache.org/jira/browse/PDFBOX-860 
>> which mentions that this is now fixed if I use the ICU4J jar, which 
>> already exists inside the PDFBox 1.4 jar as I can see, but still such 
>> words are parsed wrong. Am I missing something here ?
>
> I checked your file and get the same results with PDFBox 1.4. But from 
> version 1.5 on everything seems to be OK (apart from my own ligature 
> problems…).
> So I suggest to update your PDFBox version. Note that version 1.6 
> definitely needs the additional ICU jar (see 
> https://issues.apache.org/jira/browse/PDFBOX-970).
>
> All the best
> Thomas

Re: PDFBox extracts fi char wrong

Posted by Thomas Fischer <fi...@aon.at>.

Hi Hesham,

> I am using PDFBox version 1.4 to extract text from a PDF, but all the words having "fi" inside them are extracted wrong. You can test the following 1 page PDF sample: http://www.4shared.com/document/GAMnpE9A/the_fi_char.html
> 
> I am aware of the post: https://issues.apache.org/jira/browse/PDFBOX-860 which mentions that this is now fixed if I use the ICU4J jar, which already exists inside the PDFBox 1.4 jar as I can see, but still such words are parsed wrong. Am I missing something here ?

I checked your file and get the same results with PDFBox 1.4. But from version 1.5 on everything seems to be OK (apart from my own ligature problems…). 
So I suggest to update your PDFBox version. Note that version 1.6 definitely needs the additional ICU jar (see https://issues.apache.org/jira/browse/PDFBOX-970).

All the best
Thomas