pdfbox-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Karl Heinz Kremer <...@khk.net>
Subject Re: Fwd: Trouble reading IEEE pdf
Date Thu, 02 Feb 2017 15:36:15 GMT
Pulpit,

I did not say that in your document the XObjects are images, I said that
they usually are just images. When you analyze 100 random PDF documents,
changes are that that most of them only use the XObject construct for
images and vector graphic, not for elements that contain text. Your
documents are an exception.


Karl Heinz Kremer
PDF Acrobatics Without a Net
PDF Software Development, Training and More...

khk@khk.net
http://www.khkonsulting.com


On Thu, Feb 2, 2017 at 10:33 AM, Pulkit Kapur <pkapur@seas.upenn.edu> wrote:

> Thanks Karl for the reply.
> Thats helpful.
>
> What confuses me is this" very likely because usually such an XObject would
> just be an
> image"
> -> I am able to select the underlying text in the XObject using acrobat and
> copy/paste it.
> Thats why i am confused why pdfbox cannot access the XObject.
>
> Perhaps it is more nuanced than how i am phrasing it.
>
> Thanks,
>
> Pulkit
>
> On Thu, Feb 2, 2017 at 10:27 AM, Karl Heinz Kremer <khk@khk.net> wrote:
>
> > The document does not contain layers (or optional content groups as they
> > are called in PDF), the problem seems to be that the actual text of
> > the document is in an XObject - something that is completely legal in a
> PDF
> > file. I suspect that the text was created in one application, and then a
> > second application was used to create a new page, then placed the header
> on
> > it as "normal" text, and in a second step placed the original content
> into
> > this XObject and then placed it on the page. This is oftentimes what e.g.
> > an imposition application would do. Without having checked in the
> sources,
> > I would assume that when you extract text, PDFBox will just process the
> > Contents structure on the page, but will not recurse into XObjects that
> are
> > encountered - very likely because usually such an XObject would just be
> an
> > image.
> >
> >
> > Karl Heinz Kremer
> > PDF Acrobatics Without a Net
> > PDF Software Development, Training and More...
> >
> > khk@khk.net
> > http://www.khkonsulting.com
> >
> >
> > On Thu, Feb 2, 2017 at 10:10 AM, Pulkit Kapur <pkapur@seas.upenn.edu>
> > wrote:
> >
> > > Hi
> > >
> > > I have uploaded the pdf here:
> > > https://www.scribd.com/document/338221804/0024-iros-2016
> > >
> > > I did some more diagnosis last night and it seems that there are two
> > layers
> > > on the pdf. One which is the content and the other with headers and
> > > footers. Pdf box is only reading the headers and footers.
> > > I suspect this must be common with all conference proceedings.
> > >
> > > Thanks,
> > >
> > > Pulkit
> > >
> > > On Thu, Feb 2, 2017 at 1:21 AM, Tilman Hausherr <THausherr@t-online.de
> >
> > > wrote:
> > >
> > > > Am 02.02.2017 um 05:55 schrieb Pulkit Kapur:
> > > >
> > > >> Hi
> > > >>
> > > >> I am trying to read some past years IEEE conference proceedings i
> > have.
> > > >> I can read the pdf using acrobat and select the text.
> > > >>
> > > >> But when i try to read the text using readText function from the
> > pdfbox
> > > >> library, i only get the headers and footers in the pdf.
> > > >>
> > > >> I did check the document is not encrypted.
> > > >> Also my code works on other pdf documents but all IEEE proceedings
> > that
> > > >> are downloaded form IEEE fail to work.
> > > >>
> > > >> I have attached the pdf document with this message.
> > > >>
> > > >
> > > > Please upload the pdf somewhere, PDF attachments are not allowed
> here.
> > > >
> > > >
> > > >
> > > > Tilman
> > > >
> > >
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message