[John Dalberg]

"Graham Mayor" wrote:
We are still going round in circles with this, but you are missing the
essential point that PDF is a *graphics* format and if the original
document is not available for reference you are using what is essentially
OCR to recreate a document from the PDF- just as you might with a JPG or
TIFF file. OCR software, even at its best, is not capable of recreating
the document (any document) with 100% accuracy. In my opinion Finereader
is the best choice, but even that will not create the style structure of
the original document and you will have a lot of work on your hands to
create an editable document.


It might be in a graphics format but a pdf file should contain enough meta
data to export the file with more intelligent styles. Explain to me how an
editor like Foxit Editor is able to open a pdf file, let's you select a
text, tells you what font was used plus other attributes and let's you edit
the text. If an editor is able to do this why can't a converter dump these
attributes and create Word styles out of them? When I say a Word style, all
I want the style to include is font and size so that I can select all
instances of text that's using the same fonts and size. Surely the
converter should be able to have all similar text be lumped into a single
style.

I don't believe an OCR program will produce a better Word document than a
pdf converter.

I am not looking for extracting the same original structure. All I am
looking for is being able to choose all instances of a certain style. I
don't care if the style is a dummy style, which wasn't in the original
document, created by the converter as long as it defines something like a
font style and size and all text having the same font and size point to
that style.

I am not sure if you understand what my goal is. I don't care if the styles
produced by the converter do not resemble the styles of the original
document. If the orignial used Arial size 10 and the converter produces
font Zulu size 11, it's ok as long I can choose all text of that style. It
will take me a few seconds to choose all these text and modify them back to
use Arial size 10. *BUT* the problem is the converters *do not produce
different styles*. Accurate styles from the original (name..etc) doesn't
matter.


As you apparently didn't create the PDF in the first place, can you
obtain the original document from whoever did - presumably not?

No.



As for Acrobat's own abilities to recreate a PDF, you'll have to take
that up with Adobe. Word is not the issue here.
Had the PDF ben created from a graphical representation of the document
(as some are to make them more difficult to recreate) Acrobat would not
be able to save the PDF as an editable document.

It wasn't created from a graphical representation. It's an eBook and the
author must have been using a word processor.

John Dalberg