PDF to Html Converter

DoctorWkt · #1 (**permalink**) 29-09-2008, 01:31 AM

I've just spent a few days modifying the pdftohtml utility in poppler to produce Html output more suitable for uBook. The tool also now extracts all the images from a PDF document and hyperlinks them into the Html. You can find my version of the program here.

It's a C program and uses the poppler library, so it will probably only run on Linux. You should use the -noframes option when converting PDF files. And, like the original program, it doesn't cope well with multi-column PDF files.

µBook · #2 (**permalink**) 06-10-2008, 06:28 PM

Awesome! I've been itching to delete the pseudo-PDF support from inside uBook

Can I add a link to your page on the uBook web site?
Thanks,

DoctorWkt · #3 (**permalink**) 08-10-2008, 06:33 AM

Sure thing. I've provided the Poppler maintainers with a set of patches against the current branch, and I hope it gets integrated soon. Until then, point people at my pdftohtml website.

faenad · #4 (**permalink**) 04-12-2008, 07:05 PM

From the sample conversion on your website, this program look like to be the best converter for generating ebook friendly files.

But how to install it and use it?

The Tarball don't include a readme and the manual page just explain the command line parameters.

can you explain to us how to install and run your utility?

What I've done so far :
With Ubuntu 8.04 32bit
I extracted the tarball in a folder, and tried the followings commands with no success :
./configure
make
sudo make install

I also tried to run pdftozip and to run it in terminal but nothing happen.

DoctorWkt · #5 (**permalink**) 08-12-2008, 11:01 PM

Quote:

Originally Posted by faenad

Can you explain to us how to install and run your utility?

Yes, I was hoping that the patches would be picked up and put into the main Poppler tree, but no such luck yet. There is no ./configure, so you can simply edit the Makefile to find the location of the header and library files:

Code:

# Compile flags for FreeBSD 6
#CXXFLAGS= -O2 -Wall -I. -I/usr/local/include/poppler -I/usr/local/include
#CFLAGS= $(CXXFLAGS)
#LIBDIRS= -L/usr/local/lib

# Compile flags for Ubuntu Linux
CXXFLAGS= -O2 -Wall -I. -I/usr/include/poppler
CFLAGS= $(CXXFLAGS)
LIBDIRS=

Comment out the existing lines (as shown above) and add the ones for Linux. Then you should be able to run the "make" command to build the executable, and then move it by hand into a directory like /usr/bin.

JMonkeYJ · #6 (**permalink**) 06-03-2009, 03:20 AM

OK, I'm having some problems here. I think I managed to get it installed because I have the executable files (pdftohtml pdftozip). How do I check to make sure these are the correct files? When I run, eg

pdftohtml -v

I get the exact same output from your custom version as I do from the standard version. I have made sure that I am executing the one I just compiled rather than the standard one. Should yours output different version information?

It's a little difficult to figure out what is going wrong, but it looks like lines are still being treated as separate paragraphs. Sometimes lines are joined, but it looks like there is a tab at the join point. Also, there is no gap from one paragraph to the next, so it's basically unreadable. The command I'm running is:

pdftohtml -raw -reflow -noframes -enc Latin1

I think I've tried every combination of those options without success. Thanks for any help, this program seems very promising if I could just get it to work.

khubab1 · #7 (**permalink**) 28-08-2009, 08:40 PM

Hi, I have read an article about pdf to html conversion at www.nobleatom.com. I hope it will be helpful for you. Take a look. Thanks.