my take at a PDF text extraction utility
Java TeX
Pull request Compare This branch is even with oyvindberg:master.
Latest commit d58f241 Aug 14, 2011 yvind Raddum Berg added instructions for how to build

README

#######################################
# How to build PDFExtract from source
#######################################

# 1. create a folder for the projects and cd to it


#
# 2. install TEI P5 model
#
git clone http://github.com/elacin/TEI-P5-Java-model.git
cd TEI-P5-Java-model/
#this chooses version 0.3, which is currently used by PDFExtract
git checkout 29d668e
mvn install
cd ..

#
# 3. install patched PDFBox 
#
svn checkout http://svn.apache.org/repos/asf/pdfbox/trunk/ pdfbox
#apply patch (tested against pdfbox svn r1157684)
cd pdfbox
patch -p0 < ../PDFExtract/parent/patch/pdfbox_poms.patch
patch -p0 < ../PDFExtract/parent/patch/pdfbox-font-bounding-boxes.patch 
patch -p0 < ../PDFExtract/parent/patch/pdfbox-drawer-visibility.patch
mvn install
cd ..


#
# 4. install PDFExtract
#
git clone http://github.com/elacin/PDFExtract.git
cd PDFExtract/parent
mvn -DskipTests=true assembly:assembly #yes, some cleanup of tests is in order


# the binary distribution will end up as PDFExtract/pdfextract-cli/target/pdfextract-cli-${VERSION}-bin.tar.bz2