# make sure we can ssh in!
sudo apt-get install --assume-yes openssh-server

# create the arxivwiki user
sudo useradd -p "\$1\$tiAZODpE\$zpBSj7s87VaILoJsL/w.M/" --shell /bin/bash --groups admin,adm arxivwiki

# now logout, and log back in as arxivwiki

sudo deluser ubuntu

# Ensure apt-get isn't trying to read from a cdrom:
sudo sed -i '/deb cdrom:/d' /etc/apt/sources.list

# Start installing, via apt-get
## First update, in case we're on an old virtual machine
sudo apt-get update --assume-yes 
sudo apt-get dist-upgrade --assume-yes
## Apache
sudo apt-get install --assume-yes apache2 libapache2-svn
## subversion
sudo apt-get install --assume-yes subversion subversion-tools
## bittorrent
sudo apt-get install --assume-yes bittorrent
## zip
sudo apt-get install --assume-yes zip unzip
## java (prompts for license agreement!)
sudo apt-get install --assume-yes sun-java6-jdk sun-java6-source
## Samba drivers
sudo apt-get install --assume-yes smbfs
## lynx and curl
sudo apt-get install --assume-yes lynx curl
## patch, dos2unix
sudo apt-get install --assume-yes patch tofrodos
## some things we need to compile Virtuoso
sudo apt-get install --assume-yes autoconf automake libtool flex bison gperf gawk m4 make openssl libssl-dev
## PHP (needed for Danny Ayers sparql-demo
sudo apt-get install --assume-yes php5 php5-cli php-xsl php-xml*


# Install some more things, which we perhaps don't really need?
## MySQL (TODO: this requires user interaction, when configuring packages!)
# sudo apt-get install --assume-yes mysql-client-5.0 mysql-server-5.0 phpmyadmin
## latex
# sudo apt-get install --assume-yes texlive-latex-base imagemagick
## sendmail
# sudo apt-get install --assume-yes sendmail sasl2-bin
## DAV support for apache (not available in 7.10)
# sudo apt-get install --assume-yes libapache-mod-dav
##  something needed for jungledisk? (not available in 7.10)
# sudo apt-get install --assume-yes ia32-libs ia32-libs-gtk

# make mod_rewrite and mod_proxy available
sudo ln -s /etc/apache2/mods-available/rewrite.load /etc/apache2/mods-enabled/rewrite.load
sudo ln -s /etc/apache2/mods-available/proxy.load /etc/apache2/mods-enabled/proxy.load
sudo ln -s /etc/apache2/mods-available/proxy_http.load /etc/apache2/mods-enabled/proxy_http.load

# setup the sites
cd /etc/apache2/sites-available/
sudo rm arxivwiki.org.conf
sudo wget http://katlas.org/svn/arxivwiki/trunk/server-build/arxivwiki.org.conf
sudo ln -s /etc/apache2/sites-available/arxivwiki.org.conf /etc/apache2/sites-enabled/arxivwiki.org.conf
sudo rm proxy.conf
sudo wget http://katlas.org/svn/arxivwiki/trunk/server-build/proxy.conf
sudo ln -s /etc/apache2/sites-available/proxy.conf /etc/apache2/sites-enabled/proxy.conf
sudo apache2ctl restart

sudo mkdir -p /www/arxivwiki.org
sudo chown arxivwiki /www/arxivwiki.org/
sudo chgrp arxivwiki /www/arxivwiki.org/
cd /www/arxivwiki.org/
svn checkout http://katlas.org/svn/arxivwiki/trunk/snorql/
svn checkout http://katlas.org/svn/arxivwiki/trunk/sparql-form/

sudo mkdir -p /home/arxivwiki/samba/
sudo chown arxivwiki /home/arxivwiki/
sudo chgrp arxivwiki /home/arxivwiki/
sudo chown arxivwiki /home/arxivwiki/samba/
sudo chgrp arxivwiki /home/arxivwiki/samba/

# @deprecated, but a useful wget example!
# cd /www/arxivwiki.org/
# wget -nH http://katlas.org/svn/arxivwiki/trunk/server-build/html/ -r --cut-dirs=5 -np

# we should download the torrent at this point! at present it doesn't exist, so that's a bit ambitious.
# now unpack the tar files from the torrent
mkdir -p ~/samba/mirror/
cd ~/samba/mirror/
for file in ../torrent/*.tar; do tar xvfk $file; done

# download sesame2
cd ~
wget http://internap.dl.sourceforge.net/sourceforge/sesame/openrdf-sesame-2.0-sdk.zip
unzip openrdf-sesame-2.0-sdk.zip
rm openrdf-sesame-2.0-sdk.zip
chmod u+x ~/openrdf-sesame-2.0/bin/start-console.sh

cd ~
wget http://www.ibiblio.org/pub/mirrors/apache/tomcat/tomcat-5/v5.5.26/bin/apache-tomcat-5.5.26.zip
unzip apache-tomcat-5.5.26.zip
rm apache-tomcat-5.5.26.zip
# now patch server.xml, to enable to proxy connector on port 8082
# cd ~/apache-tomcat-5.5.26/conf
# wget http://katlas.org/svn/arxivwiki/trunk/server-build/patches/patch.server.xml
# patch server.xml patch.server.xml

# install the webapps in tomcat
sudo cp /home/arxivwiki/openrdf-sesame-2.0/war/*.war ~/apache-tomcat-5.5.26/webapps/

## .bashrc doesn't work the way I expect, lame.
echo ~/.bashrc > ~/.bash_profile
echo "export PATH=~/apache-maven-2.0.8/bin:\$PATH" >> ~/.bashrc
echo "export JAVA_HOME=/usr/lib/jvm/java-6-sun/jre/" >> ~/.bashrc
chmod a+x ~/apache-tomcat-5.5.26/bin/*.sh

# download and install rc.local, and run it right now!
cd /etc/
sudo rm /etc/rc.local
sudo wget http://katlas.org/svn/arxivwiki/trunk/server-build/rc.local
sudo dos2unix rc.local
sudo chmod u+x rc.local
sudo ./rc.local

# Download and build virtuoso. This takes hours. :-(
cd ~
wget http://superb-east.dl.sourceforge.net/sourceforge/virtuoso/virtuoso-opensource-5.0.5.tar.gz
tar xzvf virtuoso-opensource-5.0.5.tar.gz
rm virtuoso-opensource-5.0.5.tar.gz
cd virtuoso-opensource-5.0.5
./configure
make
sudo make install

# start virtuoso
cd /usr/local/virtuoso-opensource/var/lib/virtuoso/db
sudo rm virtuoso.lck
sudo ../../../../bin/virtuoso-t -f &

## You'll have to go do some virtuoso configuring at this point, which it seems we can't do from the commandline. Sorry, lame.
## Login, dba, dba.
## Turn on the SPARQL (and iSPARQL) packages.
## Set up rdf_sinks to put things in the right contexts
### create DAV directories with properties virt:rdf_graph='<context>' and virt:rdf_sponger='on'
## Load RDF data, via curl -- this can be automated below.

# download and unpackage maven
cd ~
wget http://tqft.net/files/apache-maven-2.0.8-bin.zip
unzip apache-maven-2.0.8-bin.zip
rm apache-maven-2.0.8-bin.zip

# checkout the svn repository, and build:
cd ~
svn checkout http://katlas.org/svn/arxivwiki/
cd ~/arxivwiki/trunk/metadata
mvn install
cd ~/arxivwiki/trunk/arxiv-oai2rdf
mvn install
cd ~/arxivwiki/trunk/oai2rdf
cd src/main/java/edu/mit/simile/rdfizer/oai
rm Main.java
wget http://katlas.org/svn/arxivwiki/trunk/server-build/patches/Main.java
cd ~/arxivwiki/trunk/oai2rdf
mvn install

# download everything from the arxiv OAI interface, and the citebase OAI interface
cd ~/arxivwiki/trunk/metadata
./scrape-oai.sh /www/arxivwiki.org/

# scrape the arxiv!
cd ~/arxivwiki/trunk/metadata
rm /www/arxivwiki.org/mirror/lists/identifiers.list
./build-mirror.sh /www/arxivwiki.org/

mkdir -p /www/arxivwiki.org/rdf/arxiv/info/
mkdir -p /www/arxivwiki.org/rdf/versions/info/
mkdir -p /www/arxivwiki.org/rdf/citations/info/
mkdir -p /www/arxivwiki.org/rdf/crawler-anomalies/info/
mkdir -p /www/arxivwiki.org/rdf/citebase/info/

# prepare all the rdf!
./prepare-rdf.sh /www/arxivwiki.org/

# put all the RDF somewhere useful :-)
#./create-arxivwiki-repository.sh
#./load-all-rdf.sh http://localhost:8080/openrdf-sesame ~/samba/

### errr
grep "index\.html" < all-files.txt | sed "s/^\.\/[0-9]*\/[0-9]*\/\(.*\)\/index\.html/http:\/\/arxiv.org\/abs\/\1/g" | sort > index-files.txt