Installation instructions

Our development style

We have relatively small milestones, due approximately every two months.

We're happy to mentor newcomers - pick a minor issue and we'll walk you through the steps to solve it.

Questions or ideas? Get in touch via the Google group.

Summary/TOC

Basic

Extras

Brief installation instructions

Full instructions are here.
To get a minimal working "Math enhanced Drupal", you only need to do the first few steps below.

GRAB THE CODE FROM THIS REPOSITORY

It's Drupal with our custom extensions.

git clone git://github.com/KWARC/planetary.git
ln -s /home/planetary/planetary /var/www/drupal
cd planetary
chmod a+w sites/default/files
cp sites/default/default.settings.php sites/default/settings.php
chmod a+w sites/default/settings.php

INSTALL DRUSH

git clone --branch 7.x-5.x http://git.drupal.org/project/drush.git
ln -s /home/planetary/drush/drush /usr/local/bin/drush

CREATE DATABASE

mysql -u root -p

CREATE DATABASE planetary;
GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, INDEX, ALTER ON planetary.*
TO 'planetary'@'localhost' IDENTIFIED BY 'make-something-up';

CONFIGURE APACHE

<VirtualHost *:80>
  ServerAdmin [email protected]
  ServerName xxx.yyy.zzz
  DocumentRoot /var/www/drupal/
  <Directory /var/www/drupal/>
      AllowOverride All
      Order allow,deny
      allow from all
  </Directory>
  ErrorLog /var/log/apache2/a.log
  CustomLog /var/log/apache2/a.log combined
</VirtualHost>

INSTALL REQUIRED MODULES

(By the way, you'll probably want to look through the extended installation instructions to make sure you have all the necessary configuration steps sorted out.)

For PlanetMath profile

drush -y dl migrate views features reroute_email references \
 pathauto subform token relation relation_select \
  legal userpoints userpoints_nc privatemsg \
  ctools delete_all devel backup_migrate commentrss \
 nodeaccess wysiwyg views_php sparql filefield_paths date  \
 dhtml_menu riddler apachesolr apachesolr_views

drush -y dl og-7.x-2.x-dev entityreference-7.x-1.x-dev entity-7.x-1.x-dev \
 views_bulk_operations-7.x-3.x-dev captcha-7.x-1.x-dev efq_views-7.x-1.x-dev \
 watcher-7.x-1.x-dev rdfx-7.x-2.x-dev

For Panta Rhei profile

drush -y dl views features reroute_email references \
 pathauto subform token relation relation_select \
  ctools delete_all devel backup_migrate commentrss \
 nodeaccess wysiwyg views_php filefield_paths date  \
  riddler apachesolr apachesolr_views

drush -y dl og-7.x-2.x-dev entityreference-7.x-1.x-dev entity-7.x-1.x-dev \
 views_bulk_operations-7.x-3.x-dev captcha-7.x-1.x-dev efq_views-7.x-1.x-dev \
 watcher-7.x-1.x-dev rdfx-7.x-2.x-dev

RUN PROFILE

Supported profiles are planetmath and panta:

drush site-install PROFILENAME --db-url=mysql://DBUSER:DBPASS@localhost/DBNAME \
  --site-name=SITENAME  --account-name=ADMINNAME --account-pass=ADMINPASS

GET THE ACE EDITOR, AND PUT IT IN YOUR LIBRARIES DIRECTORY

Note that latest versions of this require a library that comes with GCC 4.7, see these notes on Ubuntu 12.04, including this.

git clone git://github.com/ajaxorg/ace.git
cd ace
npm install
node Makefile.dryice.js full

GET THE SHAREJS REPOSITORY

git clone git://github.com/jucovschi/ShareJS.git
cd ShareJS
git checkout ace_services

sudo apt-get install redis-server coffeescript

Or grab the latest coffeescript if you're on an older system: http://coffeescript.org/

npm install redis
sudo npm link
cake.coffeescript build
cake.coffeescript webclient

(Alternatively, just cake depending on how you installed coffeescript.)

RUN THE SHAREJS SERVER (E.G. WITHIN GNU SCREEN) VIA

bin/exampleserver

INSTALL THE SHAREJSSERVICES MODULE

This module ships with Planetary. If you haven't already turned it on, do so now:

drush -y en sharejsservices

And enable the ACE 2.0 editor at admin/config/content/wysiwyg.

INSTALL APACHE SOLR

Pick a mirror from Apache and download Solr with a command similar to this one:

wget http://mirror.ox.ac.uk/sites/rsync.apache.org/lucene/solr/3.6.1/apache-solr-3.6.1-src.tgz

tar -zxvf apache-solr-3.6.1-src.tgz
cd apache-solr-3.6.1
sudo apt-get install ant

ant ivy-bootstrap

If you don't have the JDK installed you can do something like this (or just sudo apt-get install openjdk-6-jdk):

sudo add-apt-repository ppa:ferramroberto/java
sudo apt-get update
sudo apt-get install sun-java6-jdk

ant compile

mv solr/example/solr/conf/schema.xml solr/example/solr/conf/schema.bak
mv solr/example/solr/conf/solrconfig.xml solr/example/solr/conf/solrconfig.bak
mv solr/example/solr/conf/protwords.txt solr/example/solr/conf/protwords.bak

cd ../planetary
drush dl apachesolr

cp sites/all/modules/apachesolr/solr-conf/solr-3.x/schema.xml \
 ../apache-solr-3.6.1/solr/example/solr/conf/schema.xml
cp sites/all/modules/apachesolr/solr-conf/solr-3.x/solrconfig.xml \
 ../apache-solr-3.6.1/solr/example/solr/conf/solrconfig.xml
cp sites/all/modules/apachesolr/solr-conf/solr-3.x/protwords.txt \
 ../apache-solr-3.6.1/solr/example/solr/conf/protwords.txt

The following step is important to get the relevant webapp to build.

cd ../apache-solr-3.6.1/solr/
ant example

Then, you can run the webapp:

cd example
java -jar start.jar

Check it out via lynx http://localhost:8983/solr/admin/ or similar.

Finally, install the Drupal modules so you can use this:

drush -y en apachesolr apachesolr_search

Now you can visit admin/config/search/apachesolr to check out the configuration and control re-indexing. You can check out the details of your index at admin/reports/apachesolr/solr. In order to set Apache Solr as the default search engine (resp. hide the other search engines) you need to change the relevant radio button (resp. tick box) at admin/config/search/settings.

To set up faceted search, install facetapi:

drush -y dl facetapi-7.x-1.x-dev
drush -y en facetapi

and configure at admin/config/search/apachesolr/settings/solr/facets.

Some further information (YMMV) at http://drupalcode.org/project/apachesolr.git/blob_plain/refs/heads/7.x-1.x:/README.txt

If you want to search user pages, you should install

drush -y dl apachesolr_user
drush -y en apachesolr_user

and visit admin/config/search/apachesolr to tick the box selecting User as an indexable type.

Finally, you definitely want to do

drush -y dl apachesolr_views
drush -y en apachesolr_views

(PlanetMath will depend on this stuff, so I'll get as much of it as possible into the profile...)

There are some additional plugins but details on those will follow later. See this ticket #141 for some further notes.

INSTALL LATEXML

svn co https://svn.mathweb.org/repos/LaTeXML/branches/arXMLiv

apt-get install perlmagick libxml2 libxml2-dev libxslt1.1 libxslt1-dev \
libxml-libxml-perl libclone-perl libdata-compare-perl libio-prompt-perl \
libparse-recdescent-perl libxml-libxslt-perl libdb5.1 libdb5.1-dev \
libgdbm-dev libarchive-zip-perl unzip

sudo perl -MCPAN -e shell
install Parse::RecDescent XML::LibXSLT DB_File Data::Compare File::Which \
 Marpa::R2 Test::LeakTrace
quit

You should grab and install a current version of Mojolicious:

wget -O mojo.tar.gz https://github.com/kraih/mojo/tarball/master
tar -zvxf mojo.tar.gz

Change to the directory that was unpacked and do:

perl Makefile.PL
make
sudo make install

Once you've gotten all that sorted out, you can go to the arXMLiv directory and, again,

perl Makefile.PL
make
sudo make install

CONFIGURE LATEXML TO RUN UNDER APACHE

sudo apt-get install libapache2-mod-perl2 libplack-perl

sudo chgrp -R www-data /path/to/LaTeXML/webapp
sudo chmod -R g+w /path/to/LaTeXML/webapp

Create a "latexml" file in /etc/apache2/sites-available like this:

<VirtualHost *:80>
    ServerName latexml.example.com 
    DocumentRoot /path/to/LaTeXML/webapp
    Header set Access-Control-Allow-Origin * 
    <Perl>
      $ENV{PLACK_ENV} = 'production';
      $ENV{MOJO_HOME} = '/path/to/LaTeXML/webapp';
    </Perl>

    <Location />
      SetHandler perl-script
      PerlHandler Plack::Handler::Apache2
      PerlSetVar psgi_app /path/to/LaTeXML/webapp/ltxmojo
    </Location>

    ErrorLog /var/log/apache2/latexml.error.log
    LogLevel warn
    CustomLog /var/log/apache2/latexml.access.log combined
</VirtualHost>

and turn it on.

SET UP VIRTUOSO

drush dl libraries rdfx sparql_views
drush -y en libraries
drush -y en rdfx sparql_views views_ui rdfui

sudo aptitude install dpkg-dev build-essential autoconf automake \
 libtool flex bison gperf gawk m4 make odbcinst libxml2-dev libssl-dev \
 libreadline-dev

wget http://downloads.sourceforge.net/project/virtuoso/virtuoso/6.1.5/virtuoso-opensource-6.1.5.tar.gz
tar -zxvf virtuoso-opensource-6.1.5.tar.gz
cd virtuoso-opensource-6.1.5

./configure --prefix=/usr/local/ --with-readline --program-transform-name="s/isql/isql-v/"
nice make
sudo make install

Invoke via screen

/usr/local/bin/virtuoso-t +configfile /usr/local/var/lib/virtuoso/db/virtuoso.ini -fd

Seed your triple store with the Math Subject Classification taxonomy

wget http://msc2010.org/mscwork/msc2010.skos
curl -T msc2010.skos http://planetmath.org:8890/DAV/home/pm/rdf_sink/xml.rdf -H "Content-Type: application/rdf+xml" -u dav:PASSWORD

SET UP PYRDFA

First of all, if you're going to use our pyrdfa module, don't forget that you'll have to patch the Drupal core (see full instructions for details)!

git clone git://github.com/RDFLib/PyRDFa.git
cd PyRDFa
sudo python setup.py install

The relevant executable is in ./scripts/localRDFa.py. You can test it by pulling down some RDFa enhanced webpage and running

python /path/to/PyRDFa/scripts/localRDFa.py -x in.html > out.xml

If you run into trouble it may be because you don't have the right versions of required libraries installed. I solved these problems with the following commands (obtaining rdflib version 3.4.0-dev and html5lib version 0.95) and then rebuilt/reinstalled PyRDFa as above.

sudo pip install markdown
git clone git://github.com/RDFLib/rdflib.git
cd rdflib && sudo python setup.py install

(Further note: ideally this would be set up to run as a web service, similar to LaTeXML and so on, but that will take a moment to set up.)

LATEX INTEGRATION TO GENERATE PDFs

This is handled by the planetmath_view_pdf module: make sure the PDFLaTeX command is correct for your system (e.g. /usr/bin/pdflatex perhaps? Or /usr/local/texlive/2011/bin/x86_64-linux/pdflatex?) and create the output directory at sites/default/files/texpdf/.

INSTALL NNEXUS

git clone [email protected]:dginev/nnexus.git
wget http://li101-104.members.linode.com/nnexus.sql.gz
gunzip nnexus.sql.gz

mysql> grant all on nnexusdb.* to 'nnexususer'@localhost identified by 'PASSWORD';
mysql> create database nnexusdb;

mysql -u nnexususer --password=PASSWORD nnexusdb < nnexus.sql

sudo perl -MCPAN -e shell
cpan> install XML::Simple Data::Dumper LWP::Simple XML::Writer XML::SAX Unicode::String Graph
cpan> exit

Now edit baseconf.xml so that dbname, dbuser, and dbpass match the values you specified above.

You can then run the NNexus server (e.g. from within GNU Screen) via

./nnexusserver.pl

Alternatively, you can run it using Mojolicious and several other standard perl modules which can be installed via apt-get or cpan (cf. LaTeXML section above):

apt-get install libmojolicious-perl libxml-simple-perl \
  libunicode-string-perl libgraph-perl libjson-perl

Then, in order to run the server:

morbo --listen=http://*:3001 nnexusmojo.pl

CONFIGURE NNEXUS TO RUN UNDER APACHE

Create a "nnexus" file in /etc/apache2/sites-available like this:

<VirtualHost *:80>
    ServerName nnexus.example.com 
    DocumentRoot /path/to/nnexus
    Header set Access-Control-Allow-Origin * 
    <Perl>
      $ENV{PLACK_ENV} = 'production';
      $ENV{MOJO_HOME} = '/path/to/nnexus';
    </Perl>

    <Location />
      SetHandler perl-script
      PerlHandler Plack::Handler::Apache2
      PerlSetVar psgi_app /path/to/nnexus/nnexusmojo.pl
    </Location>

    ErrorLog /var/log/apache2/nnexus.error.log
    LogLevel warn
    CustomLog /var/log/apache2/nnexus.access.log combined
</VirtualHost>