This is the source repository for https://homosaurus.org.
Originally based on the Oregon Digital Controlled Vocabulary Manager before being heavily modified.
One needs a setup capable of running Ruby and Rails. One example is to use RVM. One should be able to just accept their GPG keys and install RVM with:
> curl -sSL https://rvm.io/mpapis.asc | gpg --import -
> curl -sSL https://rvm.io/pkuczynski.asc | gpg --import -
> \curl -sSL https://get.rvm.io | bash -s stable
To install Ruby and use it as default, one would do:
> rvm install ruby-2.6.10
> rvm --default use 2.6.10
Note that the last time I tried this, there was an issue with SSL from this thread. To get around that, I actually needed to do the following:
> sudo apt install build-essential
> mkdir ~/Downloads
> cd ~/Downloads
> wget https://www.openssl.org/source/openssl-1.1.1g.tar.gz
> tar zxvf openssl-1.1.1g.tar.gz
> cd openssl-1.1.1g
> ./config --prefix=$HOME/.openssl/openssl-1.1.1g --openssldir=$HOME/.openssl/openssl-1.1.1g
> make
> make test # you might get a few tests failing but just ignore
> make install
> rm -rf ~/.openssl/openssl-1.1.1g/certs
> ln -s /etc/ssl/certs ~/.openssl/openssl-1.1.1g/certs
> cd ~
> rvm install ruby-2.6.10 --with-openssl-dir=$HOME/.openssl/openssl-1.1.1g
> rvm --default use 2.6.10
Install a MySQL compatible server. For Ubuntu, this is:
> sudo apt-get install mysql-server
Also install libraries to be able to connect to that such as the following for Ubuntu:
> sudo apt-get install libmysqlclient-dev
It is recommended to secure that installation and have a homosaurus user with a db of the same name. Sample of commands for that:
> sudo mysql_secure_installation
> mysql -u root -p
>> CREATE USER 'homosaurus'@'%' IDENTIFIED BY 'new_password';
>> CREATE DATABASE homosaurus;
>> GRANT ALL PRIVILEGES ON homosaurus.* TO 'homosaurus'@'%';
>> ALTER DATABASE homosaurus CHARACTER SET utf8 COLLATE utf8_unicode_ci;
>> FLUSH PRIVILEGES;
This is used to cache the Library of Congress Linked Data and could be used in the future to cache other things. See the quickstart guide from https://github.com/blazegraph/database/wiki/Main_Page to run this locally. Once it is working, one can load a Library of Congress dataset using:
> wget https://lds-downloads.s3.amazonaws.com/lcsh.skos.nt.zip
> unzip lcsh.skos.nt.zip
> curl -H 'Content-Type: text/turtle' --upload-file /data/tools/blazegraph/lcsh.skos.nt -X POST "http://localhost:9999/blazegraph/sparql?context-uri=https://id.loc.gov/authorities/subjects"
Modify the various "sample" files in the config directory (removing the sample). These should be:
Copy this to database.yml and edit it with your MySQL configuration.
Copy to homosaurus.yml. This shouldn't require modification unless one wants to test the captcha stuff.
Copy to contact_emails.yml.
Copy to solr.yml.
Copy to blazegraph.yml.
One needs a Javascript runtime, The easiest is nodejs that can be installed on Ubuntu using:
> sudo apt-get install nodejs
Next one needs to install all of the Ruby gems using: bundle install from the root directory. Once that has fully
succeeded, one can optionally export a copy of the database from the mysql on the server and bring that locally.
The export and import commands are essentially:
Export:
mysqldump --lock-tables=false --set-gtid-purged=OFF --column-statistics=0 --no-tablespaces -P 3306 -h localhost -u homosaurus -p homosaurus > <dump_file>
Import:
mysql -u homosaurus -p homosaurus < <dump_file>
Once done, run rake db:migrate for rails to sync to the db state.
Solr is the index used for the search functionality for the repository. For local development, once get copy the needed instance files to the <project_root>/solr/binary folder with something like the following:
wget https://www.apache.org/dyn/closer.lua/lucene/solr/8.11.2/solr-8.11.2.tgz?action=download -P /<project_root>/solr/binary/solr-8.11.2.tgz
wget https://downloads.apache.org/lucene/solr/8.11.2/solr-8.11.2-src.tgz.sha512 -P /<project_root>/solr/binary/solr-8.11.2.tgz.sha512
The from the project root, once can just use the command solr_wrapper to start a solr instance from the command prompt.
If the index becomes corrupt (ie. from a restart where Solr wasn't shut down), then one can clear the Solr instance by
deleting everything contained in the <project_root>/solr/instance/ directory.
The rails c command will bring up the console. One will need to run the following to populate the Solr index if using
the existing DB copy:
DSolr.reindex_all
The rails s command will start the application one should be able to reach at http://localhost:3000.